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
package store
import (
"os"
make mox compile on windows, without "mox serve" but with working "mox localserve"
getting mox to compile required changing code in only a few places where
package "syscall" was used: for accessing file access times and for umask
handling. an open problem is how to start a process as an unprivileged user on
windows. that's why "mox serve" isn't implemented yet. and just finding a way
to implement it now may not be good enough in the near future: we may want to
starting using a more complete privilege separation approach, with a process
handling sensitive tasks (handling private keys, authentication), where we may
want to pass file descriptors between processes. how would that work on
windows?
anyway, getting mox to compile for windows doesn't mean it works properly on
windows. the largest issue: mox would normally open a file, rename or remove
it, and finally close it. this happens during message delivery. that doesn't
work on windows, the rename/remove would fail because the file is still open.
so this commit swaps many "remove" and "close" calls. renames are a longer
story: message delivery had two ways to deliver: with "consuming" the
(temporary) message file (which would rename it to its final destination), and
without consuming (by hardlinking the file, falling back to copying). the last
delivery to a recipient of a message (and the only one in the common case of a
single recipient) would consume the message, and the earlier recipients would
not. during delivery, the already open message file was used, to parse the
message. we still want to use that open message file, and the caller now stays
responsible for closing it, but we no longer try to rename (consume) the file.
we always hardlink (or copy) during delivery (this works on windows), and the
caller is responsible for closing and removing (in that order) the original
temporary file. this does cost one syscall more. but it makes the delivery code
(responsibilities) a bit simpler.
there is one more obvious issue: the file system path separator. mox already
used the "filepath" package to join paths in many places, but not everywhere.
and it still used strings with slashes for local file access. with this commit,
the code now uses filepath.FromSlash for path strings with slashes, uses
"filepath" in a few more places where it previously didn't. also switches from
"filepath" to regular "path" package when handling mailbox names in a few
places, because those always use forward slashes, regardless of local file
system conventions. windows can handle forward slashes when opening files, so
test code that passes path strings with forward slashes straight to go stdlib
file i/o functions are left unchanged to reduce code churn. the regular
non-test code, or test code that uses path strings in places other than
standard i/o functions, does have the paths converted for consistent paths
(otherwise we would end up with paths with mixed forward/backward slashes in
log messages).
windows cannot dup a listening socket. for "mox localserve", it isn't
important, and we can work around the issue. the current approach for "mox
serve" (forking a process and passing file descriptors of listening sockets on
"privileged" ports) won't work on windows. perhaps it isn't needed on windows,
and any user can listen on "privileged" ports? that would be welcome.
on windows, os.Open cannot open a directory, so we cannot call Sync on it after
message delivery. a cursory internet search indicates that directories cannot
be synced on windows. the story is probably much more nuanced than that, with
long deep technical details/discussions/disagreement/confusion, like on unix.
for "mox localserve" we can get away with making syncdir a no-op.
2023-10-14 11:54:07 +03:00
"path/filepath"
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
"reflect"
"strings"
"testing"
"time"
"github.com/mjl-/bstore"
"github.com/mjl-/mox/mlog"
"github.com/mjl-/mox/mox-"
)
func TestThreadingUpgrade ( t * testing . T ) {
2023-12-05 15:35:58 +03:00
log := mlog . New ( "store" , nil )
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
os . RemoveAll ( "../testdata/store/data" )
make mox compile on windows, without "mox serve" but with working "mox localserve"
getting mox to compile required changing code in only a few places where
package "syscall" was used: for accessing file access times and for umask
handling. an open problem is how to start a process as an unprivileged user on
windows. that's why "mox serve" isn't implemented yet. and just finding a way
to implement it now may not be good enough in the near future: we may want to
starting using a more complete privilege separation approach, with a process
handling sensitive tasks (handling private keys, authentication), where we may
want to pass file descriptors between processes. how would that work on
windows?
anyway, getting mox to compile for windows doesn't mean it works properly on
windows. the largest issue: mox would normally open a file, rename or remove
it, and finally close it. this happens during message delivery. that doesn't
work on windows, the rename/remove would fail because the file is still open.
so this commit swaps many "remove" and "close" calls. renames are a longer
story: message delivery had two ways to deliver: with "consuming" the
(temporary) message file (which would rename it to its final destination), and
without consuming (by hardlinking the file, falling back to copying). the last
delivery to a recipient of a message (and the only one in the common case of a
single recipient) would consume the message, and the earlier recipients would
not. during delivery, the already open message file was used, to parse the
message. we still want to use that open message file, and the caller now stays
responsible for closing it, but we no longer try to rename (consume) the file.
we always hardlink (or copy) during delivery (this works on windows), and the
caller is responsible for closing and removing (in that order) the original
temporary file. this does cost one syscall more. but it makes the delivery code
(responsibilities) a bit simpler.
there is one more obvious issue: the file system path separator. mox already
used the "filepath" package to join paths in many places, but not everywhere.
and it still used strings with slashes for local file access. with this commit,
the code now uses filepath.FromSlash for path strings with slashes, uses
"filepath" in a few more places where it previously didn't. also switches from
"filepath" to regular "path" package when handling mailbox names in a few
places, because those always use forward slashes, regardless of local file
system conventions. windows can handle forward slashes when opening files, so
test code that passes path strings with forward slashes straight to go stdlib
file i/o functions are left unchanged to reduce code churn. the regular
non-test code, or test code that uses path strings in places other than
standard i/o functions, does have the paths converted for consistent paths
(otherwise we would end up with paths with mixed forward/backward slashes in
log messages).
windows cannot dup a listening socket. for "mox localserve", it isn't
important, and we can work around the issue. the current approach for "mox
serve" (forking a process and passing file descriptors of listening sockets on
"privileged" ports) won't work on windows. perhaps it isn't needed on windows,
and any user can listen on "privileged" ports? that would be welcome.
on windows, os.Open cannot open a directory, so we cannot call Sync on it after
message delivery. a cursory internet search indicates that directories cannot
be synced on windows. the story is probably much more nuanced than that, with
long deep technical details/discussions/disagreement/confusion, like on unix.
for "mox localserve" we can get away with making syncdir a no-op.
2023-10-14 11:54:07 +03:00
mox . ConfigStaticPath = filepath . FromSlash ( "../testdata/store/mox.conf" )
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
mox . MustLoadConfig ( true , false )
2023-12-05 15:35:58 +03:00
acc , err := OpenAccount ( log , "mjl" )
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
tcheck ( t , err , "open account" )
defer func ( ) {
err = acc . Close ( )
tcheck ( t , err , "closing account" )
2024-04-16 18:33:54 +03:00
acc . CheckClosed ( )
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
} ( )
defer Switchboard ( ) ( )
// New account already has threading. Add some messages, check the threading.
deliver := func ( recv time . Time , s string , expThreadID int64 ) Message {
t . Helper ( )
2023-12-05 15:35:58 +03:00
f , err := CreateMessageTemp ( log , "account-test" )
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
tcheck ( t , err , "temp file" )
make mox compile on windows, without "mox serve" but with working "mox localserve"
getting mox to compile required changing code in only a few places where
package "syscall" was used: for accessing file access times and for umask
handling. an open problem is how to start a process as an unprivileged user on
windows. that's why "mox serve" isn't implemented yet. and just finding a way
to implement it now may not be good enough in the near future: we may want to
starting using a more complete privilege separation approach, with a process
handling sensitive tasks (handling private keys, authentication), where we may
want to pass file descriptors between processes. how would that work on
windows?
anyway, getting mox to compile for windows doesn't mean it works properly on
windows. the largest issue: mox would normally open a file, rename or remove
it, and finally close it. this happens during message delivery. that doesn't
work on windows, the rename/remove would fail because the file is still open.
so this commit swaps many "remove" and "close" calls. renames are a longer
story: message delivery had two ways to deliver: with "consuming" the
(temporary) message file (which would rename it to its final destination), and
without consuming (by hardlinking the file, falling back to copying). the last
delivery to a recipient of a message (and the only one in the common case of a
single recipient) would consume the message, and the earlier recipients would
not. during delivery, the already open message file was used, to parse the
message. we still want to use that open message file, and the caller now stays
responsible for closing it, but we no longer try to rename (consume) the file.
we always hardlink (or copy) during delivery (this works on windows), and the
caller is responsible for closing and removing (in that order) the original
temporary file. this does cost one syscall more. but it makes the delivery code
(responsibilities) a bit simpler.
there is one more obvious issue: the file system path separator. mox already
used the "filepath" package to join paths in many places, but not everywhere.
and it still used strings with slashes for local file access. with this commit,
the code now uses filepath.FromSlash for path strings with slashes, uses
"filepath" in a few more places where it previously didn't. also switches from
"filepath" to regular "path" package when handling mailbox names in a few
places, because those always use forward slashes, regardless of local file
system conventions. windows can handle forward slashes when opening files, so
test code that passes path strings with forward slashes straight to go stdlib
file i/o functions are left unchanged to reduce code churn. the regular
non-test code, or test code that uses path strings in places other than
standard i/o functions, does have the paths converted for consistent paths
(otherwise we would end up with paths with mixed forward/backward slashes in
log messages).
windows cannot dup a listening socket. for "mox localserve", it isn't
important, and we can work around the issue. the current approach for "mox
serve" (forking a process and passing file descriptors of listening sockets on
"privileged" ports) won't work on windows. perhaps it isn't needed on windows,
and any user can listen on "privileged" ports? that would be welcome.
on windows, os.Open cannot open a directory, so we cannot call Sync on it after
message delivery. a cursory internet search indicates that directories cannot
be synced on windows. the story is probably much more nuanced than that, with
long deep technical details/discussions/disagreement/confusion, like on unix.
for "mox localserve" we can get away with making syncdir a no-op.
2023-10-14 11:54:07 +03:00
defer os . Remove ( f . Name ( ) )
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
defer f . Close ( )
s = strings . ReplaceAll ( s , "\n" , "\r\n" )
m := Message {
Size : int64 ( len ( s ) ) ,
MsgPrefix : [ ] byte ( s ) ,
Received : recv ,
}
make mox compile on windows, without "mox serve" but with working "mox localserve"
getting mox to compile required changing code in only a few places where
package "syscall" was used: for accessing file access times and for umask
handling. an open problem is how to start a process as an unprivileged user on
windows. that's why "mox serve" isn't implemented yet. and just finding a way
to implement it now may not be good enough in the near future: we may want to
starting using a more complete privilege separation approach, with a process
handling sensitive tasks (handling private keys, authentication), where we may
want to pass file descriptors between processes. how would that work on
windows?
anyway, getting mox to compile for windows doesn't mean it works properly on
windows. the largest issue: mox would normally open a file, rename or remove
it, and finally close it. this happens during message delivery. that doesn't
work on windows, the rename/remove would fail because the file is still open.
so this commit swaps many "remove" and "close" calls. renames are a longer
story: message delivery had two ways to deliver: with "consuming" the
(temporary) message file (which would rename it to its final destination), and
without consuming (by hardlinking the file, falling back to copying). the last
delivery to a recipient of a message (and the only one in the common case of a
single recipient) would consume the message, and the earlier recipients would
not. during delivery, the already open message file was used, to parse the
message. we still want to use that open message file, and the caller now stays
responsible for closing it, but we no longer try to rename (consume) the file.
we always hardlink (or copy) during delivery (this works on windows), and the
caller is responsible for closing and removing (in that order) the original
temporary file. this does cost one syscall more. but it makes the delivery code
(responsibilities) a bit simpler.
there is one more obvious issue: the file system path separator. mox already
used the "filepath" package to join paths in many places, but not everywhere.
and it still used strings with slashes for local file access. with this commit,
the code now uses filepath.FromSlash for path strings with slashes, uses
"filepath" in a few more places where it previously didn't. also switches from
"filepath" to regular "path" package when handling mailbox names in a few
places, because those always use forward slashes, regardless of local file
system conventions. windows can handle forward slashes when opening files, so
test code that passes path strings with forward slashes straight to go stdlib
file i/o functions are left unchanged to reduce code churn. the regular
non-test code, or test code that uses path strings in places other than
standard i/o functions, does have the paths converted for consistent paths
(otherwise we would end up with paths with mixed forward/backward slashes in
log messages).
windows cannot dup a listening socket. for "mox localserve", it isn't
important, and we can work around the issue. the current approach for "mox
serve" (forking a process and passing file descriptors of listening sockets on
"privileged" ports) won't work on windows. perhaps it isn't needed on windows,
and any user can listen on "privileged" ports? that would be welcome.
on windows, os.Open cannot open a directory, so we cannot call Sync on it after
message delivery. a cursory internet search indicates that directories cannot
be synced on windows. the story is probably much more nuanced than that, with
long deep technical details/discussions/disagreement/confusion, like on unix.
for "mox localserve" we can get away with making syncdir a no-op.
2023-10-14 11:54:07 +03:00
err = acc . DeliverMailbox ( log , "Inbox" , & m , f )
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
tcheck ( t , err , "deliver" )
if expThreadID == 0 {
expThreadID = m . ID
}
if m . ThreadID != expThreadID {
t . Fatalf ( "got threadid %d, expected %d" , m . ThreadID , expThreadID )
}
return m
}
now := time . Now ( )
m0 := deliver ( now , "Message-ID: <m0@localhost>\nSubject: test1\n\ntest\n" , 0 )
m1 := deliver ( now , "Message-ID: <m1@localhost>\nReferences: <m0@localhost>\nSubject: test1\n\ntest\n" , m0 . ID ) // References.
m2 := deliver ( now , "Message-ID: <m2@localhost>\nReferences: <m0@localhost>\nSubject: other\n\ntest\n" , 0 ) // References, but different subject.
m3 := deliver ( now , "Message-ID: <m3@localhost>\nIn-Reply-To: <m0@localhost>\nSubject: test1\n\ntest\n" , m0 . ID ) // In-Reply-To.
m4 := deliver ( now , "Message-ID: <m4@localhost>\nSubject: re: test1\n\ntest\n" , m0 . ID ) // Subject.
m5 := deliver ( now , "Message-ID: <m5@localhost>\nSubject: test1 (fwd)\n\ntest\n" , m0 . ID ) // Subject.
m6 := deliver ( now , "Message-ID: <m6@localhost>\nSubject: [fwd: test1]\n\ntest\n" , m0 . ID ) // Subject.
m7 := deliver ( now , "Message-ID: <m7@localhost>\nSubject: test1\n\ntest\n" , 0 ) // Only subject, but not a response.
// Thread with a cyclic head, a self-referencing message.
c1 := deliver ( now , "Message-ID: <c1@localhost>\nReferences: <c2@localhost>\nSubject: cycle0\n\ntest\n" , 0 ) // Head cycle with m8.
c2 := deliver ( now , "Message-ID: <c2@localhost>\nReferences: <c1@localhost>\nSubject: cycle0\n\ntest\n" , c1 . ID ) // Head cycle with c1.
c3 := deliver ( now , "Message-ID: <c3@localhost>\nReferences: <c1@localhost>\nSubject: cycle0\n\ntest\n" , c1 . ID ) // Connected to one of the cycle elements.
c4 := deliver ( now , "Message-ID: <c4@localhost>\nReferences: <c2@localhost>\nSubject: cycle0\n\ntest\n" , c1 . ID ) // Connected to other cycle element.
c5 := deliver ( now , "Message-ID: <c5@localhost>\nReferences: <c4@localhost>\nSubject: cycle0\n\ntest\n" , c1 . ID )
c5b := deliver ( now , "Message-ID: <c5@localhost>\nReferences: <c4@localhost>\nSubject: cycle0\n\ntest\n" , c1 . ID ) // Duplicate, e.g. Sent item, internal cycle during upgrade.
c6 := deliver ( now , "Message-ID: <c6@localhost>\nReferences: <c5@localhost>\nSubject: cycle0\n\ntest\n" , c1 . ID )
c7 := deliver ( now , "Message-ID: <c7@localhost>\nReferences: <c5@localhost> <c7@localhost>\nSubject: cycle0\n\ntest\n" , c1 . ID ) // Self-referencing message that also points to actual parent.
// More than 2 messages to make a cycle.
d0 := deliver ( now , "Message-ID: <d0@localhost>\nReferences: <d2@localhost>\nSubject: cycle1\n\ntest\n" , 0 )
d1 := deliver ( now , "Message-ID: <d1@localhost>\nReferences: <d0@localhost>\nSubject: cycle1\n\ntest\n" , d0 . ID )
d2 := deliver ( now , "Message-ID: <d2@localhost>\nReferences: <d1@localhost>\nSubject: cycle1\n\ntest\n" , d0 . ID )
// Cycle with messages delivered later. During import/upgrade, they will all be one thread.
e0 := deliver ( now , "Message-ID: <e0@localhost>\nReferences: <e1@localhost>\nSubject: cycle2\n\ntest\n" , 0 )
e1 := deliver ( now , "Message-ID: <e1@localhost>\nReferences: <e2@localhost>\nSubject: cycle2\n\ntest\n" , 0 )
e2 := deliver ( now , "Message-ID: <e2@localhost>\nReferences: <e0@localhost>\nSubject: cycle2\n\ntest\n" , e0 . ID )
// Three messages in a cycle (f1, f2, f3), with one with an additional ancestor (f4) which is ignored due to the cycle. Has different threads during import.
f0 := deliver ( now , "Message-ID: <f0@localhost>\nSubject: cycle3\n\ntest\n" , 0 )
f1 := deliver ( now , "Message-ID: <f1@localhost>\nReferences: <f0@localhost> <f2@localhost>\nSubject: cycle3\n\ntest\n" , f0 . ID )
f2 := deliver ( now , "Message-ID: <f2@localhost>\nReferences: <f3@localhost>\nSubject: cycle3\n\ntest\n" , 0 )
f3 := deliver ( now , "Message-ID: <f3@localhost>\nReferences: <f1@localhost>\nSubject: cycle3\n\ntest\n" , f0 . ID )
// Duplicate single message (no larger thread).
g0 := deliver ( now , "Message-ID: <g0@localhost>\nSubject: dup\n\ntest\n" , 0 )
g0b := deliver ( now , "Message-ID: <g0@localhost>\nSubject: dup\n\ntest\n" , g0 . ID )
// Duplicate message with a child message.
h0 := deliver ( now , "Message-ID: <h0@localhost>\nSubject: dup2\n\ntest\n" , 0 )
h0b := deliver ( now , "Message-ID: <h0@localhost>\nSubject: dup2\n\ntest\n" , h0 . ID )
h1 := deliver ( now , "Message-ID: <h1@localhost>\nReferences: <h0@localhost>\nSubject: dup2\n\ntest\n" , h0 . ID )
// Message has itself as reference.
s0 := deliver ( now , "Message-ID: <s0@localhost>\nReferences: <s0@localhost>\nSubject: self-referencing message\n\ntest\n" , 0 )
// Message with \0 in subject, should get an empty base subject.
b0 := deliver ( now , "Message-ID: <b0@localhost>\nSubject: bad\u0000subject\n\ntest\n" , 0 )
b1 := deliver ( now , "Message-ID: <b1@localhost>\nSubject: bad\u0000subject\n\ntest\n" , 0 ) // Not matched.
// Interleaved duplicate threaded messages. First child, then parent, then duplicate parent, then duplicat child again.
i0 := deliver ( now , "Message-ID: <i0@localhost>\nReferences: <i1@localhost>\nSubject: interleaved duplicate\n\ntest\n" , 0 )
i1 := deliver ( now , "Message-ID: <i1@localhost>\nSubject: interleaved duplicate\n\ntest\n" , 0 )
i2 := deliver ( now , "Message-ID: <i1@localhost>\nSubject: interleaved duplicate\n\ntest\n" , i1 . ID )
i3 := deliver ( now , "Message-ID: <i0@localhost>\nReferences: <i1@localhost>\nSubject: interleaved duplicate\n\ntest\n" , i0 . ID )
j0 := deliver ( now , "Message-ID: <j0@localhost>\nReferences: <>\nSubject: empty id in references\n\ntest\n" , 0 )
dbpath := acc . DBPath
err = acc . Close ( )
tcheck ( t , err , "close account" )
2024-04-16 18:33:54 +03:00
acc . CheckClosed ( )
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
// Now clear the threading upgrade, and the threading fields and close the account.
// We open the database file directly, so we don't trigger the consistency checker.
db , err := bstore . Open ( ctxbg , dbpath , & bstore . Options { Timeout : 5 * time . Second , Perm : 0660 } , DBTypes ... )
err = db . Write ( ctxbg , func ( tx * bstore . Tx ) error {
up := Upgrade { ID : 1 }
err := tx . Delete ( & up )
tcheck ( t , err , "delete upgrade" )
q := bstore . QueryTx [ Message ] ( tx )
_ , err = q . UpdateFields ( map [ string ] any {
"MessageID" : "" ,
"SubjectBase" : "" ,
"ThreadID" : int64 ( 0 ) ,
"ThreadParentIDs" : [ ] int64 ( nil ) ,
"ThreadMissingLink" : false ,
} )
return err
} )
tcheck ( t , err , "reset threading fields" )
err = db . Close ( )
tcheck ( t , err , "closing db" )
// Open the account again, that should get the account upgraded. Wait for upgrade to finish.
2023-12-05 15:35:58 +03:00
acc , err = OpenAccount ( log , "mjl" )
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
tcheck ( t , err , "open account" )
err = acc . ThreadingWait ( log )
tcheck ( t , err , "wait for threading" )
check := func ( id int64 , expThreadID int64 , expParentIDs [ ] int64 , expMissingLink bool ) {
t . Helper ( )
m := Message { ID : id }
err := acc . DB . Get ( ctxbg , & m )
tcheck ( t , err , "get message" )
if m . ThreadID != expThreadID || ! reflect . DeepEqual ( m . ThreadParentIDs , expParentIDs ) || m . ThreadMissingLink != expMissingLink {
t . Fatalf ( "got thread id %d, parent ids %v, missing link %v, expected %d %v %v" , m . ThreadID , m . ThreadParentIDs , m . ThreadMissingLink , expThreadID , expParentIDs , expMissingLink )
}
}
parents0 := [ ] int64 { m0 . ID }
check ( m0 . ID , m0 . ID , nil , false )
check ( m1 . ID , m0 . ID , parents0 , false )
check ( m2 . ID , m2 . ID , nil , true )
check ( m3 . ID , m0 . ID , parents0 , false )
check ( m4 . ID , m0 . ID , parents0 , true )
check ( m5 . ID , m0 . ID , parents0 , true )
check ( m6 . ID , m0 . ID , parents0 , true )
check ( m7 . ID , m7 . ID , nil , false )
check ( c1 . ID , c1 . ID , nil , true ) // Head of cycle, hence missing link
check ( c2 . ID , c1 . ID , [ ] int64 { c1 . ID } , false )
check ( c3 . ID , c1 . ID , [ ] int64 { c1 . ID } , false )
check ( c4 . ID , c1 . ID , [ ] int64 { c2 . ID , c1 . ID } , false )
check ( c5 . ID , c1 . ID , [ ] int64 { c4 . ID , c2 . ID , c1 . ID } , false )
check ( c5b . ID , c1 . ID , [ ] int64 { c5 . ID , c4 . ID , c2 . ID , c1 . ID } , true )
check ( c6 . ID , c1 . ID , [ ] int64 { c5 . ID , c4 . ID , c2 . ID , c1 . ID } , false )
check ( c7 . ID , c1 . ID , [ ] int64 { c5 . ID , c4 . ID , c2 . ID , c1 . ID } , true )
check ( d0 . ID , d0 . ID , nil , true )
check ( d1 . ID , d0 . ID , [ ] int64 { d0 . ID } , false )
check ( d2 . ID , d0 . ID , [ ] int64 { d1 . ID , d0 . ID } , false )
check ( e0 . ID , e0 . ID , nil , true )
check ( e1 . ID , e0 . ID , [ ] int64 { e2 . ID , e0 . ID } , false )
check ( e2 . ID , e0 . ID , [ ] int64 { e0 . ID } , false )
check ( f0 . ID , f0 . ID , nil , false )
check ( f1 . ID , f1 . ID , nil , true )
check ( f2 . ID , f1 . ID , [ ] int64 { f3 . ID , f1 . ID } , false )
check ( f3 . ID , f1 . ID , [ ] int64 { f1 . ID } , false )
check ( g0 . ID , g0 . ID , nil , false )
check ( g0b . ID , g0 . ID , [ ] int64 { g0 . ID } , true )
check ( h0 . ID , h0 . ID , nil , false )
check ( h0b . ID , h0 . ID , [ ] int64 { h0 . ID } , true )
check ( h1 . ID , h0 . ID , [ ] int64 { h0 . ID } , false )
check ( s0 . ID , s0 . ID , nil , true )
check ( b0 . ID , b0 . ID , nil , false )
check ( b1 . ID , b1 . ID , nil , false )
check ( i0 . ID , i1 . ID , [ ] int64 { i1 . ID } , false )
check ( i1 . ID , i1 . ID , nil , false )
check ( i2 . ID , i1 . ID , [ ] int64 { i1 . ID } , true )
check ( i3 . ID , i1 . ID , [ ] int64 { i0 . ID , i1 . ID } , true )
check ( j0 . ID , j0 . ID , nil , false )
}
