add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
package webmail
import (
"context"
replace http basic auth for web interfaces with session cookie & csrf-based auth
the http basic auth we had was very simple to reason about, and to implement.
but it has a major downside:
there is no way to logout, browsers keep sending credentials. ideally, browsers
themselves would show a button to stop sending credentials.
a related downside: the http auth mechanism doesn't indicate for which server
paths the credentials are.
another downside: the original password is sent to the server with each
request. though sending original passwords to web servers seems to be
considered normal.
our new approach uses session cookies, along with csrf values when we can. the
sessions are server-side managed, automatically extended on each use. this
makes it easy to invalidate sessions and keeps the frontend simpler (than with
long- vs short-term sessions and refreshing). the cookies are httponly,
samesite=strict, scoped to the path of the web interface. cookies are set
"secure" when set over https. the cookie is set by a successful call to Login.
a call to Logout invalidates a session. changing a password invalidates all
sessions for a user, but keeps the session with which the password was changed
alive. the csrf value is also random, and associated with the session cookie.
the csrf must be sent as header for api calls, or as parameter for direct form
posts (where we cannot set a custom header). rest-like calls made directly by
the browser, e.g. for images, don't have a csrf protection. the csrf value is
returned by the Login api call and stored in localstorage.
api calls without credentials return code "user:noAuth", and with bad
credentials return "user:badAuth". the api client recognizes this and triggers
a login. after a login, all auth-failed api calls are automatically retried.
only for "user:badAuth" is an error message displayed in the login form (e.g.
session expired).
in an ideal world, browsers would take care of most session management. a
server would indicate authentication is needed (like http basic auth), and the
browsers uses trusted ui to request credentials for the server & path. the
browser could use safer mechanism than sending original passwords to the
server, such as scram, along with a standard way to create sessions. for now,
web developers have to do authentication themselves: from showing the login
prompt, ensuring the right session/csrf cookies/localstorage/headers/etc are
sent with each request.
webauthn is a newer way to do authentication, perhaps we'll implement it in the
future. though hardware tokens aren't an attractive option for many users, and
it may be overkill as long as we still do old-fashioned authentication in smtp
& imap where passwords can be sent to the server.
for issue #58
2024-01-04 15:10:48 +03:00
cryptorand "crypto/rand"
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
"encoding/base64"
"encoding/json"
"errors"
"fmt"
"io"
2024-02-08 16:49:01 +03:00
"log/slog"
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
"mime"
"mime/multipart"
2023-10-15 16:05:20 +03:00
"net"
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
"net/http"
"net/mail"
"net/textproto"
"os"
implement "requiretls", rfc 8689
with requiretls, the tls verification mode/rules for email deliveries can be
changed by the sender/submitter. in two ways:
1. "requiretls" smtp extension to always enforce verified tls (with mta-sts or
dnssec+dane), along the entire delivery path until delivery into the final
destination mailbox (so entire transport is verified-tls-protected).
2. "tls-required: no" message header, to ignore any tls and tls verification
errors even if the recipient domain has a policy that requires tls verification
(mta-sts and/or dnssec+dane), allowing delivery of non-sensitive messages in
case of misconfiguration/interoperability issues (at least useful for sending
tls reports).
we enable requiretls by default (only when tls is active), for smtp and
submission. it can be disabled through the config.
for each delivery attempt, we now store (per recipient domain, in the account
of the sender) whether the smtp server supports starttls and requiretls. this
support is shown (after having sent a first message) in the webmail when
sending a message (the previous 3 bars under the address input field are now 5
bars, the first for starttls support, the last for requiretls support). when
all recipient domains for a message are known to implement requiretls,
requiretls is automatically selected for sending (instead of "default" tls
behaviour). users can also select the "fallback to insecure" to add the
"tls-required: no" header.
new metrics are added for insight into requiretls errors and (some, not yet
all) cases where tls-required-no ignored a tls/verification error.
the admin can change the requiretls status for messages in the queue. so with
default delivery attempts, when verified tls is required by failing, an admin
could potentially change the field to "tls-required: no"-behaviour.
messages received (over smtp) with the requiretls option, get a comment added
to their Received header line, just before "id", after "with".
2023-10-24 11:06:16 +03:00
"runtime/debug"
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
"sort"
"strings"
2023-10-15 16:05:20 +03:00
"sync"
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
"time"
_ "embed"
"golang.org/x/exp/maps"
"github.com/mjl-/bstore"
"github.com/mjl-/sherpa"
"github.com/mjl-/sherpadoc"
"github.com/mjl-/sherpaprom"
"github.com/mjl-/mox/dkim"
"github.com/mjl-/mox/dns"
"github.com/mjl-/mox/message"
implement "requiretls", rfc 8689
with requiretls, the tls verification mode/rules for email deliveries can be
changed by the sender/submitter. in two ways:
1. "requiretls" smtp extension to always enforce verified tls (with mta-sts or
dnssec+dane), along the entire delivery path until delivery into the final
destination mailbox (so entire transport is verified-tls-protected).
2. "tls-required: no" message header, to ignore any tls and tls verification
errors even if the recipient domain has a policy that requires tls verification
(mta-sts and/or dnssec+dane), allowing delivery of non-sensitive messages in
case of misconfiguration/interoperability issues (at least useful for sending
tls reports).
we enable requiretls by default (only when tls is active), for smtp and
submission. it can be disabled through the config.
for each delivery attempt, we now store (per recipient domain, in the account
of the sender) whether the smtp server supports starttls and requiretls. this
support is shown (after having sent a first message) in the webmail when
sending a message (the previous 3 bars under the address input field are now 5
bars, the first for starttls support, the last for requiretls support). when
all recipient domains for a message are known to implement requiretls,
requiretls is automatically selected for sending (instead of "default" tls
behaviour). users can also select the "fallback to insecure" to add the
"tls-required: no" header.
new metrics are added for insight into requiretls errors and (some, not yet
all) cases where tls-required-no ignored a tls/verification error.
the admin can change the requiretls status for messages in the queue. so with
default delivery attempts, when verified tls is required by failing, an admin
could potentially change the field to "tls-required: no"-behaviour.
messages received (over smtp) with the requiretls option, get a comment added
to their Received header line, just before "id", after "with".
2023-10-24 11:06:16 +03:00
"github.com/mjl-/mox/metrics"
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
"github.com/mjl-/mox/mox-"
"github.com/mjl-/mox/moxio"
"github.com/mjl-/mox/moxvar"
2023-10-15 16:05:20 +03:00
"github.com/mjl-/mox/mtasts"
"github.com/mjl-/mox/mtastsdb"
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
"github.com/mjl-/mox/queue"
"github.com/mjl-/mox/smtp"
2023-10-15 16:05:20 +03:00
"github.com/mjl-/mox/smtpclient"
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
"github.com/mjl-/mox/store"
replace http basic auth for web interfaces with session cookie & csrf-based auth
the http basic auth we had was very simple to reason about, and to implement.
but it has a major downside:
there is no way to logout, browsers keep sending credentials. ideally, browsers
themselves would show a button to stop sending credentials.
a related downside: the http auth mechanism doesn't indicate for which server
paths the credentials are.
another downside: the original password is sent to the server with each
request. though sending original passwords to web servers seems to be
considered normal.
our new approach uses session cookies, along with csrf values when we can. the
sessions are server-side managed, automatically extended on each use. this
makes it easy to invalidate sessions and keeps the frontend simpler (than with
long- vs short-term sessions and refreshing). the cookies are httponly,
samesite=strict, scoped to the path of the web interface. cookies are set
"secure" when set over https. the cookie is set by a successful call to Login.
a call to Logout invalidates a session. changing a password invalidates all
sessions for a user, but keeps the session with which the password was changed
alive. the csrf value is also random, and associated with the session cookie.
the csrf must be sent as header for api calls, or as parameter for direct form
posts (where we cannot set a custom header). rest-like calls made directly by
the browser, e.g. for images, don't have a csrf protection. the csrf value is
returned by the Login api call and stored in localstorage.
api calls without credentials return code "user:noAuth", and with bad
credentials return "user:badAuth". the api client recognizes this and triggers
a login. after a login, all auth-failed api calls are automatically retried.
only for "user:badAuth" is an error message displayed in the login form (e.g.
session expired).
in an ideal world, browsers would take care of most session management. a
server would indicate authentication is needed (like http basic auth), and the
browsers uses trusted ui to request credentials for the server & path. the
browser could use safer mechanism than sending original passwords to the
server, such as scram, along with a standard way to create sessions. for now,
web developers have to do authentication themselves: from showing the login
prompt, ensuring the right session/csrf cookies/localstorage/headers/etc are
sent with each request.
webauthn is a newer way to do authentication, perhaps we'll implement it in the
future. though hardware tokens aren't an attractive option for many users, and
it may be overkill as long as we still do old-fashioned authentication in smtp
& imap where passwords can be sent to the server.
for issue #58
2024-01-04 15:10:48 +03:00
"github.com/mjl-/mox/webauth"
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
)
//go:embed api.json
var webmailapiJSON [ ] byte
type Webmail struct {
replace http basic auth for web interfaces with session cookie & csrf-based auth
the http basic auth we had was very simple to reason about, and to implement.
but it has a major downside:
there is no way to logout, browsers keep sending credentials. ideally, browsers
themselves would show a button to stop sending credentials.
a related downside: the http auth mechanism doesn't indicate for which server
paths the credentials are.
another downside: the original password is sent to the server with each
request. though sending original passwords to web servers seems to be
considered normal.
our new approach uses session cookies, along with csrf values when we can. the
sessions are server-side managed, automatically extended on each use. this
makes it easy to invalidate sessions and keeps the frontend simpler (than with
long- vs short-term sessions and refreshing). the cookies are httponly,
samesite=strict, scoped to the path of the web interface. cookies are set
"secure" when set over https. the cookie is set by a successful call to Login.
a call to Logout invalidates a session. changing a password invalidates all
sessions for a user, but keeps the session with which the password was changed
alive. the csrf value is also random, and associated with the session cookie.
the csrf must be sent as header for api calls, or as parameter for direct form
posts (where we cannot set a custom header). rest-like calls made directly by
the browser, e.g. for images, don't have a csrf protection. the csrf value is
returned by the Login api call and stored in localstorage.
api calls without credentials return code "user:noAuth", and with bad
credentials return "user:badAuth". the api client recognizes this and triggers
a login. after a login, all auth-failed api calls are automatically retried.
only for "user:badAuth" is an error message displayed in the login form (e.g.
session expired).
in an ideal world, browsers would take care of most session management. a
server would indicate authentication is needed (like http basic auth), and the
browsers uses trusted ui to request credentials for the server & path. the
browser could use safer mechanism than sending original passwords to the
server, such as scram, along with a standard way to create sessions. for now,
web developers have to do authentication themselves: from showing the login
prompt, ensuring the right session/csrf cookies/localstorage/headers/etc are
sent with each request.
webauthn is a newer way to do authentication, perhaps we'll implement it in the
future. though hardware tokens aren't an attractive option for many users, and
it may be overkill as long as we still do old-fashioned authentication in smtp
& imap where passwords can be sent to the server.
for issue #58
2024-01-04 15:10:48 +03:00
maxMessageSize int64 // From listener.
cookiePath string // From listener.
isForwarded bool // From listener, whether we look at X-Forwarded-* headers.
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
}
func mustParseAPI ( api string , buf [ ] byte ) ( doc sherpadoc . Section ) {
err := json . Unmarshal ( buf , & doc )
if err != nil {
2023-12-05 15:35:58 +03:00
pkglog . Fatalx ( "parsing webmail api docs" , err , slog . String ( "api" , api ) )
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
}
return doc
}
var webmailDoc = mustParseAPI ( "webmail" , webmailapiJSON )
var sherpaHandlerOpts * sherpa . HandlerOpts
replace http basic auth for web interfaces with session cookie & csrf-based auth
the http basic auth we had was very simple to reason about, and to implement.
but it has a major downside:
there is no way to logout, browsers keep sending credentials. ideally, browsers
themselves would show a button to stop sending credentials.
a related downside: the http auth mechanism doesn't indicate for which server
paths the credentials are.
another downside: the original password is sent to the server with each
request. though sending original passwords to web servers seems to be
considered normal.
our new approach uses session cookies, along with csrf values when we can. the
sessions are server-side managed, automatically extended on each use. this
makes it easy to invalidate sessions and keeps the frontend simpler (than with
long- vs short-term sessions and refreshing). the cookies are httponly,
samesite=strict, scoped to the path of the web interface. cookies are set
"secure" when set over https. the cookie is set by a successful call to Login.
a call to Logout invalidates a session. changing a password invalidates all
sessions for a user, but keeps the session with which the password was changed
alive. the csrf value is also random, and associated with the session cookie.
the csrf must be sent as header for api calls, or as parameter for direct form
posts (where we cannot set a custom header). rest-like calls made directly by
the browser, e.g. for images, don't have a csrf protection. the csrf value is
returned by the Login api call and stored in localstorage.
api calls without credentials return code "user:noAuth", and with bad
credentials return "user:badAuth". the api client recognizes this and triggers
a login. after a login, all auth-failed api calls are automatically retried.
only for "user:badAuth" is an error message displayed in the login form (e.g.
session expired).
in an ideal world, browsers would take care of most session management. a
server would indicate authentication is needed (like http basic auth), and the
browsers uses trusted ui to request credentials for the server & path. the
browser could use safer mechanism than sending original passwords to the
server, such as scram, along with a standard way to create sessions. for now,
web developers have to do authentication themselves: from showing the login
prompt, ensuring the right session/csrf cookies/localstorage/headers/etc are
sent with each request.
webauthn is a newer way to do authentication, perhaps we'll implement it in the
future. though hardware tokens aren't an attractive option for many users, and
it may be overkill as long as we still do old-fashioned authentication in smtp
& imap where passwords can be sent to the server.
for issue #58
2024-01-04 15:10:48 +03:00
func makeSherpaHandler ( maxMessageSize int64 , cookiePath string , isForwarded bool ) ( http . Handler , error ) {
return sherpa . NewHandler ( "/api/" , moxvar . Version , Webmail { maxMessageSize , cookiePath , isForwarded } , & webmailDoc , sherpaHandlerOpts )
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
}
func init ( ) {
collector , err := sherpaprom . NewCollector ( "moxwebmail" , nil )
if err != nil {
2023-12-05 15:35:58 +03:00
pkglog . Fatalx ( "creating sherpa prometheus collector" , err )
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
}
replace http basic auth for web interfaces with session cookie & csrf-based auth
the http basic auth we had was very simple to reason about, and to implement.
but it has a major downside:
there is no way to logout, browsers keep sending credentials. ideally, browsers
themselves would show a button to stop sending credentials.
a related downside: the http auth mechanism doesn't indicate for which server
paths the credentials are.
another downside: the original password is sent to the server with each
request. though sending original passwords to web servers seems to be
considered normal.
our new approach uses session cookies, along with csrf values when we can. the
sessions are server-side managed, automatically extended on each use. this
makes it easy to invalidate sessions and keeps the frontend simpler (than with
long- vs short-term sessions and refreshing). the cookies are httponly,
samesite=strict, scoped to the path of the web interface. cookies are set
"secure" when set over https. the cookie is set by a successful call to Login.
a call to Logout invalidates a session. changing a password invalidates all
sessions for a user, but keeps the session with which the password was changed
alive. the csrf value is also random, and associated with the session cookie.
the csrf must be sent as header for api calls, or as parameter for direct form
posts (where we cannot set a custom header). rest-like calls made directly by
the browser, e.g. for images, don't have a csrf protection. the csrf value is
returned by the Login api call and stored in localstorage.
api calls without credentials return code "user:noAuth", and with bad
credentials return "user:badAuth". the api client recognizes this and triggers
a login. after a login, all auth-failed api calls are automatically retried.
only for "user:badAuth" is an error message displayed in the login form (e.g.
session expired).
in an ideal world, browsers would take care of most session management. a
server would indicate authentication is needed (like http basic auth), and the
browsers uses trusted ui to request credentials for the server & path. the
browser could use safer mechanism than sending original passwords to the
server, such as scram, along with a standard way to create sessions. for now,
web developers have to do authentication themselves: from showing the login
prompt, ensuring the right session/csrf cookies/localstorage/headers/etc are
sent with each request.
webauthn is a newer way to do authentication, perhaps we'll implement it in the
future. though hardware tokens aren't an attractive option for many users, and
it may be overkill as long as we still do old-fashioned authentication in smtp
& imap where passwords can be sent to the server.
for issue #58
2024-01-04 15:10:48 +03:00
sherpaHandlerOpts = & sherpa . HandlerOpts { Collector : collector , AdjustFunctionNames : "none" , NoCORS : true }
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
// Just to validate.
replace http basic auth for web interfaces with session cookie & csrf-based auth
the http basic auth we had was very simple to reason about, and to implement.
but it has a major downside:
there is no way to logout, browsers keep sending credentials. ideally, browsers
themselves would show a button to stop sending credentials.
a related downside: the http auth mechanism doesn't indicate for which server
paths the credentials are.
another downside: the original password is sent to the server with each
request. though sending original passwords to web servers seems to be
considered normal.
our new approach uses session cookies, along with csrf values when we can. the
sessions are server-side managed, automatically extended on each use. this
makes it easy to invalidate sessions and keeps the frontend simpler (than with
long- vs short-term sessions and refreshing). the cookies are httponly,
samesite=strict, scoped to the path of the web interface. cookies are set
"secure" when set over https. the cookie is set by a successful call to Login.
a call to Logout invalidates a session. changing a password invalidates all
sessions for a user, but keeps the session with which the password was changed
alive. the csrf value is also random, and associated with the session cookie.
the csrf must be sent as header for api calls, or as parameter for direct form
posts (where we cannot set a custom header). rest-like calls made directly by
the browser, e.g. for images, don't have a csrf protection. the csrf value is
returned by the Login api call and stored in localstorage.
api calls without credentials return code "user:noAuth", and with bad
credentials return "user:badAuth". the api client recognizes this and triggers
a login. after a login, all auth-failed api calls are automatically retried.
only for "user:badAuth" is an error message displayed in the login form (e.g.
session expired).
in an ideal world, browsers would take care of most session management. a
server would indicate authentication is needed (like http basic auth), and the
browsers uses trusted ui to request credentials for the server & path. the
browser could use safer mechanism than sending original passwords to the
server, such as scram, along with a standard way to create sessions. for now,
web developers have to do authentication themselves: from showing the login
prompt, ensuring the right session/csrf cookies/localstorage/headers/etc are
sent with each request.
webauthn is a newer way to do authentication, perhaps we'll implement it in the
future. though hardware tokens aren't an attractive option for many users, and
it may be overkill as long as we still do old-fashioned authentication in smtp
& imap where passwords can be sent to the server.
for issue #58
2024-01-04 15:10:48 +03:00
_ , err = makeSherpaHandler ( 0 , "" , false )
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
if err != nil {
2023-12-05 15:35:58 +03:00
pkglog . Fatalx ( "sherpa handler" , err )
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
}
}
replace http basic auth for web interfaces with session cookie & csrf-based auth
the http basic auth we had was very simple to reason about, and to implement.
but it has a major downside:
there is no way to logout, browsers keep sending credentials. ideally, browsers
themselves would show a button to stop sending credentials.
a related downside: the http auth mechanism doesn't indicate for which server
paths the credentials are.
another downside: the original password is sent to the server with each
request. though sending original passwords to web servers seems to be
considered normal.
our new approach uses session cookies, along with csrf values when we can. the
sessions are server-side managed, automatically extended on each use. this
makes it easy to invalidate sessions and keeps the frontend simpler (than with
long- vs short-term sessions and refreshing). the cookies are httponly,
samesite=strict, scoped to the path of the web interface. cookies are set
"secure" when set over https. the cookie is set by a successful call to Login.
a call to Logout invalidates a session. changing a password invalidates all
sessions for a user, but keeps the session with which the password was changed
alive. the csrf value is also random, and associated with the session cookie.
the csrf must be sent as header for api calls, or as parameter for direct form
posts (where we cannot set a custom header). rest-like calls made directly by
the browser, e.g. for images, don't have a csrf protection. the csrf value is
returned by the Login api call and stored in localstorage.
api calls without credentials return code "user:noAuth", and with bad
credentials return "user:badAuth". the api client recognizes this and triggers
a login. after a login, all auth-failed api calls are automatically retried.
only for "user:badAuth" is an error message displayed in the login form (e.g.
session expired).
in an ideal world, browsers would take care of most session management. a
server would indicate authentication is needed (like http basic auth), and the
browsers uses trusted ui to request credentials for the server & path. the
browser could use safer mechanism than sending original passwords to the
server, such as scram, along with a standard way to create sessions. for now,
web developers have to do authentication themselves: from showing the login
prompt, ensuring the right session/csrf cookies/localstorage/headers/etc are
sent with each request.
webauthn is a newer way to do authentication, perhaps we'll implement it in the
future. though hardware tokens aren't an attractive option for many users, and
it may be overkill as long as we still do old-fashioned authentication in smtp
& imap where passwords can be sent to the server.
for issue #58
2024-01-04 15:10:48 +03:00
// LoginPrep returns a login token, and also sets it as cookie. Both must be
// present in the call to Login.
func ( w Webmail ) LoginPrep ( ctx context . Context ) string {
log := pkglog . WithContext ( ctx )
reqInfo := ctx . Value ( requestInfoCtxKey ) . ( requestInfo )
var data [ 8 ] byte
_ , err := cryptorand . Read ( data [ : ] )
xcheckf ( ctx , err , "generate token" )
loginToken := base64 . RawURLEncoding . EncodeToString ( data [ : ] )
webauth . LoginPrep ( ctx , log , "webmail" , w . cookiePath , w . isForwarded , reqInfo . Response , reqInfo . Request , loginToken )
return loginToken
}
// Login returns a session token for the credentials, or fails with error code
// "user:badLogin". Call LoginPrep to get a loginToken.
func ( w Webmail ) Login ( ctx context . Context , loginToken , username , password string ) store . CSRFToken {
log := pkglog . WithContext ( ctx )
reqInfo := ctx . Value ( requestInfoCtxKey ) . ( requestInfo )
csrfToken , err := webauth . Login ( ctx , log , webauth . Accounts , "webmail" , w . cookiePath , w . isForwarded , reqInfo . Response , reqInfo . Request , loginToken , username , password )
if _ , ok := err . ( * sherpa . Error ) ; ok {
panic ( err )
}
xcheckf ( ctx , err , "login" )
return csrfToken
}
// Logout invalidates the session token.
func ( w Webmail ) Logout ( ctx context . Context ) {
log := pkglog . WithContext ( ctx )
reqInfo := ctx . Value ( requestInfoCtxKey ) . ( requestInfo )
err := webauth . Logout ( ctx , log , webauth . Accounts , "webmail" , w . cookiePath , w . isForwarded , reqInfo . Response , reqInfo . Request , reqInfo . AccountName , reqInfo . SessionToken )
xcheckf ( ctx , err , "logout" )
}
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
// Token returns a token to use for an SSE connection. A token can only be used for
// a single SSE connection. Tokens are stored in memory for a maximum of 1 minute,
// with at most 10 unused tokens (the most recently created) per account.
func ( Webmail ) Token ( ctx context . Context ) string {
reqInfo := ctx . Value ( requestInfoCtxKey ) . ( requestInfo )
replace http basic auth for web interfaces with session cookie & csrf-based auth
the http basic auth we had was very simple to reason about, and to implement.
but it has a major downside:
there is no way to logout, browsers keep sending credentials. ideally, browsers
themselves would show a button to stop sending credentials.
a related downside: the http auth mechanism doesn't indicate for which server
paths the credentials are.
another downside: the original password is sent to the server with each
request. though sending original passwords to web servers seems to be
considered normal.
our new approach uses session cookies, along with csrf values when we can. the
sessions are server-side managed, automatically extended on each use. this
makes it easy to invalidate sessions and keeps the frontend simpler (than with
long- vs short-term sessions and refreshing). the cookies are httponly,
samesite=strict, scoped to the path of the web interface. cookies are set
"secure" when set over https. the cookie is set by a successful call to Login.
a call to Logout invalidates a session. changing a password invalidates all
sessions for a user, but keeps the session with which the password was changed
alive. the csrf value is also random, and associated with the session cookie.
the csrf must be sent as header for api calls, or as parameter for direct form
posts (where we cannot set a custom header). rest-like calls made directly by
the browser, e.g. for images, don't have a csrf protection. the csrf value is
returned by the Login api call and stored in localstorage.
api calls without credentials return code "user:noAuth", and with bad
credentials return "user:badAuth". the api client recognizes this and triggers
a login. after a login, all auth-failed api calls are automatically retried.
only for "user:badAuth" is an error message displayed in the login form (e.g.
session expired).
in an ideal world, browsers would take care of most session management. a
server would indicate authentication is needed (like http basic auth), and the
browsers uses trusted ui to request credentials for the server & path. the
browser could use safer mechanism than sending original passwords to the
server, such as scram, along with a standard way to create sessions. for now,
web developers have to do authentication themselves: from showing the login
prompt, ensuring the right session/csrf cookies/localstorage/headers/etc are
sent with each request.
webauthn is a newer way to do authentication, perhaps we'll implement it in the
future. though hardware tokens aren't an attractive option for many users, and
it may be overkill as long as we still do old-fashioned authentication in smtp
& imap where passwords can be sent to the server.
for issue #58
2024-01-04 15:10:48 +03:00
return sseTokens . xgenerate ( ctx , reqInfo . AccountName , reqInfo . LoginAddress , reqInfo . SessionToken )
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
}
// Requests sends a new request for an open SSE connection. Any currently active
// request for the connection will be canceled, but this is done asynchrously, so
// the SSE connection may still send results for the previous request. Callers
// should take care to ignore such results. If req.Cancel is set, no new request is
// started.
func ( Webmail ) Request ( ctx context . Context , req Request ) {
reqInfo := ctx . Value ( requestInfoCtxKey ) . ( requestInfo )
if ! req . Cancel && req . Page . Count <= 0 {
xcheckuserf ( ctx , errors . New ( "Page.Count must be >= 1" ) , "checking request" )
}
sse , ok := sseGet ( req . SSEID , reqInfo . AccountName )
if ! ok {
xcheckuserf ( ctx , errors . New ( "unknown sseid" ) , "looking up connection" )
}
sse . Request <- req
}
// ParsedMessage returns enough to render the textual body of a message. It is
// assumed the client already has other fields through MessageItem.
func ( Webmail ) ParsedMessage ( ctx context . Context , msgID int64 ) ( pm ParsedMessage ) {
2023-12-05 15:35:58 +03:00
log := pkglog . WithContext ( ctx )
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
reqInfo := ctx . Value ( requestInfoCtxKey ) . ( requestInfo )
2023-12-05 15:35:58 +03:00
acc , err := store . OpenAccount ( log , reqInfo . AccountName )
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
xcheckf ( ctx , err , "open account" )
defer func ( ) {
err := acc . Close ( )
log . Check ( err , "closing account" )
} ( )
var m store . Message
xdbread ( ctx , acc , func ( tx * bstore . Tx ) {
m = xmessageID ( ctx , tx , msgID )
} )
state := msgState { acc : acc }
defer state . clear ( )
pm , err = parsedMessage ( log , m , & state , true , false )
xcheckf ( ctx , err , "parsing message" )
return
}
// Attachment is a MIME part is an existing message that is not intended as
// viewable text or HTML part.
type Attachment struct {
Path [ ] int // Indices into top-level message.Part.Parts.
// File name based on "name" attribute of "Content-Type", or the "filename"
// attribute of "Content-Disposition".
Filename string
Part message . Part
}
// SubmitMessage is an email message to be sent to one or more recipients.
// Addresses are formatted as just email address, or with a name like "name
// <user@host>".
type SubmitMessage struct {
From string
To [ ] string
Cc [ ] string
Bcc [ ] string
Subject string
TextBody string
Attachments [ ] File
ForwardAttachments ForwardAttachments
IsForward bool
2024-02-10 19:55:56 +03:00
ResponseMessageID int64 // If set, this was a reply or forward, based on IsForward.
ReplyTo string // If non-empty, Reply-To header to add to message.
UserAgent string // User-Agent header added if not empty.
RequireTLS * bool // For "Require TLS" extension during delivery.
FutureRelease * time . Time // If set, time (in the future) when message should be delivered from queue.
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
}
// ForwardAttachments references attachments by a list of message.Part paths.
type ForwardAttachments struct {
MessageID int64 // Only relevant if MessageID is not 0.
Paths [ ] [ ] int // List of attachments, each path is a list of indices into the top-level message.Part.Parts.
}
// File is a new attachment (not from an existing message that is being
// forwarded) to send with a SubmitMessage.
type File struct {
Filename string
DataURI string // Full data of the attachment, with base64 encoding and including content-type.
}
// parseAddress expects either a plain email address like "user@domain", or a
// single address as used in a message header, like "name <user@domain>".
2023-11-09 23:15:27 +03:00
func parseAddress ( msghdr string ) ( message . NameAddress , error ) {
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
a , err := mail . ParseAddress ( msghdr )
if err != nil {
2024-03-09 11:51:24 +03:00
return message . NameAddress { } , err
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
}
// todo: parse more fully according to ../rfc/5322:959
path , err := smtp . ParseAddress ( a . Address )
if err != nil {
2023-11-09 23:15:27 +03:00
return message . NameAddress { } , err
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
}
2023-11-09 23:15:27 +03:00
return message . NameAddress { DisplayName : a . Name , Address : path } , nil
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
}
func xmailboxID ( ctx context . Context , tx * bstore . Tx , mailboxID int64 ) store . Mailbox {
if mailboxID == 0 {
xcheckuserf ( ctx , errors . New ( "invalid zero mailbox ID" ) , "getting mailbox" )
}
mb := store . Mailbox { ID : mailboxID }
err := tx . Get ( & mb )
if err == bstore . ErrAbsent {
xcheckuserf ( ctx , err , "getting mailbox" )
}
xcheckf ( ctx , err , "getting mailbox" )
return mb
}
// xmessageID returns a non-expunged message or panics with a sherpa error.
func xmessageID ( ctx context . Context , tx * bstore . Tx , messageID int64 ) store . Message {
if messageID == 0 {
xcheckuserf ( ctx , errors . New ( "invalid zero message id" ) , "getting message" )
}
m := store . Message { ID : messageID }
err := tx . Get ( & m )
if err == bstore . ErrAbsent {
xcheckuserf ( ctx , errors . New ( "message does not exist" ) , "getting message" )
} else if err == nil && m . Expunged {
xcheckuserf ( ctx , errors . New ( "message was removed" ) , "getting message" )
}
xcheckf ( ctx , err , "getting message" )
return m
}
// MessageSubmit sends a message by submitting it the outgoing email queue. The
// message is sent to all addresses listed in the To, Cc and Bcc addresses, without
// Bcc message header.
//
// If a Sent mailbox is configured, messages are added to it after submitting
// to the delivery queue.
func ( w Webmail ) MessageSubmit ( ctx context . Context , m SubmitMessage ) {
// Similar between ../smtpserver/server.go:/submit\( and ../webmail/webmail.go:/MessageSubmit\(
// todo: consider making this an HTTP POST, so we can upload as regular form, which is probably more efficient for encoding for the client and we can stream the data in.
2024-01-01 20:30:31 +03:00
// Prevent any accidental control characters, or attempts at getting bare \r or \n
// into messages.
for _ , l := range [ ] [ ] string { m . To , m . Cc , m . Bcc , { m . From , m . Subject , m . ReplyTo , m . UserAgent } } {
for _ , s := range l {
for _ , c := range s {
if c < 0x20 {
xcheckuserf ( ctx , errors . New ( "control characters not allowed" ) , "checking header values" )
}
}
}
}
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
reqInfo := ctx . Value ( requestInfoCtxKey ) . ( requestInfo )
2023-12-05 15:35:58 +03:00
log := pkglog . WithContext ( ctx ) . With ( slog . String ( "account" , reqInfo . AccountName ) )
acc , err := store . OpenAccount ( log , reqInfo . AccountName )
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
xcheckf ( ctx , err , "open account" )
defer func ( ) {
err := acc . Close ( )
log . Check ( err , "closing account" )
} ( )
log . Debug ( "message submit" )
fromAddr , err := parseAddress ( m . From )
xcheckuserf ( ctx , err , "parsing From address" )
2023-11-09 23:15:27 +03:00
var replyTo * message . NameAddress
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
if m . ReplyTo != "" {
a , err := parseAddress ( m . ReplyTo )
xcheckuserf ( ctx , err , "parsing Reply-To address" )
replyTo = & a
}
var recipients [ ] smtp . Address
2023-11-09 23:15:27 +03:00
var toAddrs [ ] message . NameAddress
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
for _ , s := range m . To {
addr , err := parseAddress ( s )
xcheckuserf ( ctx , err , "parsing To address" )
toAddrs = append ( toAddrs , addr )
recipients = append ( recipients , addr . Address )
}
2023-11-09 23:15:27 +03:00
var ccAddrs [ ] message . NameAddress
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
for _ , s := range m . Cc {
addr , err := parseAddress ( s )
xcheckuserf ( ctx , err , "parsing Cc address" )
ccAddrs = append ( ccAddrs , addr )
recipients = append ( recipients , addr . Address )
}
for _ , s := range m . Bcc {
addr , err := parseAddress ( s )
xcheckuserf ( ctx , err , "parsing Bcc address" )
recipients = append ( recipients , addr . Address )
}
// Check if from address is allowed for account.
fromAccName , _ , _ , err := mox . FindAccount ( fromAddr . Address . Localpart , fromAddr . Address . Domain , false )
if err == nil && fromAccName != reqInfo . AccountName {
err = mox . ErrAccountNotFound
}
if err != nil && ( errors . Is ( err , mox . ErrAccountNotFound ) || errors . Is ( err , mox . ErrDomainNotFound ) ) {
metricSubmission . WithLabelValues ( "badfrom" ) . Inc ( )
2024-03-09 11:51:24 +03:00
xcheckuserf ( ctx , errors . New ( "address not found" ) , ` looking up "from" address for account ` )
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
}
xcheckf ( ctx , err , "checking if from address is allowed" )
if len ( recipients ) == 0 {
2024-03-09 11:51:24 +03:00
xcheckuserf ( ctx , errors . New ( "no recipients" ) , "composing message" )
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
}
// Check outgoing message rate limit.
xdbread ( ctx , acc , func ( tx * bstore . Tx ) {
rcpts := make ( [ ] smtp . Path , len ( recipients ) )
for i , r := range recipients {
rcpts [ i ] = smtp . Path { Localpart : r . Localpart , IPDomain : dns . IPDomain { Domain : r . Domain } }
}
msglimit , rcptlimit , err := acc . SendLimitReached ( tx , rcpts )
if msglimit >= 0 {
metricSubmission . WithLabelValues ( "messagelimiterror" ) . Inc ( )
xcheckuserf ( ctx , errors . New ( "send message limit reached" ) , "checking outgoing rate limit" )
} else if rcptlimit >= 0 {
metricSubmission . WithLabelValues ( "recipientlimiterror" ) . Inc ( )
xcheckuserf ( ctx , errors . New ( "send message limit reached" ) , "checking outgoing rate limit" )
}
xcheckf ( ctx , err , "checking send limit" )
} )
// We only use smtputf8 if we have to, with a utf-8 localpart. For IDNA, we use ASCII domains.
smtputf8 := false
for _ , a := range recipients {
if a . Localpart . IsInternational ( ) {
smtputf8 = true
break
}
}
if ! smtputf8 && fromAddr . Address . Localpart . IsInternational ( ) {
// todo: may want to warn user that they should consider sending with a ascii-only localpart, in case receiver doesn't support smtputf8.
smtputf8 = true
}
// Create file to compose message into.
2023-12-05 15:35:58 +03:00
dataFile , err := store . CreateMessageTemp ( log , "webmail-submit" )
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
xcheckf ( ctx , err , "creating temporary file for message" )
2023-11-01 20:57:38 +03:00
defer store . CloseRemoveTempFile ( log , dataFile , "message to submit" )
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
// If writing to the message file fails, we abort immediately.
2024-03-16 22:54:10 +03:00
xc := message . NewComposer ( dataFile , w . maxMessageSize , smtputf8 )
2023-11-09 23:15:27 +03:00
defer func ( ) {
x := recover ( )
if x == nil {
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
return
}
2023-11-09 23:15:27 +03:00
if err , ok := x . ( error ) ; ok && errors . Is ( err , message . ErrMessageSize ) {
xcheckuserf ( ctx , err , "making message" )
} else if ok && errors . Is ( err , message . ErrCompose ) {
xcheckf ( ctx , err , "making message" )
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
}
2023-11-09 23:15:27 +03:00
panic ( x )
} ( )
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
// todo spec: can we add an Authentication-Results header that indicates this is an authenticated message? the "auth" method is for SMTP AUTH, which this isn't. ../rfc/8601 https://www.iana.org/assignments/email-auth/email-auth.xhtml
// Each queued message gets a Received header.
// We don't have access to the local IP for adding.
// We cannot use VIA, because there is no registered method. We would like to use
// it to add the ascii domain name in case of smtputf8 and IDNA host name.
recvFrom := message . HeaderCommentDomain ( mox . Conf . Static . HostnameDomain , smtputf8 )
recvBy := mox . Conf . Static . HostnameDomain . XName ( smtputf8 )
recvID := mox . ReceivedID ( mox . CidFromCtx ( ctx ) )
recvHdrFor := func ( rcptTo string ) string {
recvHdr := & message . HeaderWriter { }
// For additional Received-header clauses, see:
// https://www.iana.org/assignments/mail-parameters/mail-parameters.xhtml#table-mail-parameters-8
// Note: we don't have "via" or "with", there is no registered for webmail.
recvHdr . Add ( " " , "Received:" , "from" , recvFrom , "by" , recvBy , "id" , recvID ) // ../rfc/5321:3158
if reqInfo . Request . TLS != nil {
2023-12-05 15:35:58 +03:00
recvHdr . Add ( " " , mox . TLSReceivedComment ( log , * reqInfo . Request . TLS ) ... )
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
}
recvHdr . Add ( " " , "for" , "<" + rcptTo + ">;" , time . Now ( ) . Format ( message . RFC5322Z ) )
return recvHdr . String ( )
}
// Outer message headers.
2023-11-09 23:15:27 +03:00
xc . HeaderAddrs ( "From" , [ ] message . NameAddress { fromAddr } )
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
if replyTo != nil {
2023-11-09 23:15:27 +03:00
xc . HeaderAddrs ( "Reply-To" , [ ] message . NameAddress { * replyTo } )
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
}
2023-11-09 23:15:27 +03:00
xc . HeaderAddrs ( "To" , toAddrs )
xc . HeaderAddrs ( "Cc" , ccAddrs )
if m . Subject != "" {
xc . Subject ( m . Subject )
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
}
messageID := fmt . Sprintf ( "<%s>" , mox . MessageIDGen ( smtputf8 ) )
2023-11-09 23:15:27 +03:00
xc . Header ( "Message-Id" , messageID )
xc . Header ( "Date" , time . Now ( ) . Format ( message . RFC5322Z ) )
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
// Add In-Reply-To and References headers.
if m . ResponseMessageID > 0 {
xdbread ( ctx , acc , func ( tx * bstore . Tx ) {
rm := xmessageID ( ctx , tx , m . ResponseMessageID )
msgr := acc . MessageReader ( rm )
defer func ( ) {
err := msgr . Close ( )
log . Check ( err , "closing message reader" )
} ( )
rp , err := rm . LoadPart ( msgr )
xcheckf ( ctx , err , "load parsed message" )
h , err := rp . Header ( )
xcheckf ( ctx , err , "parsing header" )
if rp . Envelope == nil {
return
}
2023-11-09 23:15:27 +03:00
xc . Header ( "In-Reply-To" , rp . Envelope . MessageID )
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
ref := h . Get ( "References" )
if ref == "" {
ref = h . Get ( "In-Reply-To" )
}
if ref != "" {
2023-11-09 23:15:27 +03:00
xc . Header ( "References" , ref + "\r\n\t" + rp . Envelope . MessageID )
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
} else {
2023-11-09 23:15:27 +03:00
xc . Header ( "References" , rp . Envelope . MessageID )
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
}
} )
}
if m . UserAgent != "" {
2023-11-09 23:15:27 +03:00
xc . Header ( "User-Agent" , m . UserAgent )
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
}
implement "requiretls", rfc 8689
with requiretls, the tls verification mode/rules for email deliveries can be
changed by the sender/submitter. in two ways:
1. "requiretls" smtp extension to always enforce verified tls (with mta-sts or
dnssec+dane), along the entire delivery path until delivery into the final
destination mailbox (so entire transport is verified-tls-protected).
2. "tls-required: no" message header, to ignore any tls and tls verification
errors even if the recipient domain has a policy that requires tls verification
(mta-sts and/or dnssec+dane), allowing delivery of non-sensitive messages in
case of misconfiguration/interoperability issues (at least useful for sending
tls reports).
we enable requiretls by default (only when tls is active), for smtp and
submission. it can be disabled through the config.
for each delivery attempt, we now store (per recipient domain, in the account
of the sender) whether the smtp server supports starttls and requiretls. this
support is shown (after having sent a first message) in the webmail when
sending a message (the previous 3 bars under the address input field are now 5
bars, the first for starttls support, the last for requiretls support). when
all recipient domains for a message are known to implement requiretls,
requiretls is automatically selected for sending (instead of "default" tls
behaviour). users can also select the "fallback to insecure" to add the
"tls-required: no" header.
new metrics are added for insight into requiretls errors and (some, not yet
all) cases where tls-required-no ignored a tls/verification error.
the admin can change the requiretls status for messages in the queue. so with
default delivery attempts, when verified tls is required by failing, an admin
could potentially change the field to "tls-required: no"-behaviour.
messages received (over smtp) with the requiretls option, get a comment added
to their Received header line, just before "id", after "with".
2023-10-24 11:06:16 +03:00
if m . RequireTLS != nil && ! * m . RequireTLS {
2023-11-09 23:15:27 +03:00
xc . Header ( "TLS-Required" , "No" )
implement "requiretls", rfc 8689
with requiretls, the tls verification mode/rules for email deliveries can be
changed by the sender/submitter. in two ways:
1. "requiretls" smtp extension to always enforce verified tls (with mta-sts or
dnssec+dane), along the entire delivery path until delivery into the final
destination mailbox (so entire transport is verified-tls-protected).
2. "tls-required: no" message header, to ignore any tls and tls verification
errors even if the recipient domain has a policy that requires tls verification
(mta-sts and/or dnssec+dane), allowing delivery of non-sensitive messages in
case of misconfiguration/interoperability issues (at least useful for sending
tls reports).
we enable requiretls by default (only when tls is active), for smtp and
submission. it can be disabled through the config.
for each delivery attempt, we now store (per recipient domain, in the account
of the sender) whether the smtp server supports starttls and requiretls. this
support is shown (after having sent a first message) in the webmail when
sending a message (the previous 3 bars under the address input field are now 5
bars, the first for starttls support, the last for requiretls support). when
all recipient domains for a message are known to implement requiretls,
requiretls is automatically selected for sending (instead of "default" tls
behaviour). users can also select the "fallback to insecure" to add the
"tls-required: no" header.
new metrics are added for insight into requiretls errors and (some, not yet
all) cases where tls-required-no ignored a tls/verification error.
the admin can change the requiretls status for messages in the queue. so with
default delivery attempts, when verified tls is required by failing, an admin
could potentially change the field to "tls-required: no"-behaviour.
messages received (over smtp) with the requiretls option, get a comment added
to their Received header line, just before "id", after "with".
2023-10-24 11:06:16 +03:00
}
2023-11-09 23:15:27 +03:00
xc . Header ( "MIME-Version" , "1.0" )
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
if len ( m . Attachments ) > 0 || len ( m . ForwardAttachments . Paths ) > 0 {
2023-11-09 23:15:27 +03:00
mp := multipart . NewWriter ( xc )
xc . Header ( "Content-Type" , fmt . Sprintf ( ` multipart/mixed; boundary="%s" ` , mp . Boundary ( ) ) )
xc . Line ( )
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
2023-11-09 23:15:27 +03:00
textBody , ct , cte := xc . TextPart ( m . TextBody )
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
textHdr := textproto . MIMEHeader { }
2023-10-14 15:14:13 +03:00
textHdr . Set ( "Content-Type" , ct )
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
textHdr . Set ( "Content-Transfer-Encoding" , cte )
2023-10-14 15:14:13 +03:00
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
textp , err := mp . CreatePart ( textHdr )
xcheckf ( ctx , err , "adding text part to message" )
2023-11-09 23:15:27 +03:00
_ , err = textp . Write ( textBody )
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
xcheckf ( ctx , err , "writing text part" )
xaddPart := func ( ct , filename string ) io . Writer {
ahdr := textproto . MIMEHeader { }
2023-10-14 15:14:13 +03:00
cd := mime . FormatMediaType ( "attachment" , map [ string ] string { "filename" : filename } )
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
ahdr . Set ( "Content-Type" , ct )
ahdr . Set ( "Content-Transfer-Encoding" , "base64" )
2023-10-14 15:14:13 +03:00
ahdr . Set ( "Content-Disposition" , cd )
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
ap , err := mp . CreatePart ( ahdr )
xcheckf ( ctx , err , "adding attachment part to message" )
return ap
}
xaddAttachmentBase64 := func ( ct , filename string , base64Data [ ] byte ) {
ap := xaddPart ( ct , filename )
for len ( base64Data ) > 0 {
line := base64Data
n := len ( line )
if n > 78 {
n = 78
}
line , base64Data = base64Data [ : n ] , base64Data [ n : ]
_ , err := ap . Write ( line )
xcheckf ( ctx , err , "writing attachment" )
_ , err = ap . Write ( [ ] byte ( "\r\n" ) )
xcheckf ( ctx , err , "writing attachment" )
}
}
xaddAttachment := func ( ct , filename string , r io . Reader ) {
ap := xaddPart ( ct , filename )
wc := moxio . Base64Writer ( ap )
_ , err := io . Copy ( wc , r )
xcheckf ( ctx , err , "adding attachment" )
err = wc . Close ( )
xcheckf ( ctx , err , "flushing attachment" )
}
for _ , a := range m . Attachments {
s := a . DataURI
if ! strings . HasPrefix ( s , "data:" ) {
xcheckuserf ( ctx , errors . New ( "missing data: in datauri" ) , "parsing attachment" )
}
s = s [ len ( "data:" ) : ]
t := strings . SplitN ( s , "," , 2 )
if len ( t ) != 2 {
xcheckuserf ( ctx , errors . New ( "missing comma in datauri" ) , "parsing attachment" )
}
if ! strings . HasSuffix ( t [ 0 ] , "base64" ) {
xcheckuserf ( ctx , errors . New ( "missing base64 in datauri" ) , "parsing attachment" )
}
ct := strings . TrimSuffix ( t [ 0 ] , "base64" )
ct = strings . TrimSuffix ( ct , ";" )
2023-10-14 15:14:13 +03:00
if ct == "" {
ct = "application/octet-stream"
}
filename := a . Filename
if filename == "" {
filename = "unnamed.bin"
}
params := map [ string ] string { "name" : filename }
ct = mime . FormatMediaType ( ct , params )
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
// Ensure base64 is valid, then we'll write the original string.
_ , err := io . Copy ( io . Discard , base64 . NewDecoder ( base64 . StdEncoding , strings . NewReader ( t [ 1 ] ) ) )
xcheckuserf ( ctx , err , "parsing attachment as base64" )
2023-10-14 15:14:13 +03:00
xaddAttachmentBase64 ( ct , filename , [ ] byte ( t [ 1 ] ) )
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
}
if len ( m . ForwardAttachments . Paths ) > 0 {
acc . WithRLock ( func ( ) {
xdbread ( ctx , acc , func ( tx * bstore . Tx ) {
fm := xmessageID ( ctx , tx , m . ForwardAttachments . MessageID )
msgr := acc . MessageReader ( fm )
defer func ( ) {
err := msgr . Close ( )
log . Check ( err , "closing message reader" )
} ( )
fp , err := fm . LoadPart ( msgr )
xcheckf ( ctx , err , "load parsed message" )
for _ , path := range m . ForwardAttachments . Paths {
ap := fp
for _ , xp := range path {
if xp < 0 || xp >= len ( ap . Parts ) {
xcheckuserf ( ctx , errors . New ( "unknown part" ) , "looking up attachment" )
}
ap = ap . Parts [ xp ]
}
2023-10-14 15:14:13 +03:00
filename := tryDecodeParam ( log , ap . ContentTypeParams [ "name" ] )
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
if filename == "" {
filename = "unnamed.bin"
}
2023-10-14 15:14:13 +03:00
params := map [ string ] string { "name" : filename }
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
if pcharset := ap . ContentTypeParams [ "charset" ] ; pcharset != "" {
2023-10-14 15:14:13 +03:00
params [ "charset" ] = pcharset
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
}
2023-10-14 15:14:13 +03:00
ct := strings . ToLower ( ap . MediaType + "/" + ap . MediaSubType )
ct = mime . FormatMediaType ( ct , params )
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
xaddAttachment ( ct , filename , ap . Reader ( ) )
}
} )
} )
}
err = mp . Close ( )
xcheckf ( ctx , err , "writing mime multipart" )
} else {
2023-11-09 23:15:27 +03:00
textBody , ct , cte := xc . TextPart ( m . TextBody )
xc . Header ( "Content-Type" , ct )
xc . Header ( "Content-Transfer-Encoding" , cte )
xc . Line ( )
xc . Write ( [ ] byte ( textBody ) )
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
}
2023-11-09 23:15:27 +03:00
xc . Flush ( )
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
// Add DKIM-Signature headers.
var msgPrefix string
fd := fromAddr . Address . Domain
confDom , _ := mox . Conf . Domain ( fd )
2023-12-05 23:13:57 +03:00
selectors := mox . DKIMSelectors ( confDom . DKIM )
if len ( selectors ) > 0 {
dkimHeaders , err := dkim . Sign ( ctx , log . Logger , fromAddr . Address . Localpart , fd , selectors , smtputf8 , dataFile )
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
if err != nil {
metricServerErrors . WithLabelValues ( "dkimsign" ) . Inc ( )
}
xcheckf ( ctx , err , "sign dkim" )
msgPrefix = dkimHeaders
}
fromPath := smtp . Path {
Localpart : fromAddr . Address . Localpart ,
IPDomain : dns . IPDomain { Domain : fromAddr . Address . Domain } ,
}
2024-03-05 22:10:28 +03:00
qml := make ( [ ] queue . Msg , len ( recipients ) )
queue: deliver to multiple recipients in a single smtp transaction
transferring the data only once. we only do this when the recipient domains
are the same. when queuing, we now take care to set the same NextAttempt
timestamp, so queued messages are actually eligable for combined delivery.
this adds a DeliverMultiple to the smtp client. for pipelined requests, it will
send all RCPT TO (and MAIL and DATA) in one go, and handles the various
responses and error conditions, returning either an overal error, or per
recipient smtp responses. the results of the smtp LIMITS extension are also
available in the smtp client now.
this also takes the "LIMITS RCPTMAX" smtp extension into account: if the server
only accepts a single recipient, we won't send multiple.
if a server doesn't announce a RCPTMAX limit, but still has one (like mox does
for non-spf-verified transactions), we'll recognize code 452 and 552 (for
historic reasons) as temporary error, and try again in a separate transaction
immediately after. we don't yet implement "LIMITS MAILMAX", doesn't seem likely
in practice.
2024-03-07 12:07:53 +03:00
now := time . Now ( )
2024-03-05 22:10:28 +03:00
for i , rcpt := range recipients {
queue: deliver to multiple recipients in a single smtp transaction
transferring the data only once. we only do this when the recipient domains
are the same. when queuing, we now take care to set the same NextAttempt
timestamp, so queued messages are actually eligable for combined delivery.
this adds a DeliverMultiple to the smtp client. for pipelined requests, it will
send all RCPT TO (and MAIL and DATA) in one go, and handles the various
responses and error conditions, returning either an overal error, or per
recipient smtp responses. the results of the smtp LIMITS extension are also
available in the smtp client now.
this also takes the "LIMITS RCPTMAX" smtp extension into account: if the server
only accepts a single recipient, we won't send multiple.
if a server doesn't announce a RCPTMAX limit, but still has one (like mox does
for non-spf-verified transactions), we'll recognize code 452 and 552 (for
historic reasons) as temporary error, and try again in a separate transaction
immediately after. we don't yet implement "LIMITS MAILMAX", doesn't seem likely
in practice.
2024-03-07 12:07:53 +03:00
// Don't use per-recipient unique message prefix when multiple recipients are
// present, or the queue cannot deliver it in a single smtp transaction.
var recvRcpt string
if len ( recipients ) == 1 {
recvRcpt = rcpt . Pack ( smtputf8 )
}
rcptMsgPrefix := recvHdrFor ( recvRcpt ) + msgPrefix
2023-11-09 23:15:27 +03:00
msgSize := int64 ( len ( rcptMsgPrefix ) ) + xc . Size
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
toPath := smtp . Path {
Localpart : rcpt . Localpart ,
IPDomain : dns . IPDomain { Domain : rcpt . Domain } ,
}
2024-03-16 22:54:10 +03:00
qm := queue . MakeMsg ( fromPath , toPath , xc . Has8bit , xc . SMTPUTF8 , msgSize , messageID , [ ] byte ( rcptMsgPrefix ) , m . RequireTLS , now )
2024-02-10 19:55:56 +03:00
if m . FutureRelease != nil {
ival := time . Until ( * m . FutureRelease )
if ival < 0 {
xcheckuserf ( ctx , errors . New ( "date/time is in the past" ) , "scheduling delivery" )
} else if ival > queue . FutureReleaseIntervalMax {
xcheckuserf ( ctx , fmt . Errorf ( "date/time can not be further than %v in the future" , queue . FutureReleaseIntervalMax ) , "scheduling delivery" )
}
qm . NextAttempt = * m . FutureRelease
qm . FutureReleaseRequest = "until;" + m . FutureRelease . Format ( time . RFC3339 )
// todo: possibly add a header to the message stored in the Sent mailbox to indicate it was scheduled for later delivery.
}
2024-03-05 22:10:28 +03:00
qml [ i ] = qm
}
2024-03-09 11:51:24 +03:00
err = queue . Add ( ctx , log , reqInfo . AccountName , dataFile , qml ... )
if err != nil {
2024-03-05 22:10:28 +03:00
metricSubmission . WithLabelValues ( "queueerror" ) . Inc ( )
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
}
2024-03-05 22:10:28 +03:00
xcheckf ( ctx , err , "adding messages to the delivery queue" )
metricSubmission . WithLabelValues ( "ok" ) . Inc ( )
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
var modseq store . ModSeq // Only set if needed.
// Append message to Sent mailbox and mark original messages as answered/forwarded.
acc . WithRLock ( func ( ) {
var changes [ ] store . Change
metricked := false
defer func ( ) {
if x := recover ( ) ; x != nil {
if ! metricked {
metricServerErrors . WithLabelValues ( "submit" ) . Inc ( )
}
panic ( x )
}
} ( )
xdbwrite ( ctx , acc , func ( tx * bstore . Tx ) {
if m . ResponseMessageID > 0 {
rm := xmessageID ( ctx , tx , m . ResponseMessageID )
oflags := rm . Flags
if m . IsForward {
rm . Forwarded = true
} else {
rm . Answered = true
}
if ! rm . Junk && ! rm . Notjunk {
rm . Notjunk = true
}
if rm . Flags != oflags {
modseq , err = acc . NextModSeq ( tx )
xcheckf ( ctx , err , "next modseq" )
rm . ModSeq = modseq
err := tx . Update ( & rm )
xcheckf ( ctx , err , "updating flags of replied/forwarded message" )
changes = append ( changes , rm . ChangeFlags ( oflags ) )
err = acc . RetrainMessages ( ctx , log , tx , [ ] store . Message { rm } , false )
xcheckf ( ctx , err , "retraining messages after reply/forward" )
}
}
2023-12-16 13:53:14 +03:00
sentmb , err := bstore . QueryTx [ store . Mailbox ] ( tx ) . FilterEqual ( "Sent" , true ) . Get ( )
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
if err == bstore . ErrAbsent {
// There is no mailbox designated as Sent mailbox, so we're done.
return
}
xcheckf ( ctx , err , "message submitted to queue, adding to Sent mailbox" )
if modseq == 0 {
modseq , err = acc . NextModSeq ( tx )
xcheckf ( ctx , err , "next modseq" )
}
sentm := store . Message {
CreateSeq : modseq ,
ModSeq : modseq ,
MailboxID : sentmb . ID ,
MailboxOrigID : sentmb . ID ,
Flags : store . Flags { Notjunk : true , Seen : true } ,
2023-11-09 23:15:27 +03:00
Size : int64 ( len ( msgPrefix ) ) + xc . Size ,
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
MsgPrefix : [ ] byte ( msgPrefix ) ,
}
2023-12-20 22:54:12 +03:00
if ok , maxSize , err := acc . CanAddMessageSize ( tx , sentm . Size ) ; err != nil {
xcheckf ( ctx , err , "checking quota" )
} else if ! ok {
xcheckuserf ( ctx , fmt . Errorf ( "account over maximum total message size %d" , maxSize ) , "checking quota" )
}
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
// Update mailbox before delivery, which changes uidnext.
sentmb . Add ( sentm . MailboxCounts ( ) )
err = tx . Update ( & sentmb )
xcheckf ( ctx , err , "updating sent mailbox for counts" )
2023-12-20 22:54:12 +03:00
err = acc . DeliverMessage ( log , tx , & sentm , dataFile , true , false , false , true )
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
if err != nil {
metricSubmission . WithLabelValues ( "storesenterror" ) . Inc ( )
metricked = true
}
xcheckf ( ctx , err , "message submitted to queue, appending message to Sent mailbox" )
changes = append ( changes , sentm . ChangeAddUID ( ) , sentmb . ChangeCounts ( ) )
} )
store . BroadcastChanges ( acc , changes )
} )
}
// MessageMove moves messages to another mailbox. If the message is already in
// the mailbox an error is returned.
func ( Webmail ) MessageMove ( ctx context . Context , messageIDs [ ] int64 , mailboxID int64 ) {
2023-12-05 15:35:58 +03:00
log := pkglog . WithContext ( ctx )
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
reqInfo := ctx . Value ( requestInfoCtxKey ) . ( requestInfo )
2023-12-05 15:35:58 +03:00
acc , err := store . OpenAccount ( log , reqInfo . AccountName )
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
xcheckf ( ctx , err , "open account" )
defer func ( ) {
err := acc . Close ( )
log . Check ( err , "closing account" )
} ( )
acc . WithRLock ( func ( ) {
retrain := make ( [ ] store . Message , 0 , len ( messageIDs ) )
removeChanges := map [ int64 ] store . ChangeRemoveUIDs { }
// n adds, 1 remove, 2 mailboxcounts, optimistic and at least for a single message.
changes := make ( [ ] store . Change , 0 , len ( messageIDs ) + 3 )
xdbwrite ( ctx , acc , func ( tx * bstore . Tx ) {
var mbSrc store . Mailbox
var modseq store . ModSeq
mbDst := xmailboxID ( ctx , tx , mailboxID )
if len ( messageIDs ) == 0 {
return
}
keywords := map [ string ] struct { } { }
for _ , mid := range messageIDs {
m := xmessageID ( ctx , tx , mid )
// We may have loaded this mailbox in the previous iteration of this loop.
if m . MailboxID != mbSrc . ID {
if mbSrc . ID != 0 {
err = tx . Update ( & mbSrc )
xcheckf ( ctx , err , "updating source mailbox counts" )
changes = append ( changes , mbSrc . ChangeCounts ( ) )
}
mbSrc = xmailboxID ( ctx , tx , m . MailboxID )
}
if mbSrc . ID == mailboxID {
// Client should filter out messages that are already in mailbox.
xcheckuserf ( ctx , errors . New ( "already in destination mailbox" ) , "moving message" )
}
if modseq == 0 {
modseq , err = acc . NextModSeq ( tx )
xcheckf ( ctx , err , "assigning next modseq" )
}
ch := removeChanges [ m . MailboxID ]
ch . UIDs = append ( ch . UIDs , m . UID )
ch . ModSeq = modseq
ch . MailboxID = m . MailboxID
removeChanges [ m . MailboxID ] = ch
// Copy of message record that we'll insert when UID is freed up.
om := m
om . PrepareExpunge ( )
om . ID = 0 // Assign new ID.
om . ModSeq = modseq
mbSrc . Sub ( m . MailboxCounts ( ) )
if mbDst . Trash {
m . Seen = true
}
conf , _ := acc . Conf ( )
m . MailboxID = mbDst . ID
2023-08-09 17:47:29 +03:00
if m . IsReject && m . MailboxDestinedID != 0 {
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
// Incorrectly delivered to Rejects mailbox. Adjust MailboxOrigID so this message
// is used for reputation calculation during future deliveries.
m . MailboxOrigID = m . MailboxDestinedID
2023-08-09 17:47:29 +03:00
m . IsReject = false
2023-09-22 16:43:25 +03:00
m . Seen = false
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
}
m . UID = mbDst . UIDNext
m . ModSeq = modseq
mbDst . UIDNext ++
2023-09-21 16:19:11 +03:00
m . JunkFlagsForMailbox ( mbDst , conf )
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
err = tx . Update ( & m )
xcheckf ( ctx , err , "updating moved message in database" )
// Now that UID is unused, we can insert the old record again.
err = tx . Insert ( & om )
xcheckf ( ctx , err , "inserting record for expunge after moving message" )
mbDst . Add ( m . MailboxCounts ( ) )
changes = append ( changes , m . ChangeAddUID ( ) )
retrain = append ( retrain , m )
for _ , kw := range m . Keywords {
keywords [ kw ] = struct { } { }
}
}
err = tx . Update ( & mbSrc )
xcheckf ( ctx , err , "updating source mailbox counts" )
changes = append ( changes , mbSrc . ChangeCounts ( ) , mbDst . ChangeCounts ( ) )
// Ensure destination mailbox has keywords of the moved messages.
var mbKwChanged bool
mbDst . Keywords , mbKwChanged = store . MergeKeywords ( mbDst . Keywords , maps . Keys ( keywords ) )
if mbKwChanged {
changes = append ( changes , mbDst . ChangeKeywords ( ) )
}
err = tx . Update ( & mbDst )
xcheckf ( ctx , err , "updating mailbox with uidnext" )
err = acc . RetrainMessages ( ctx , log , tx , retrain , false )
xcheckf ( ctx , err , "retraining messages after move" )
} )
// Ensure UIDs of the removed message are in increasing order. It is quite common
// for all messages to be from a single source mailbox, meaning this is just one
// change, for which we preallocated space.
for _ , ch := range removeChanges {
sort . Slice ( ch . UIDs , func ( i , j int ) bool {
return ch . UIDs [ i ] < ch . UIDs [ j ]
} )
changes = append ( changes , ch )
}
store . BroadcastChanges ( acc , changes )
} )
}
// MessageDelete permanently deletes messages, without moving them to the Trash mailbox.
func ( Webmail ) MessageDelete ( ctx context . Context , messageIDs [ ] int64 ) {
2023-12-05 15:35:58 +03:00
log := pkglog . WithContext ( ctx )
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
reqInfo := ctx . Value ( requestInfoCtxKey ) . ( requestInfo )
2023-12-05 15:35:58 +03:00
acc , err := store . OpenAccount ( log , reqInfo . AccountName )
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
xcheckf ( ctx , err , "open account" )
defer func ( ) {
err := acc . Close ( )
log . Check ( err , "closing account" )
} ( )
if len ( messageIDs ) == 0 {
return
}
acc . WithWLock ( func ( ) {
removeChanges := map [ int64 ] store . ChangeRemoveUIDs { }
changes := make ( [ ] store . Change , 0 , len ( messageIDs ) + 1 ) // n remove, 1 mailbox counts
xdbwrite ( ctx , acc , func ( tx * bstore . Tx ) {
var modseq store . ModSeq
var mb store . Mailbox
remove := make ( [ ] store . Message , 0 , len ( messageIDs ) )
2023-12-20 22:54:12 +03:00
var totalSize int64
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
for _ , mid := range messageIDs {
m := xmessageID ( ctx , tx , mid )
2023-12-20 22:54:12 +03:00
totalSize += m . Size
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
if m . MailboxID != mb . ID {
if mb . ID != 0 {
err := tx . Update ( & mb )
xcheckf ( ctx , err , "updating mailbox counts" )
changes = append ( changes , mb . ChangeCounts ( ) )
}
mb = xmailboxID ( ctx , tx , m . MailboxID )
}
qmr := bstore . QueryTx [ store . Recipient ] ( tx )
qmr . FilterEqual ( "MessageID" , m . ID )
_ , err = qmr . Delete ( )
xcheckf ( ctx , err , "removing message recipients" )
mb . Sub ( m . MailboxCounts ( ) )
if modseq == 0 {
modseq , err = acc . NextModSeq ( tx )
xcheckf ( ctx , err , "assigning next modseq" )
}
m . Expunged = true
m . ModSeq = modseq
err = tx . Update ( & m )
xcheckf ( ctx , err , "marking message as expunged" )
ch := removeChanges [ m . MailboxID ]
ch . UIDs = append ( ch . UIDs , m . UID )
ch . MailboxID = m . MailboxID
ch . ModSeq = modseq
removeChanges [ m . MailboxID ] = ch
remove = append ( remove , m )
}
if mb . ID != 0 {
err := tx . Update ( & mb )
xcheckf ( ctx , err , "updating count in mailbox" )
changes = append ( changes , mb . ChangeCounts ( ) )
}
2023-12-20 22:54:12 +03:00
err = acc . AddMessageSize ( log , tx , - totalSize )
xcheckf ( ctx , err , "updating disk usage" )
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
// Mark removed messages as not needing training, then retrain them, so if they
// were trained, they get untrained.
for i := range remove {
remove [ i ] . Junk = false
remove [ i ] . Notjunk = false
}
err = acc . RetrainMessages ( ctx , log , tx , remove , true )
xcheckf ( ctx , err , "untraining deleted messages" )
} )
for _ , ch := range removeChanges {
sort . Slice ( ch . UIDs , func ( i , j int ) bool {
return ch . UIDs [ i ] < ch . UIDs [ j ]
} )
changes = append ( changes , ch )
}
store . BroadcastChanges ( acc , changes )
} )
for _ , mID := range messageIDs {
p := acc . MessagePath ( mID )
err := os . Remove ( p )
log . Check ( err , "removing message file for expunge" )
}
}
// FlagsAdd adds flags, either system flags like \Seen or custom keywords. The
// flags should be lower-case, but will be converted and verified.
func ( Webmail ) FlagsAdd ( ctx context . Context , messageIDs [ ] int64 , flaglist [ ] string ) {
2023-12-05 15:35:58 +03:00
log := pkglog . WithContext ( ctx )
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
reqInfo := ctx . Value ( requestInfoCtxKey ) . ( requestInfo )
2023-12-05 15:35:58 +03:00
acc , err := store . OpenAccount ( log , reqInfo . AccountName )
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
xcheckf ( ctx , err , "open account" )
defer func ( ) {
err := acc . Close ( )
log . Check ( err , "closing account" )
} ( )
flags , keywords , err := store . ParseFlagsKeywords ( flaglist )
xcheckuserf ( ctx , err , "parsing flags" )
acc . WithRLock ( func ( ) {
var changes [ ] store . Change
xdbwrite ( ctx , acc , func ( tx * bstore . Tx ) {
var modseq store . ModSeq
var retrain [ ] store . Message
var mb , origmb store . Mailbox
for _ , mid := range messageIDs {
m := xmessageID ( ctx , tx , mid )
if mb . ID != m . MailboxID {
if mb . ID != 0 {
err := tx . Update ( & mb )
xcheckf ( ctx , err , "updating mailbox" )
if mb . MailboxCounts != origmb . MailboxCounts {
changes = append ( changes , mb . ChangeCounts ( ) )
}
if mb . KeywordsChanged ( origmb ) {
changes = append ( changes , mb . ChangeKeywords ( ) )
}
}
mb = xmailboxID ( ctx , tx , m . MailboxID )
origmb = mb
}
mb . Keywords , _ = store . MergeKeywords ( mb . Keywords , keywords )
mb . Sub ( m . MailboxCounts ( ) )
oflags := m . Flags
m . Flags = m . Flags . Set ( flags , flags )
var kwChanged bool
m . Keywords , kwChanged = store . MergeKeywords ( m . Keywords , keywords )
mb . Add ( m . MailboxCounts ( ) )
if m . Flags == oflags && ! kwChanged {
continue
}
if modseq == 0 {
modseq , err = acc . NextModSeq ( tx )
xcheckf ( ctx , err , "assigning next modseq" )
}
m . ModSeq = modseq
err = tx . Update ( & m )
xcheckf ( ctx , err , "updating message" )
changes = append ( changes , m . ChangeFlags ( oflags ) )
retrain = append ( retrain , m )
}
if mb . ID != 0 {
err := tx . Update ( & mb )
xcheckf ( ctx , err , "updating mailbox" )
if mb . MailboxCounts != origmb . MailboxCounts {
changes = append ( changes , mb . ChangeCounts ( ) )
}
if mb . KeywordsChanged ( origmb ) {
changes = append ( changes , mb . ChangeKeywords ( ) )
}
}
err = acc . RetrainMessages ( ctx , log , tx , retrain , false )
xcheckf ( ctx , err , "retraining messages" )
} )
store . BroadcastChanges ( acc , changes )
} )
}
// FlagsClear clears flags, either system flags like \Seen or custom keywords.
func ( Webmail ) FlagsClear ( ctx context . Context , messageIDs [ ] int64 , flaglist [ ] string ) {
2023-12-05 15:35:58 +03:00
log := pkglog . WithContext ( ctx )
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
reqInfo := ctx . Value ( requestInfoCtxKey ) . ( requestInfo )
2023-12-05 15:35:58 +03:00
acc , err := store . OpenAccount ( log , reqInfo . AccountName )
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
xcheckf ( ctx , err , "open account" )
defer func ( ) {
err := acc . Close ( )
log . Check ( err , "closing account" )
} ( )
flags , keywords , err := store . ParseFlagsKeywords ( flaglist )
xcheckuserf ( ctx , err , "parsing flags" )
acc . WithRLock ( func ( ) {
var retrain [ ] store . Message
var changes [ ] store . Change
xdbwrite ( ctx , acc , func ( tx * bstore . Tx ) {
var modseq store . ModSeq
var mb , origmb store . Mailbox
for _ , mid := range messageIDs {
m := xmessageID ( ctx , tx , mid )
if mb . ID != m . MailboxID {
if mb . ID != 0 {
err := tx . Update ( & mb )
xcheckf ( ctx , err , "updating counts for mailbox" )
if mb . MailboxCounts != origmb . MailboxCounts {
changes = append ( changes , mb . ChangeCounts ( ) )
}
// note: cannot remove keywords from mailbox by removing keywords from message.
}
mb = xmailboxID ( ctx , tx , m . MailboxID )
origmb = mb
}
oflags := m . Flags
mb . Sub ( m . MailboxCounts ( ) )
m . Flags = m . Flags . Set ( flags , store . Flags { } )
var changed bool
m . Keywords , changed = store . RemoveKeywords ( m . Keywords , keywords )
mb . Add ( m . MailboxCounts ( ) )
if m . Flags == oflags && ! changed {
continue
}
if modseq == 0 {
modseq , err = acc . NextModSeq ( tx )
xcheckf ( ctx , err , "assigning next modseq" )
}
m . ModSeq = modseq
err = tx . Update ( & m )
xcheckf ( ctx , err , "updating message" )
changes = append ( changes , m . ChangeFlags ( oflags ) )
retrain = append ( retrain , m )
}
if mb . ID != 0 {
err := tx . Update ( & mb )
xcheckf ( ctx , err , "updating keywords in mailbox" )
if mb . MailboxCounts != origmb . MailboxCounts {
changes = append ( changes , mb . ChangeCounts ( ) )
}
// note: cannot remove keywords from mailbox by removing keywords from message.
}
err = acc . RetrainMessages ( ctx , log , tx , retrain , false )
xcheckf ( ctx , err , "retraining messages" )
} )
store . BroadcastChanges ( acc , changes )
} )
}
// MailboxCreate creates a new mailbox.
func ( Webmail ) MailboxCreate ( ctx context . Context , name string ) {
2023-12-05 15:35:58 +03:00
log := pkglog . WithContext ( ctx )
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
reqInfo := ctx . Value ( requestInfoCtxKey ) . ( requestInfo )
2023-12-05 15:35:58 +03:00
acc , err := store . OpenAccount ( log , reqInfo . AccountName )
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
xcheckf ( ctx , err , "open account" )
defer func ( ) {
err := acc . Close ( )
log . Check ( err , "closing account" )
} ( )
name , _ , err = store . CheckMailboxName ( name , false )
xcheckuserf ( ctx , err , "checking mailbox name" )
acc . WithWLock ( func ( ) {
var changes [ ] store . Change
xdbwrite ( ctx , acc , func ( tx * bstore . Tx ) {
var exists bool
var err error
changes , _ , exists , err = acc . MailboxCreate ( tx , name )
if exists {
xcheckuserf ( ctx , errors . New ( "mailbox already exists" ) , "creating mailbox" )
}
xcheckf ( ctx , err , "creating mailbox" )
} )
store . BroadcastChanges ( acc , changes )
} )
}
// MailboxDelete deletes a mailbox and all its messages.
func ( Webmail ) MailboxDelete ( ctx context . Context , mailboxID int64 ) {
2023-12-05 15:35:58 +03:00
log := pkglog . WithContext ( ctx )
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
reqInfo := ctx . Value ( requestInfoCtxKey ) . ( requestInfo )
2023-12-05 15:35:58 +03:00
acc , err := store . OpenAccount ( log , reqInfo . AccountName )
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
xcheckf ( ctx , err , "open account" )
defer func ( ) {
err := acc . Close ( )
log . Check ( err , "closing account" )
} ( )
// Messages to remove after having broadcasted the removal of messages.
var removeMessageIDs [ ] int64
acc . WithWLock ( func ( ) {
var changes [ ] store . Change
xdbwrite ( ctx , acc , func ( tx * bstore . Tx ) {
mb := xmailboxID ( ctx , tx , mailboxID )
if mb . Name == "Inbox" {
// Inbox is special in IMAP and cannot be removed.
xcheckuserf ( ctx , errors . New ( "cannot remove special Inbox" ) , "checking mailbox" )
}
var hasChildren bool
var err error
changes , removeMessageIDs , hasChildren , err = acc . MailboxDelete ( ctx , log , tx , mb )
if hasChildren {
xcheckuserf ( ctx , errors . New ( "mailbox has children" ) , "deleting mailbox" )
}
xcheckf ( ctx , err , "deleting mailbox" )
} )
store . BroadcastChanges ( acc , changes )
} )
for _ , mID := range removeMessageIDs {
p := acc . MessagePath ( mID )
err := os . Remove ( p )
2023-12-05 15:35:58 +03:00
log . Check ( err , "removing message file for mailbox delete" , slog . String ( "path" , p ) )
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
}
}
// MailboxEmpty empties a mailbox, removing all messages from the mailbox, but not
// its child mailboxes.
func ( Webmail ) MailboxEmpty ( ctx context . Context , mailboxID int64 ) {
2023-12-05 15:35:58 +03:00
log := pkglog . WithContext ( ctx )
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
reqInfo := ctx . Value ( requestInfoCtxKey ) . ( requestInfo )
2023-12-05 15:35:58 +03:00
acc , err := store . OpenAccount ( log , reqInfo . AccountName )
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
xcheckf ( ctx , err , "open account" )
defer func ( ) {
err := acc . Close ( )
log . Check ( err , "closing account" )
} ( )
var expunged [ ] store . Message
acc . WithWLock ( func ( ) {
var changes [ ] store . Change
xdbwrite ( ctx , acc , func ( tx * bstore . Tx ) {
mb := xmailboxID ( ctx , tx , mailboxID )
modseq , err := acc . NextModSeq ( tx )
xcheckf ( ctx , err , "next modseq" )
// Mark messages as expunged.
qm := bstore . QueryTx [ store . Message ] ( tx )
qm . FilterNonzero ( store . Message { MailboxID : mb . ID } )
qm . FilterEqual ( "Expunged" , false )
qm . SortAsc ( "UID" )
qm . Gather ( & expunged )
_ , err = qm . UpdateNonzero ( store . Message { ModSeq : modseq , Expunged : true } )
xcheckf ( ctx , err , "deleting messages" )
// Remove Recipients.
anyIDs := make ( [ ] any , len ( expunged ) )
for i , m := range expunged {
anyIDs [ i ] = m . ID
}
qmr := bstore . QueryTx [ store . Recipient ] ( tx )
qmr . FilterEqual ( "MessageID" , anyIDs ... )
_ , err = qmr . Delete ( )
xcheckf ( ctx , err , "removing message recipients" )
// Adjust mailbox counts, gather UIDs for broadcasted change, prepare for untraining.
2023-12-20 22:54:12 +03:00
var totalSize int64
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
uids := make ( [ ] store . UID , len ( expunged ) )
for i , m := range expunged {
m . Expunged = false // Gather returns updated values.
mb . Sub ( m . MailboxCounts ( ) )
2023-12-20 22:54:12 +03:00
totalSize += m . Size
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
uids [ i ] = m . UID
expunged [ i ] . Junk = false
expunged [ i ] . Notjunk = false
}
err = tx . Update ( & mb )
xcheckf ( ctx , err , "updating mailbox for counts" )
2023-12-20 22:54:12 +03:00
err = acc . AddMessageSize ( log , tx , - totalSize )
xcheckf ( ctx , err , "updating disk usage" )
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
err = acc . RetrainMessages ( ctx , log , tx , expunged , true )
xcheckf ( ctx , err , "retraining expunged messages" )
chremove := store . ChangeRemoveUIDs { MailboxID : mb . ID , UIDs : uids , ModSeq : modseq }
changes = [ ] store . Change { chremove , mb . ChangeCounts ( ) }
} )
store . BroadcastChanges ( acc , changes )
} )
for _ , m := range expunged {
p := acc . MessagePath ( m . ID )
err := os . Remove ( p )
2023-12-05 15:35:58 +03:00
log . Check ( err , "removing message file after emptying mailbox" , slog . String ( "path" , p ) )
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
}
}
// MailboxRename renames a mailbox, possibly moving it to a new parent. The mailbox
// ID and its messages are unchanged.
func ( Webmail ) MailboxRename ( ctx context . Context , mailboxID int64 , newName string ) {
2023-12-05 15:35:58 +03:00
log := pkglog . WithContext ( ctx )
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
reqInfo := ctx . Value ( requestInfoCtxKey ) . ( requestInfo )
2023-12-05 15:35:58 +03:00
acc , err := store . OpenAccount ( log , reqInfo . AccountName )
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
xcheckf ( ctx , err , "open account" )
defer func ( ) {
err := acc . Close ( )
log . Check ( err , "closing account" )
} ( )
// Renaming Inbox is special for IMAP. For IMAP we have to implement it per the
// standard. We can just say no.
newName , _ , err = store . CheckMailboxName ( newName , false )
xcheckuserf ( ctx , err , "checking new mailbox name" )
acc . WithWLock ( func ( ) {
var changes [ ] store . Change
xdbwrite ( ctx , acc , func ( tx * bstore . Tx ) {
mbsrc := xmailboxID ( ctx , tx , mailboxID )
var err error
var isInbox , notExists , alreadyExists bool
changes , isInbox , notExists , alreadyExists , err = acc . MailboxRename ( tx , mbsrc , newName )
if isInbox || notExists || alreadyExists {
xcheckuserf ( ctx , err , "renaming mailbox" )
}
xcheckf ( ctx , err , "renaming mailbox" )
} )
store . BroadcastChanges ( acc , changes )
} )
}
// CompleteRecipient returns autocomplete matches for a recipient, returning the
// matches, most recently used first, and whether this is the full list and further
// requests for longer prefixes aren't necessary.
func ( Webmail ) CompleteRecipient ( ctx context . Context , search string ) ( [ ] string , bool ) {
2023-12-05 15:35:58 +03:00
log := pkglog . WithContext ( ctx )
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
reqInfo := ctx . Value ( requestInfoCtxKey ) . ( requestInfo )
2023-12-05 15:35:58 +03:00
acc , err := store . OpenAccount ( log , reqInfo . AccountName )
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
xcheckf ( ctx , err , "open account" )
defer func ( ) {
err := acc . Close ( )
log . Check ( err , "closing account" )
} ( )
search = strings . ToLower ( search )
var matches [ ] string
all := true
acc . WithRLock ( func ( ) {
xdbread ( ctx , acc , func ( tx * bstore . Tx ) {
type key struct {
2024-03-08 23:08:40 +03:00
localpart string
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
domain string
}
seen := map [ key ] bool { }
q := bstore . QueryTx [ store . Recipient ] ( tx )
q . SortDesc ( "Sent" )
err := q . ForEach ( func ( r store . Recipient ) error {
k := key { r . Localpart , r . Domain }
if seen [ k ] {
return nil
}
// todo: we should have the address including name available in the database for searching. Will result in better matching, and also for the name.
2024-03-08 23:08:40 +03:00
address := fmt . Sprintf ( "<%s@%s>" , r . Localpart , r . Domain )
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
if ! strings . Contains ( strings . ToLower ( address ) , search ) {
return nil
}
if len ( matches ) >= 20 {
all = false
return bstore . StopForEach
}
// Look in the message that was sent for a name along with the address.
m := store . Message { ID : r . MessageID }
err := tx . Get ( & m )
xcheckf ( ctx , err , "get sent message" )
if ! m . Expunged && m . ParsedBuf != nil {
var part message . Part
err := json . Unmarshal ( m . ParsedBuf , & part )
xcheckf ( ctx , err , "parsing part" )
dom , err := dns . ParseDomain ( r . Domain )
xcheckf ( ctx , err , "parsing domain of recipient" )
var found bool
2024-03-08 23:08:40 +03:00
lp := r . Localpart
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
checkAddrs := func ( l [ ] message . Address ) {
if found {
return
}
for _ , a := range l {
if a . Name != "" && a . User == lp && strings . EqualFold ( a . Host , dom . ASCII ) {
found = true
address = addressString ( a , false )
return
}
}
}
if part . Envelope != nil {
env := part . Envelope
checkAddrs ( env . To )
checkAddrs ( env . CC )
checkAddrs ( env . BCC )
}
}
matches = append ( matches , address )
seen [ k ] = true
return nil
} )
xcheckf ( ctx , err , "listing recipients" )
} )
} )
return matches , all
}
// addressString returns an address into a string as it could be used in a message header.
func addressString ( a message . Address , smtputf8 bool ) string {
host := a . Host
dom , err := dns . ParseDomain ( a . Host )
if err == nil {
if smtputf8 && dom . Unicode != "" {
host = dom . Unicode
} else {
host = dom . ASCII
}
}
s := "<" + a . User + "@" + host + ">"
if a . Name != "" {
// todo: properly encoded/escaped name
s = a . Name + " " + s
}
return s
}
// MailboxSetSpecialUse sets the special use flags of a mailbox.
func ( Webmail ) MailboxSetSpecialUse ( ctx context . Context , mb store . Mailbox ) {
2023-12-05 15:35:58 +03:00
log := pkglog . WithContext ( ctx )
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
reqInfo := ctx . Value ( requestInfoCtxKey ) . ( requestInfo )
2023-12-05 15:35:58 +03:00
acc , err := store . OpenAccount ( log , reqInfo . AccountName )
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
xcheckf ( ctx , err , "open account" )
defer func ( ) {
err := acc . Close ( )
log . Check ( err , "closing account" )
} ( )
acc . WithWLock ( func ( ) {
var changes [ ] store . Change
xdbwrite ( ctx , acc , func ( tx * bstore . Tx ) {
xmb := xmailboxID ( ctx , tx , mb . ID )
// We only allow a single mailbox for each flag (JMAP requirement). So for any flag
// we set, we clear it for the mailbox(es) that had it, if any.
clearPrevious := func ( clear bool , specialUse string ) {
if ! clear {
return
}
var ombl [ ] store . Mailbox
q := bstore . QueryTx [ store . Mailbox ] ( tx )
q . FilterNotEqual ( "ID" , mb . ID )
q . FilterEqual ( specialUse , true )
q . Gather ( & ombl )
_ , err := q . UpdateField ( specialUse , false )
xcheckf ( ctx , err , "updating previous special-use mailboxes" )
for _ , omb := range ombl {
changes = append ( changes , omb . ChangeSpecialUse ( ) )
}
}
clearPrevious ( mb . Archive , "Archive" )
clearPrevious ( mb . Draft , "Draft" )
clearPrevious ( mb . Junk , "Junk" )
clearPrevious ( mb . Sent , "Sent" )
clearPrevious ( mb . Trash , "Trash" )
xmb . SpecialUse = mb . SpecialUse
err = tx . Update ( & xmb )
xcheckf ( ctx , err , "updating special-use flags for mailbox" )
changes = append ( changes , xmb . ChangeSpecialUse ( ) )
} )
store . BroadcastChanges ( acc , changes )
} )
}
implement message threading in backend and webmail
we match messages to their parents based on the "references" and "in-reply-to"
headers (requiring the same base subject), and in absense of those headers we
also by only base subject (against messages received max 4 weeks ago).
we store a threadid with messages. all messages in a thread have the same
threadid. messages also have a "thread parent ids", which holds all id's of
parent messages up to the thread root. then there is "thread missing link",
which is set when a referenced immediate parent wasn't found (but possibly
earlier ancestors can still be found and will be in thread parent ids".
threads can be muted: newly delivered messages are automatically marked as
read/seen. threads can be marked as collapsed: if set, the webmail collapses
the thread to a single item in the basic threading view (default is to expand
threads). the muted and collapsed fields are copied from their parent on
message delivery.
the threading is implemented in the webmail. the non-threading mode still works
as before. the new default threading mode "unread" automatically expands only
the threads with at least one unread (not seen) meessage. the basic threading
mode "on" expands all threads except when explicitly collapsed (as saved in the
thread collapsed field). new shortcuts for navigation/interaction threads have
been added, e.g. go to previous/next thread root, toggle collapse/expand of
thread (or double click), toggle mute of thread. some previous shortcuts have
changed, see the help for details.
the message threading are added with an explicit account upgrade step,
automatically started when an account is opened. the upgrade is done in the
background because it will take too long for large mailboxes to block account
operations. the upgrade takes two steps: 1. updating all message records in the
database to add a normalized message-id and thread base subject (with "re:",
"fwd:" and several other schemes stripped). 2. going through all messages in
the database again, reading the "references" and "in-reply-to" headers from
disk, and matching against their parents. this second step is also done at the
end of each import of mbox/maildir mailboxes. new deliveries are matched
immediately against other existing messages, currently no attempt is made to
rematch previously delivered messages (which could be useful for related
messages being delivered out of order).
the threading is not yet exposed over imap.
2023-09-13 09:51:50 +03:00
// ThreadCollapse saves the ThreadCollapse field for the messages and its
// children. The messageIDs are typically thread roots. But not all roots
// (without parent) of a thread need to have the same collapsed state.
func ( Webmail ) ThreadCollapse ( ctx context . Context , messageIDs [ ] int64 , collapse bool ) {
2023-12-05 15:35:58 +03:00
log := pkglog . WithContext ( ctx )
implement message threading in backend and webmail
we match messages to their parents based on the "references" and "in-reply-to"
headers (requiring the same base subject), and in absense of those headers we
also by only base subject (against messages received max 4 weeks ago).
we store a threadid with messages. all messages in a thread have the same
threadid. messages also have a "thread parent ids", which holds all id's of
parent messages up to the thread root. then there is "thread missing link",
which is set when a referenced immediate parent wasn't found (but possibly
earlier ancestors can still be found and will be in thread parent ids".
threads can be muted: newly delivered messages are automatically marked as
read/seen. threads can be marked as collapsed: if set, the webmail collapses
the thread to a single item in the basic threading view (default is to expand
threads). the muted and collapsed fields are copied from their parent on
message delivery.
the threading is implemented in the webmail. the non-threading mode still works
as before. the new default threading mode "unread" automatically expands only
the threads with at least one unread (not seen) meessage. the basic threading
mode "on" expands all threads except when explicitly collapsed (as saved in the
thread collapsed field). new shortcuts for navigation/interaction threads have
been added, e.g. go to previous/next thread root, toggle collapse/expand of
thread (or double click), toggle mute of thread. some previous shortcuts have
changed, see the help for details.
the message threading are added with an explicit account upgrade step,
automatically started when an account is opened. the upgrade is done in the
background because it will take too long for large mailboxes to block account
operations. the upgrade takes two steps: 1. updating all message records in the
database to add a normalized message-id and thread base subject (with "re:",
"fwd:" and several other schemes stripped). 2. going through all messages in
the database again, reading the "references" and "in-reply-to" headers from
disk, and matching against their parents. this second step is also done at the
end of each import of mbox/maildir mailboxes. new deliveries are matched
immediately against other existing messages, currently no attempt is made to
rematch previously delivered messages (which could be useful for related
messages being delivered out of order).
the threading is not yet exposed over imap.
2023-09-13 09:51:50 +03:00
reqInfo := ctx . Value ( requestInfoCtxKey ) . ( requestInfo )
2023-12-05 15:35:58 +03:00
acc , err := store . OpenAccount ( log , reqInfo . AccountName )
implement message threading in backend and webmail
we match messages to their parents based on the "references" and "in-reply-to"
headers (requiring the same base subject), and in absense of those headers we
also by only base subject (against messages received max 4 weeks ago).
we store a threadid with messages. all messages in a thread have the same
threadid. messages also have a "thread parent ids", which holds all id's of
parent messages up to the thread root. then there is "thread missing link",
which is set when a referenced immediate parent wasn't found (but possibly
earlier ancestors can still be found and will be in thread parent ids".
threads can be muted: newly delivered messages are automatically marked as
read/seen. threads can be marked as collapsed: if set, the webmail collapses
the thread to a single item in the basic threading view (default is to expand
threads). the muted and collapsed fields are copied from their parent on
message delivery.
the threading is implemented in the webmail. the non-threading mode still works
as before. the new default threading mode "unread" automatically expands only
the threads with at least one unread (not seen) meessage. the basic threading
mode "on" expands all threads except when explicitly collapsed (as saved in the
thread collapsed field). new shortcuts for navigation/interaction threads have
been added, e.g. go to previous/next thread root, toggle collapse/expand of
thread (or double click), toggle mute of thread. some previous shortcuts have
changed, see the help for details.
the message threading are added with an explicit account upgrade step,
automatically started when an account is opened. the upgrade is done in the
background because it will take too long for large mailboxes to block account
operations. the upgrade takes two steps: 1. updating all message records in the
database to add a normalized message-id and thread base subject (with "re:",
"fwd:" and several other schemes stripped). 2. going through all messages in
the database again, reading the "references" and "in-reply-to" headers from
disk, and matching against their parents. this second step is also done at the
end of each import of mbox/maildir mailboxes. new deliveries are matched
immediately against other existing messages, currently no attempt is made to
rematch previously delivered messages (which could be useful for related
messages being delivered out of order).
the threading is not yet exposed over imap.
2023-09-13 09:51:50 +03:00
xcheckf ( ctx , err , "open account" )
defer func ( ) {
err := acc . Close ( )
log . Check ( err , "closing account" )
} ( )
if len ( messageIDs ) == 0 {
xcheckuserf ( ctx , errors . New ( "no messages" ) , "setting collapse" )
}
acc . WithWLock ( func ( ) {
changes := make ( [ ] store . Change , 0 , len ( messageIDs ) )
xdbwrite ( ctx , acc , func ( tx * bstore . Tx ) {
// Gather ThreadIDs to list all potential messages, for a way to get all potential
// (child) messages. Further refined in FilterFn.
threadIDs := map [ int64 ] struct { } { }
msgIDs := map [ int64 ] struct { } { }
for _ , id := range messageIDs {
m := store . Message { ID : id }
err := tx . Get ( & m )
if err == bstore . ErrAbsent {
xcheckuserf ( ctx , err , "get message" )
}
xcheckf ( ctx , err , "get message" )
threadIDs [ m . ThreadID ] = struct { } { }
msgIDs [ id ] = struct { } { }
}
var updated [ ] store . Message
q := bstore . QueryTx [ store . Message ] ( tx )
q . FilterEqual ( "ThreadID" , slicesAny ( maps . Keys ( threadIDs ) ) ... )
q . FilterNotEqual ( "ThreadCollapsed" , collapse )
q . FilterFn ( func ( tm store . Message ) bool {
for _ , id := range tm . ThreadParentIDs {
if _ , ok := msgIDs [ id ] ; ok {
return true
}
}
_ , ok := msgIDs [ tm . ID ]
return ok
} )
q . Gather ( & updated )
q . SortAsc ( "ID" ) // Consistent order for testing.
_ , err = q . UpdateFields ( map [ string ] any { "ThreadCollapsed" : collapse } )
xcheckf ( ctx , err , "updating collapse in database" )
for _ , m := range updated {
changes = append ( changes , m . ChangeThread ( ) )
}
} )
store . BroadcastChanges ( acc , changes )
} )
}
// ThreadMute saves the ThreadMute field for the messages and their children.
// If messages are muted, they are also marked collapsed.
func ( Webmail ) ThreadMute ( ctx context . Context , messageIDs [ ] int64 , mute bool ) {
2023-12-05 15:35:58 +03:00
log := pkglog . WithContext ( ctx )
implement message threading in backend and webmail
we match messages to their parents based on the "references" and "in-reply-to"
headers (requiring the same base subject), and in absense of those headers we
also by only base subject (against messages received max 4 weeks ago).
we store a threadid with messages. all messages in a thread have the same
threadid. messages also have a "thread parent ids", which holds all id's of
parent messages up to the thread root. then there is "thread missing link",
which is set when a referenced immediate parent wasn't found (but possibly
earlier ancestors can still be found and will be in thread parent ids".
threads can be muted: newly delivered messages are automatically marked as
read/seen. threads can be marked as collapsed: if set, the webmail collapses
the thread to a single item in the basic threading view (default is to expand
threads). the muted and collapsed fields are copied from their parent on
message delivery.
the threading is implemented in the webmail. the non-threading mode still works
as before. the new default threading mode "unread" automatically expands only
the threads with at least one unread (not seen) meessage. the basic threading
mode "on" expands all threads except when explicitly collapsed (as saved in the
thread collapsed field). new shortcuts for navigation/interaction threads have
been added, e.g. go to previous/next thread root, toggle collapse/expand of
thread (or double click), toggle mute of thread. some previous shortcuts have
changed, see the help for details.
the message threading are added with an explicit account upgrade step,
automatically started when an account is opened. the upgrade is done in the
background because it will take too long for large mailboxes to block account
operations. the upgrade takes two steps: 1. updating all message records in the
database to add a normalized message-id and thread base subject (with "re:",
"fwd:" and several other schemes stripped). 2. going through all messages in
the database again, reading the "references" and "in-reply-to" headers from
disk, and matching against their parents. this second step is also done at the
end of each import of mbox/maildir mailboxes. new deliveries are matched
immediately against other existing messages, currently no attempt is made to
rematch previously delivered messages (which could be useful for related
messages being delivered out of order).
the threading is not yet exposed over imap.
2023-09-13 09:51:50 +03:00
reqInfo := ctx . Value ( requestInfoCtxKey ) . ( requestInfo )
2023-12-05 15:35:58 +03:00
acc , err := store . OpenAccount ( log , reqInfo . AccountName )
implement message threading in backend and webmail
we match messages to their parents based on the "references" and "in-reply-to"
headers (requiring the same base subject), and in absense of those headers we
also by only base subject (against messages received max 4 weeks ago).
we store a threadid with messages. all messages in a thread have the same
threadid. messages also have a "thread parent ids", which holds all id's of
parent messages up to the thread root. then there is "thread missing link",
which is set when a referenced immediate parent wasn't found (but possibly
earlier ancestors can still be found and will be in thread parent ids".
threads can be muted: newly delivered messages are automatically marked as
read/seen. threads can be marked as collapsed: if set, the webmail collapses
the thread to a single item in the basic threading view (default is to expand
threads). the muted and collapsed fields are copied from their parent on
message delivery.
the threading is implemented in the webmail. the non-threading mode still works
as before. the new default threading mode "unread" automatically expands only
the threads with at least one unread (not seen) meessage. the basic threading
mode "on" expands all threads except when explicitly collapsed (as saved in the
thread collapsed field). new shortcuts for navigation/interaction threads have
been added, e.g. go to previous/next thread root, toggle collapse/expand of
thread (or double click), toggle mute of thread. some previous shortcuts have
changed, see the help for details.
the message threading are added with an explicit account upgrade step,
automatically started when an account is opened. the upgrade is done in the
background because it will take too long for large mailboxes to block account
operations. the upgrade takes two steps: 1. updating all message records in the
database to add a normalized message-id and thread base subject (with "re:",
"fwd:" and several other schemes stripped). 2. going through all messages in
the database again, reading the "references" and "in-reply-to" headers from
disk, and matching against their parents. this second step is also done at the
end of each import of mbox/maildir mailboxes. new deliveries are matched
immediately against other existing messages, currently no attempt is made to
rematch previously delivered messages (which could be useful for related
messages being delivered out of order).
the threading is not yet exposed over imap.
2023-09-13 09:51:50 +03:00
xcheckf ( ctx , err , "open account" )
defer func ( ) {
err := acc . Close ( )
log . Check ( err , "closing account" )
} ( )
if len ( messageIDs ) == 0 {
xcheckuserf ( ctx , errors . New ( "no messages" ) , "setting mute" )
}
acc . WithWLock ( func ( ) {
changes := make ( [ ] store . Change , 0 , len ( messageIDs ) )
xdbwrite ( ctx , acc , func ( tx * bstore . Tx ) {
threadIDs := map [ int64 ] struct { } { }
msgIDs := map [ int64 ] struct { } { }
for _ , id := range messageIDs {
m := store . Message { ID : id }
err := tx . Get ( & m )
if err == bstore . ErrAbsent {
xcheckuserf ( ctx , err , "get message" )
}
xcheckf ( ctx , err , "get message" )
threadIDs [ m . ThreadID ] = struct { } { }
msgIDs [ id ] = struct { } { }
}
var updated [ ] store . Message
q := bstore . QueryTx [ store . Message ] ( tx )
q . FilterEqual ( "ThreadID" , slicesAny ( maps . Keys ( threadIDs ) ) ... )
q . FilterFn ( func ( tm store . Message ) bool {
if tm . ThreadMuted == mute && ( ! mute || tm . ThreadCollapsed ) {
return false
}
for _ , id := range tm . ThreadParentIDs {
if _ , ok := msgIDs [ id ] ; ok {
return true
}
}
_ , ok := msgIDs [ tm . ID ]
return ok
} )
q . Gather ( & updated )
fields := map [ string ] any { "ThreadMuted" : mute }
if mute {
fields [ "ThreadCollapsed" ] = true
}
_ , err = q . UpdateFields ( fields )
xcheckf ( ctx , err , "updating mute in database" )
for _ , m := range updated {
changes = append ( changes , m . ChangeThread ( ) )
}
} )
store . BroadcastChanges ( acc , changes )
} )
}
2023-10-15 16:05:20 +03:00
// SecurityResult indicates whether a security feature is supported.
type SecurityResult string
const (
SecurityResultError SecurityResult = "error"
SecurityResultNo SecurityResult = "no"
SecurityResultYes SecurityResult = "yes"
// Unknown whether supported. Finding out may only be (reasonably) possible when
// trying (e.g. SMTP STARTTLS). Once tried, the result may be cached for future
// lookups.
SecurityResultUnknown SecurityResult = "unknown"
)
implement "requiretls", rfc 8689
with requiretls, the tls verification mode/rules for email deliveries can be
changed by the sender/submitter. in two ways:
1. "requiretls" smtp extension to always enforce verified tls (with mta-sts or
dnssec+dane), along the entire delivery path until delivery into the final
destination mailbox (so entire transport is verified-tls-protected).
2. "tls-required: no" message header, to ignore any tls and tls verification
errors even if the recipient domain has a policy that requires tls verification
(mta-sts and/or dnssec+dane), allowing delivery of non-sensitive messages in
case of misconfiguration/interoperability issues (at least useful for sending
tls reports).
we enable requiretls by default (only when tls is active), for smtp and
submission. it can be disabled through the config.
for each delivery attempt, we now store (per recipient domain, in the account
of the sender) whether the smtp server supports starttls and requiretls. this
support is shown (after having sent a first message) in the webmail when
sending a message (the previous 3 bars under the address input field are now 5
bars, the first for starttls support, the last for requiretls support). when
all recipient domains for a message are known to implement requiretls,
requiretls is automatically selected for sending (instead of "default" tls
behaviour). users can also select the "fallback to insecure" to add the
"tls-required: no" header.
new metrics are added for insight into requiretls errors and (some, not yet
all) cases where tls-required-no ignored a tls/verification error.
the admin can change the requiretls status for messages in the queue. so with
default delivery attempts, when verified tls is required by failing, an admin
could potentially change the field to "tls-required: no"-behaviour.
messages received (over smtp) with the requiretls option, get a comment added
to their Received header line, just before "id", after "with".
2023-10-24 11:06:16 +03:00
// RecipientSecurity is a quick analysis of the security properties of delivery to
// the recipient (domain).
2023-10-15 16:05:20 +03:00
type RecipientSecurity struct {
implement "requiretls", rfc 8689
with requiretls, the tls verification mode/rules for email deliveries can be
changed by the sender/submitter. in two ways:
1. "requiretls" smtp extension to always enforce verified tls (with mta-sts or
dnssec+dane), along the entire delivery path until delivery into the final
destination mailbox (so entire transport is verified-tls-protected).
2. "tls-required: no" message header, to ignore any tls and tls verification
errors even if the recipient domain has a policy that requires tls verification
(mta-sts and/or dnssec+dane), allowing delivery of non-sensitive messages in
case of misconfiguration/interoperability issues (at least useful for sending
tls reports).
we enable requiretls by default (only when tls is active), for smtp and
submission. it can be disabled through the config.
for each delivery attempt, we now store (per recipient domain, in the account
of the sender) whether the smtp server supports starttls and requiretls. this
support is shown (after having sent a first message) in the webmail when
sending a message (the previous 3 bars under the address input field are now 5
bars, the first for starttls support, the last for requiretls support). when
all recipient domains for a message are known to implement requiretls,
requiretls is automatically selected for sending (instead of "default" tls
behaviour). users can also select the "fallback to insecure" to add the
"tls-required: no" header.
new metrics are added for insight into requiretls errors and (some, not yet
all) cases where tls-required-no ignored a tls/verification error.
the admin can change the requiretls status for messages in the queue. so with
default delivery attempts, when verified tls is required by failing, an admin
could potentially change the field to "tls-required: no"-behaviour.
messages received (over smtp) with the requiretls option, get a comment added
to their Received header line, just before "id", after "with".
2023-10-24 11:06:16 +03:00
// Whether recipient domain supports (opportunistic) STARTTLS, as seen during most
// recent delivery attempt. Will be "unknown" if no delivery to the domain has been
// attempted yet.
STARTTLS SecurityResult
// Whether we have a stored enforced MTA-STS policy, or domain has MTA-STS DNS
// record.
MTASTS SecurityResult
// Whether MX lookup response was DNSSEC-signed.
DNSSEC SecurityResult
// Whether first delivery destination has DANE records.
DANE SecurityResult
// Whether recipient domain is known to implement the REQUIRETLS SMTP extension.
// Will be "unknown" if no delivery to the domain has been attempted yet.
RequireTLS SecurityResult
2023-10-15 16:05:20 +03:00
}
// RecipientSecurity looks up security properties of the address in the
// single-address message addressee (as it appears in a To/Cc/Bcc/etc header).
func ( Webmail ) RecipientSecurity ( ctx context . Context , messageAddressee string ) ( RecipientSecurity , error ) {
2023-12-05 15:35:58 +03:00
resolver := dns . StrictResolver { Pkg : "webmail" , Log : pkglog . WithContext ( ctx ) . Logger }
2023-10-15 16:05:20 +03:00
return recipientSecurity ( ctx , resolver , messageAddressee )
}
implement "requiretls", rfc 8689
with requiretls, the tls verification mode/rules for email deliveries can be
changed by the sender/submitter. in two ways:
1. "requiretls" smtp extension to always enforce verified tls (with mta-sts or
dnssec+dane), along the entire delivery path until delivery into the final
destination mailbox (so entire transport is verified-tls-protected).
2. "tls-required: no" message header, to ignore any tls and tls verification
errors even if the recipient domain has a policy that requires tls verification
(mta-sts and/or dnssec+dane), allowing delivery of non-sensitive messages in
case of misconfiguration/interoperability issues (at least useful for sending
tls reports).
we enable requiretls by default (only when tls is active), for smtp and
submission. it can be disabled through the config.
for each delivery attempt, we now store (per recipient domain, in the account
of the sender) whether the smtp server supports starttls and requiretls. this
support is shown (after having sent a first message) in the webmail when
sending a message (the previous 3 bars under the address input field are now 5
bars, the first for starttls support, the last for requiretls support). when
all recipient domains for a message are known to implement requiretls,
requiretls is automatically selected for sending (instead of "default" tls
behaviour). users can also select the "fallback to insecure" to add the
"tls-required: no" header.
new metrics are added for insight into requiretls errors and (some, not yet
all) cases where tls-required-no ignored a tls/verification error.
the admin can change the requiretls status for messages in the queue. so with
default delivery attempts, when verified tls is required by failing, an admin
could potentially change the field to "tls-required: no"-behaviour.
messages received (over smtp) with the requiretls option, get a comment added
to their Received header line, just before "id", after "with".
2023-10-24 11:06:16 +03:00
// logPanic can be called with a defer from a goroutine to prevent the entire program from being shutdown in case of a panic.
func logPanic ( ctx context . Context ) {
x := recover ( )
if x == nil {
return
}
2023-12-05 15:35:58 +03:00
log := pkglog . WithContext ( ctx )
log . Error ( "recover from panic" , slog . Any ( "panic" , x ) )
implement "requiretls", rfc 8689
with requiretls, the tls verification mode/rules for email deliveries can be
changed by the sender/submitter. in two ways:
1. "requiretls" smtp extension to always enforce verified tls (with mta-sts or
dnssec+dane), along the entire delivery path until delivery into the final
destination mailbox (so entire transport is verified-tls-protected).
2. "tls-required: no" message header, to ignore any tls and tls verification
errors even if the recipient domain has a policy that requires tls verification
(mta-sts and/or dnssec+dane), allowing delivery of non-sensitive messages in
case of misconfiguration/interoperability issues (at least useful for sending
tls reports).
we enable requiretls by default (only when tls is active), for smtp and
submission. it can be disabled through the config.
for each delivery attempt, we now store (per recipient domain, in the account
of the sender) whether the smtp server supports starttls and requiretls. this
support is shown (after having sent a first message) in the webmail when
sending a message (the previous 3 bars under the address input field are now 5
bars, the first for starttls support, the last for requiretls support). when
all recipient domains for a message are known to implement requiretls,
requiretls is automatically selected for sending (instead of "default" tls
behaviour). users can also select the "fallback to insecure" to add the
"tls-required: no" header.
new metrics are added for insight into requiretls errors and (some, not yet
all) cases where tls-required-no ignored a tls/verification error.
the admin can change the requiretls status for messages in the queue. so with
default delivery attempts, when verified tls is required by failing, an admin
could potentially change the field to "tls-required: no"-behaviour.
messages received (over smtp) with the requiretls option, get a comment added
to their Received header line, just before "id", after "with".
2023-10-24 11:06:16 +03:00
debug . PrintStack ( )
metrics . PanicInc ( metrics . Webmail )
}
2023-10-15 16:05:20 +03:00
// separate function for testing with mocked resolver.
func recipientSecurity ( ctx context . Context , resolver dns . Resolver , messageAddressee string ) ( RecipientSecurity , error ) {
2023-12-05 15:35:58 +03:00
log := pkglog . WithContext ( ctx )
2023-10-15 16:05:20 +03:00
rs := RecipientSecurity {
SecurityResultUnknown ,
SecurityResultUnknown ,
SecurityResultUnknown ,
implement "requiretls", rfc 8689
with requiretls, the tls verification mode/rules for email deliveries can be
changed by the sender/submitter. in two ways:
1. "requiretls" smtp extension to always enforce verified tls (with mta-sts or
dnssec+dane), along the entire delivery path until delivery into the final
destination mailbox (so entire transport is verified-tls-protected).
2. "tls-required: no" message header, to ignore any tls and tls verification
errors even if the recipient domain has a policy that requires tls verification
(mta-sts and/or dnssec+dane), allowing delivery of non-sensitive messages in
case of misconfiguration/interoperability issues (at least useful for sending
tls reports).
we enable requiretls by default (only when tls is active), for smtp and
submission. it can be disabled through the config.
for each delivery attempt, we now store (per recipient domain, in the account
of the sender) whether the smtp server supports starttls and requiretls. this
support is shown (after having sent a first message) in the webmail when
sending a message (the previous 3 bars under the address input field are now 5
bars, the first for starttls support, the last for requiretls support). when
all recipient domains for a message are known to implement requiretls,
requiretls is automatically selected for sending (instead of "default" tls
behaviour). users can also select the "fallback to insecure" to add the
"tls-required: no" header.
new metrics are added for insight into requiretls errors and (some, not yet
all) cases where tls-required-no ignored a tls/verification error.
the admin can change the requiretls status for messages in the queue. so with
default delivery attempts, when verified tls is required by failing, an admin
could potentially change the field to "tls-required: no"-behaviour.
messages received (over smtp) with the requiretls option, get a comment added
to their Received header line, just before "id", after "with".
2023-10-24 11:06:16 +03:00
SecurityResultUnknown ,
SecurityResultUnknown ,
2023-10-15 16:05:20 +03:00
}
msgAddr , err := mail . ParseAddress ( messageAddressee )
if err != nil {
return rs , fmt . Errorf ( "parsing message addressee: %v" , err )
}
addr , err := smtp . ParseAddress ( msgAddr . Address )
if err != nil {
return rs , fmt . Errorf ( "parsing address: %v" , err )
}
var wg sync . WaitGroup
// MTA-STS.
wg . Add ( 1 )
go func ( ) {
implement "requiretls", rfc 8689
with requiretls, the tls verification mode/rules for email deliveries can be
changed by the sender/submitter. in two ways:
1. "requiretls" smtp extension to always enforce verified tls (with mta-sts or
dnssec+dane), along the entire delivery path until delivery into the final
destination mailbox (so entire transport is verified-tls-protected).
2. "tls-required: no" message header, to ignore any tls and tls verification
errors even if the recipient domain has a policy that requires tls verification
(mta-sts and/or dnssec+dane), allowing delivery of non-sensitive messages in
case of misconfiguration/interoperability issues (at least useful for sending
tls reports).
we enable requiretls by default (only when tls is active), for smtp and
submission. it can be disabled through the config.
for each delivery attempt, we now store (per recipient domain, in the account
of the sender) whether the smtp server supports starttls and requiretls. this
support is shown (after having sent a first message) in the webmail when
sending a message (the previous 3 bars under the address input field are now 5
bars, the first for starttls support, the last for requiretls support). when
all recipient domains for a message are known to implement requiretls,
requiretls is automatically selected for sending (instead of "default" tls
behaviour). users can also select the "fallback to insecure" to add the
"tls-required: no" header.
new metrics are added for insight into requiretls errors and (some, not yet
all) cases where tls-required-no ignored a tls/verification error.
the admin can change the requiretls status for messages in the queue. so with
default delivery attempts, when verified tls is required by failing, an admin
could potentially change the field to "tls-required: no"-behaviour.
messages received (over smtp) with the requiretls option, get a comment added
to their Received header line, just before "id", after "with".
2023-10-24 11:06:16 +03:00
defer logPanic ( ctx )
2023-10-15 16:05:20 +03:00
defer wg . Done ( )
2023-12-05 15:35:58 +03:00
policy , _ , _ , err := mtastsdb . Get ( ctx , log . Logger , resolver , addr . Domain )
2023-10-15 16:05:20 +03:00
if policy != nil && policy . Mode == mtasts . ModeEnforce {
rs . MTASTS = SecurityResultYes
} else if err == nil {
rs . MTASTS = SecurityResultNo
} else {
rs . MTASTS = SecurityResultError
}
} ( )
// DNSSEC and DANE.
wg . Add ( 1 )
go func ( ) {
implement "requiretls", rfc 8689
with requiretls, the tls verification mode/rules for email deliveries can be
changed by the sender/submitter. in two ways:
1. "requiretls" smtp extension to always enforce verified tls (with mta-sts or
dnssec+dane), along the entire delivery path until delivery into the final
destination mailbox (so entire transport is verified-tls-protected).
2. "tls-required: no" message header, to ignore any tls and tls verification
errors even if the recipient domain has a policy that requires tls verification
(mta-sts and/or dnssec+dane), allowing delivery of non-sensitive messages in
case of misconfiguration/interoperability issues (at least useful for sending
tls reports).
we enable requiretls by default (only when tls is active), for smtp and
submission. it can be disabled through the config.
for each delivery attempt, we now store (per recipient domain, in the account
of the sender) whether the smtp server supports starttls and requiretls. this
support is shown (after having sent a first message) in the webmail when
sending a message (the previous 3 bars under the address input field are now 5
bars, the first for starttls support, the last for requiretls support). when
all recipient domains for a message are known to implement requiretls,
requiretls is automatically selected for sending (instead of "default" tls
behaviour). users can also select the "fallback to insecure" to add the
"tls-required: no" header.
new metrics are added for insight into requiretls errors and (some, not yet
all) cases where tls-required-no ignored a tls/verification error.
the admin can change the requiretls status for messages in the queue. so with
default delivery attempts, when verified tls is required by failing, an admin
could potentially change the field to "tls-required: no"-behaviour.
messages received (over smtp) with the requiretls option, get a comment added
to their Received header line, just before "id", after "with".
2023-10-24 11:06:16 +03:00
defer logPanic ( ctx )
2023-10-15 16:05:20 +03:00
defer wg . Done ( )
2023-12-05 15:35:58 +03:00
_ , origNextHopAuthentic , expandedNextHopAuthentic , _ , hosts , _ , err := smtpclient . GatherDestinations ( ctx , log . Logger , resolver , dns . IPDomain { Domain : addr . Domain } )
2023-10-15 16:05:20 +03:00
if err != nil {
rs . DNSSEC = SecurityResultError
return
}
if origNextHopAuthentic && expandedNextHopAuthentic {
rs . DNSSEC = SecurityResultYes
} else {
rs . DNSSEC = SecurityResultNo
}
if ! origNextHopAuthentic {
rs . DANE = SecurityResultNo
return
}
// We're only looking at the first host to deliver to (typically first mx destination).
if len ( hosts ) == 0 || hosts [ 0 ] . Domain . IsZero ( ) {
return // Should not happen.
}
host := hosts [ 0 ]
// Resolve the IPs. Required for DANE to prevent bad DNS servers from causing an
// error result instead of no-DANE result.
2023-12-05 15:35:58 +03:00
authentic , expandedAuthentic , expandedHost , _ , _ , err := smtpclient . GatherIPs ( ctx , log . Logger , resolver , host , map [ string ] [ ] net . IP { } )
2023-10-15 16:05:20 +03:00
if err != nil {
rs . DANE = SecurityResultError
return
}
if ! authentic {
rs . DANE = SecurityResultNo
return
}
2023-12-05 15:35:58 +03:00
daneRequired , _ , _ , err := smtpclient . GatherTLSA ( ctx , log . Logger , resolver , host . Domain , expandedAuthentic , expandedHost )
2023-10-15 16:05:20 +03:00
if err != nil {
rs . DANE = SecurityResultError
return
} else if daneRequired {
rs . DANE = SecurityResultYes
} else {
rs . DANE = SecurityResultNo
}
} ( )
implement "requiretls", rfc 8689
with requiretls, the tls verification mode/rules for email deliveries can be
changed by the sender/submitter. in two ways:
1. "requiretls" smtp extension to always enforce verified tls (with mta-sts or
dnssec+dane), along the entire delivery path until delivery into the final
destination mailbox (so entire transport is verified-tls-protected).
2. "tls-required: no" message header, to ignore any tls and tls verification
errors even if the recipient domain has a policy that requires tls verification
(mta-sts and/or dnssec+dane), allowing delivery of non-sensitive messages in
case of misconfiguration/interoperability issues (at least useful for sending
tls reports).
we enable requiretls by default (only when tls is active), for smtp and
submission. it can be disabled through the config.
for each delivery attempt, we now store (per recipient domain, in the account
of the sender) whether the smtp server supports starttls and requiretls. this
support is shown (after having sent a first message) in the webmail when
sending a message (the previous 3 bars under the address input field are now 5
bars, the first for starttls support, the last for requiretls support). when
all recipient domains for a message are known to implement requiretls,
requiretls is automatically selected for sending (instead of "default" tls
behaviour). users can also select the "fallback to insecure" to add the
"tls-required: no" header.
new metrics are added for insight into requiretls errors and (some, not yet
all) cases where tls-required-no ignored a tls/verification error.
the admin can change the requiretls status for messages in the queue. so with
default delivery attempts, when verified tls is required by failing, an admin
could potentially change the field to "tls-required: no"-behaviour.
messages received (over smtp) with the requiretls option, get a comment added
to their Received header line, just before "id", after "with".
2023-10-24 11:06:16 +03:00
// STARTTLS and RequireTLS
reqInfo := ctx . Value ( requestInfoCtxKey ) . ( requestInfo )
2023-12-05 15:35:58 +03:00
acc , err := store . OpenAccount ( log , reqInfo . AccountName )
implement "requiretls", rfc 8689
with requiretls, the tls verification mode/rules for email deliveries can be
changed by the sender/submitter. in two ways:
1. "requiretls" smtp extension to always enforce verified tls (with mta-sts or
dnssec+dane), along the entire delivery path until delivery into the final
destination mailbox (so entire transport is verified-tls-protected).
2. "tls-required: no" message header, to ignore any tls and tls verification
errors even if the recipient domain has a policy that requires tls verification
(mta-sts and/or dnssec+dane), allowing delivery of non-sensitive messages in
case of misconfiguration/interoperability issues (at least useful for sending
tls reports).
we enable requiretls by default (only when tls is active), for smtp and
submission. it can be disabled through the config.
for each delivery attempt, we now store (per recipient domain, in the account
of the sender) whether the smtp server supports starttls and requiretls. this
support is shown (after having sent a first message) in the webmail when
sending a message (the previous 3 bars under the address input field are now 5
bars, the first for starttls support, the last for requiretls support). when
all recipient domains for a message are known to implement requiretls,
requiretls is automatically selected for sending (instead of "default" tls
behaviour). users can also select the "fallback to insecure" to add the
"tls-required: no" header.
new metrics are added for insight into requiretls errors and (some, not yet
all) cases where tls-required-no ignored a tls/verification error.
the admin can change the requiretls status for messages in the queue. so with
default delivery attempts, when verified tls is required by failing, an admin
could potentially change the field to "tls-required: no"-behaviour.
messages received (over smtp) with the requiretls option, get a comment added
to their Received header line, just before "id", after "with".
2023-10-24 11:06:16 +03:00
xcheckf ( ctx , err , "open account" )
defer func ( ) {
if acc != nil {
err := acc . Close ( )
log . Check ( err , "closing account" )
}
} ( )
err = acc . DB . Read ( ctx , func ( tx * bstore . Tx ) error {
q := bstore . QueryTx [ store . RecipientDomainTLS ] ( tx )
q . FilterNonzero ( store . RecipientDomainTLS { Domain : addr . Domain . Name ( ) } )
rd , err := q . Get ( )
if err == bstore . ErrAbsent {
return nil
} else if err != nil {
rs . STARTTLS = SecurityResultError
rs . RequireTLS = SecurityResultError
2023-12-05 15:35:58 +03:00
log . Errorx ( "looking up recipient domain" , err , slog . Any ( "domain" , addr . Domain ) )
implement "requiretls", rfc 8689
with requiretls, the tls verification mode/rules for email deliveries can be
changed by the sender/submitter. in two ways:
1. "requiretls" smtp extension to always enforce verified tls (with mta-sts or
dnssec+dane), along the entire delivery path until delivery into the final
destination mailbox (so entire transport is verified-tls-protected).
2. "tls-required: no" message header, to ignore any tls and tls verification
errors even if the recipient domain has a policy that requires tls verification
(mta-sts and/or dnssec+dane), allowing delivery of non-sensitive messages in
case of misconfiguration/interoperability issues (at least useful for sending
tls reports).
we enable requiretls by default (only when tls is active), for smtp and
submission. it can be disabled through the config.
for each delivery attempt, we now store (per recipient domain, in the account
of the sender) whether the smtp server supports starttls and requiretls. this
support is shown (after having sent a first message) in the webmail when
sending a message (the previous 3 bars under the address input field are now 5
bars, the first for starttls support, the last for requiretls support). when
all recipient domains for a message are known to implement requiretls,
requiretls is automatically selected for sending (instead of "default" tls
behaviour). users can also select the "fallback to insecure" to add the
"tls-required: no" header.
new metrics are added for insight into requiretls errors and (some, not yet
all) cases where tls-required-no ignored a tls/verification error.
the admin can change the requiretls status for messages in the queue. so with
default delivery attempts, when verified tls is required by failing, an admin
could potentially change the field to "tls-required: no"-behaviour.
messages received (over smtp) with the requiretls option, get a comment added
to their Received header line, just before "id", after "with".
2023-10-24 11:06:16 +03:00
return nil
}
if rd . STARTTLS {
rs . STARTTLS = SecurityResultYes
} else {
rs . STARTTLS = SecurityResultNo
}
if rd . RequireTLS {
rs . RequireTLS = SecurityResultYes
} else {
rs . RequireTLS = SecurityResultNo
}
return nil
} )
xcheckf ( ctx , err , "lookup recipient domain" )
// Close account as soon as possible, not after waiting for MTA-STS/DNSSEC/DANE
// checks to complete, which can take a while.
err = acc . Close ( )
log . Check ( err , "closing account" )
acc = nil
2023-10-15 16:05:20 +03:00
wg . Wait ( )
implement "requiretls", rfc 8689
with requiretls, the tls verification mode/rules for email deliveries can be
changed by the sender/submitter. in two ways:
1. "requiretls" smtp extension to always enforce verified tls (with mta-sts or
dnssec+dane), along the entire delivery path until delivery into the final
destination mailbox (so entire transport is verified-tls-protected).
2. "tls-required: no" message header, to ignore any tls and tls verification
errors even if the recipient domain has a policy that requires tls verification
(mta-sts and/or dnssec+dane), allowing delivery of non-sensitive messages in
case of misconfiguration/interoperability issues (at least useful for sending
tls reports).
we enable requiretls by default (only when tls is active), for smtp and
submission. it can be disabled through the config.
for each delivery attempt, we now store (per recipient domain, in the account
of the sender) whether the smtp server supports starttls and requiretls. this
support is shown (after having sent a first message) in the webmail when
sending a message (the previous 3 bars under the address input field are now 5
bars, the first for starttls support, the last for requiretls support). when
all recipient domains for a message are known to implement requiretls,
requiretls is automatically selected for sending (instead of "default" tls
behaviour). users can also select the "fallback to insecure" to add the
"tls-required: no" header.
new metrics are added for insight into requiretls errors and (some, not yet
all) cases where tls-required-no ignored a tls/verification error.
the admin can change the requiretls status for messages in the queue. so with
default delivery attempts, when verified tls is required by failing, an admin
could potentially change the field to "tls-required: no"-behaviour.
messages received (over smtp) with the requiretls option, get a comment added
to their Received header line, just before "id", after "with".
2023-10-24 11:06:16 +03:00
2023-10-15 16:05:20 +03:00
return rs , nil
}
2024-02-09 13:21:33 +03:00
// DecodeMIMEWords decodes Q/B-encoded words for a mime headers into UTF-8 text.
func ( Webmail ) DecodeMIMEWords ( ctx context . Context , text string ) string {
s , err := wordDecoder . DecodeHeader ( text )
xcheckuserf ( ctx , err , "decoding mime q/b-word encoded header" )
return s
}
implement message threading in backend and webmail
we match messages to their parents based on the "references" and "in-reply-to"
headers (requiring the same base subject), and in absense of those headers we
also by only base subject (against messages received max 4 weeks ago).
we store a threadid with messages. all messages in a thread have the same
threadid. messages also have a "thread parent ids", which holds all id's of
parent messages up to the thread root. then there is "thread missing link",
which is set when a referenced immediate parent wasn't found (but possibly
earlier ancestors can still be found and will be in thread parent ids".
threads can be muted: newly delivered messages are automatically marked as
read/seen. threads can be marked as collapsed: if set, the webmail collapses
the thread to a single item in the basic threading view (default is to expand
threads). the muted and collapsed fields are copied from their parent on
message delivery.
the threading is implemented in the webmail. the non-threading mode still works
as before. the new default threading mode "unread" automatically expands only
the threads with at least one unread (not seen) meessage. the basic threading
mode "on" expands all threads except when explicitly collapsed (as saved in the
thread collapsed field). new shortcuts for navigation/interaction threads have
been added, e.g. go to previous/next thread root, toggle collapse/expand of
thread (or double click), toggle mute of thread. some previous shortcuts have
changed, see the help for details.
the message threading are added with an explicit account upgrade step,
automatically started when an account is opened. the upgrade is done in the
background because it will take too long for large mailboxes to block account
operations. the upgrade takes two steps: 1. updating all message records in the
database to add a normalized message-id and thread base subject (with "re:",
"fwd:" and several other schemes stripped). 2. going through all messages in
the database again, reading the "references" and "in-reply-to" headers from
disk, and matching against their parents. this second step is also done at the
end of each import of mbox/maildir mailboxes. new deliveries are matched
immediately against other existing messages, currently no attempt is made to
rematch previously delivered messages (which could be useful for related
messages being delivered out of order).
the threading is not yet exposed over imap.
2023-09-13 09:51:50 +03:00
func slicesAny [ T any ] ( l [ ] T ) [ ] any {
r := make ( [ ] any , len ( l ) )
for i , v := range l {
r [ i ] = v
}
return r
}
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
// SSETypes exists to ensure the generated API contains the types, for use in SSE events.
implement message threading in backend and webmail
we match messages to their parents based on the "references" and "in-reply-to"
headers (requiring the same base subject), and in absense of those headers we
also by only base subject (against messages received max 4 weeks ago).
we store a threadid with messages. all messages in a thread have the same
threadid. messages also have a "thread parent ids", which holds all id's of
parent messages up to the thread root. then there is "thread missing link",
which is set when a referenced immediate parent wasn't found (but possibly
earlier ancestors can still be found and will be in thread parent ids".
threads can be muted: newly delivered messages are automatically marked as
read/seen. threads can be marked as collapsed: if set, the webmail collapses
the thread to a single item in the basic threading view (default is to expand
threads). the muted and collapsed fields are copied from their parent on
message delivery.
the threading is implemented in the webmail. the non-threading mode still works
as before. the new default threading mode "unread" automatically expands only
the threads with at least one unread (not seen) meessage. the basic threading
mode "on" expands all threads except when explicitly collapsed (as saved in the
thread collapsed field). new shortcuts for navigation/interaction threads have
been added, e.g. go to previous/next thread root, toggle collapse/expand of
thread (or double click), toggle mute of thread. some previous shortcuts have
changed, see the help for details.
the message threading are added with an explicit account upgrade step,
automatically started when an account is opened. the upgrade is done in the
background because it will take too long for large mailboxes to block account
operations. the upgrade takes two steps: 1. updating all message records in the
database to add a normalized message-id and thread base subject (with "re:",
"fwd:" and several other schemes stripped). 2. going through all messages in
the database again, reading the "references" and "in-reply-to" headers from
disk, and matching against their parents. this second step is also done at the
end of each import of mbox/maildir mailboxes. new deliveries are matched
immediately against other existing messages, currently no attempt is made to
rematch previously delivered messages (which could be useful for related
messages being delivered out of order).
the threading is not yet exposed over imap.
2023-09-13 09:51:50 +03:00
func ( Webmail ) SSETypes ( ) ( start EventStart , viewErr EventViewErr , viewReset EventViewReset , viewMsgs EventViewMsgs , viewChanges EventViewChanges , msgAdd ChangeMsgAdd , msgRemove ChangeMsgRemove , msgFlags ChangeMsgFlags , msgThread ChangeMsgThread , mailboxRemove ChangeMailboxRemove , mailboxAdd ChangeMailboxAdd , mailboxRename ChangeMailboxRename , mailboxCounts ChangeMailboxCounts , mailboxSpecialUse ChangeMailboxSpecialUse , mailboxKeywords ChangeMailboxKeywords , flags store . Flags ) {
add webmail
it was far down on the roadmap, but implemented earlier, because it's
interesting, and to help prepare for a jmap implementation. for jmap we need to
implement more client-like functionality than with just imap. internal data
structures need to change. jmap has lots of other requirements, so it's already
a big project. by implementing a webmail now, some of the required data
structure changes become clear and can be made now, so the later jmap
implementation can do things similarly to the webmail code. the webmail
frontend and webmail are written together, making their interface/api much
smaller and simpler than jmap.
one of the internal changes is that we now keep track of per-mailbox
total/unread/unseen/deleted message counts and mailbox sizes. keeping this
data consistent after any change to the stored messages (through the code base)
is tricky, so mox now has a consistency check that verifies the counts are
correct, which runs only during tests, each time an internal account reference
is closed. we have a few more internal "changes" that are propagated for the
webmail frontend (that imap doesn't have a way to propagate on a connection),
like changes to the special-use flags on mailboxes, and used keywords in a
mailbox. more changes that will be required have revealed themselves while
implementing the webmail, and will be implemented next.
the webmail user interface is modeled after the mail clients i use or have
used: thunderbird, macos mail, mutt; and webmails i normally only use for
testing: gmail, proton, yahoo, outlook. a somewhat technical user is assumed,
but still the goal is to make this webmail client easy to use for everyone. the
user interface looks like most other mail clients: a list of mailboxes, a
search bar, a message list view, and message details. there is a top/bottom and
a left/right layout for the list/message view, default is automatic based on
screen size. the panes can be resized by the user. buttons for actions are just
text, not icons. clicking a button briefly shows the shortcut for the action in
the bottom right, helping with learning to operate quickly. any text that is
underdotted has a title attribute that causes more information to be displayed,
e.g. what a button does or a field is about. to highlight potential phishing
attempts, any text (anywhere in the webclient) that switches unicode "blocks"
(a rough approximation to (language) scripts) within a word is underlined
orange. multiple messages can be selected with familiar ui interaction:
clicking while holding control and/or shift keys. keyboard navigation works
with arrows/page up/down and home/end keys, and also with a few basic vi-like
keys for list/message navigation. we prefer showing the text instead of
html (with inlined images only) version of a message. html messages are shown
in an iframe served from an endpoint with CSP headers to prevent dangerous
resources (scripts, external images) from being loaded. the html is also
sanitized, with javascript removed. a user can choose to load external
resources (e.g. images for tracking purposes).
the frontend is just (strict) typescript, no external frameworks. all
incoming/outgoing data is typechecked, both the api request parameters and
response types, and the data coming in over SSE. the types and checking code
are generated with sherpats, which uses the api definitions generated by
sherpadoc based on the Go code. so types from the backend are automatically
propagated to the frontend. since there is no framework to automatically
propagate properties and rerender components, changes coming in over the SSE
connection are propagated explicitly with regular function calls. the ui is
separated into "views", each with a "root" dom element that is added to the
visible document. these views have additional functions for getting changes
propagated, often resulting in the view updating its (internal) ui state (dom).
we keep the frontend compilation simple, it's just a few typescript files that
get compiled (combined and types stripped) into a single js file, no additional
runtime code needed or complicated build processes used. the webmail is served
is served from a compressed, cachable html file that includes style and the
javascript, currently just over 225kb uncompressed, under 60kb compressed (not
minified, including comments). we include the generated js files in the
repository, to keep Go's easily buildable self-contained binaries.
authentication is basic http, as with the account and admin pages. most data
comes in over one long-term SSE connection to the backend. api requests signal
which mailbox/search/messages are requested over the SSE connection. fetching
individual messages, and making changes, are done through api calls. the
operations are similar to imap, so some code has been moved from package
imapserver to package store. the future jmap implementation will benefit from
these changes too. more functionality will probably be moved to the store
package in the future.
the quickstart enables webmail on the internal listener by default (for new
installs). users can enable it on the public listener if they want to. mox
localserve enables it too. to enable webmail on existing installs, add settings
like the following to the listeners in mox.conf, similar to AccountHTTP(S):
WebmailHTTP:
Enabled: true
WebmailHTTPS:
Enabled: true
special thanks to liesbeth, gerben, andrii for early user feedback.
there is plenty still to do, see the list at the top of webmail/webmail.ts.
feedback welcome as always.
2023-08-07 22:57:03 +03:00
return
}