YANZ UPLOAD ANJG
2021-05-04 10:50:51 -0700 Kevin McCarthy (23ccbe39)
* Update UPDATING file for 2.0.7.
M UPDATING
2021-05-03 13:11:30 -0700 Kevin McCarthy (7c4779ac)
* Fix seqset iterator when it ends in a comma.
If the seqset ended with a comma, the substr_end marker would be just
before the trailing nul. In the next call, the loop to skip the
marker would iterate right past the end of string too.
The fix is simple: place the substr_end marker and skip past it
immediately.
M imap/util.c
2021-04-28 10:59:03 -0700 Kevin McCarthy (e0f4d46b)
* Filter unprintables for 2231 encoded continuations too.
Changeset de8d8e8f a long time ago added the filter for 2047 and 2231
encoded values, to prevent strange issues. However, it looks like it
missed the case where the value is broken into continuations. Those
are assembled and charset converted in rfc2231_join_continuations().
Add a filter there too.
M rfc2231.c
2021-04-24 15:20:02 -0700 Kevin McCarthy (8801dfe9)
* Turn off pylint in gitlab-ci.
It dynamically installs the latest pylint via pip, meaning it can
start failing arbitrarily.
It also appears to halt the build even on warnings.
Since this is just checking a single contrib script, it's not worth my
trouble.
M .gitlab-ci.yml
2021-03-16 14:56:13 -0700 Kevin McCarthy (8eb29a4c)
* Check if mutt_prepare_template() fails for -H command line arg.
If the function fails, it will free the envelope, leading to a segv
just below.
Thanks to Phil Pennock for reporting the problem.
M main.c
2021-03-10 18:31:42 -0800 Kevin McCarthy (dcf1d11f)
* Fix resolve_types() newline removal and restoration.
If the line had a '\n', the previous code would leave it off. The
"buf[nl] = '\n'" at the end of the function was overwriting the old
nul character, but there was a new one where the original newline was.
However, if there was not a '\n', the code was removing the nul string
terminator, which could cause very bad things for searches, (which are
the only things that use fmt after resolve_types). Since this bug has
been there for 12 years, I'm guessing the lines always contain a
trailing '\n'.
Now, there is a potential problem with behavior changes here... This
could have an effect on the QuoteRegexp or SearchRE matches. It turns
out SearchRE is compiled with REG_NEWLINE, but the other isn't. Since
QuoteRegexp is (typically) only looking at the start of a line, I'm
going to take the risk of restoring the '\n'.
M pager.c
2021-03-06 11:15:07 -0800 Kevin McCarthy (98f8cb83)
* automatic post-release commit for mutt-2.0.6
M ChangeLog
M VERSION
2021-03-06 11:12:08 -0800 Kevin McCarthy (814f6d20)
* Update UPDATING file for 2.0.6 release.
M UPDATING
2021-02-25 14:19:19 -0800 Kevin McCarthy (de51ab4a)
* Fix (un)setenv to not return an error with unset env vars.
The code was returning -1 in those cases. This can cause hook
processors, such as send2-hook, to sleep 1 second without an error
message being set and displayed. It also causes the hook processor to
abort executing subsequent matching commands.
-1 should be reserved for syntax errors. For setenv querying, change
the return value to 0. For unsetenv, generate a message and also
change the return value to 0.
checks if err->data is set to determine whether to
print anything, and the behavior is used in the 'set' commands, so
this is established correct behavior.
M init.c
2021-02-05 15:34:37 -0800 Kevin McCarthy (0f8a2a13)
* Save CurrentFolder in background edit scope.
The CurrentFolder can actually be used to resolve an fcc-hook at fcc
time. We would want to use the folder they started composing in, to
be consistent with expectations.
M doc/manual.xml.head
M send.c
M send.h
2021-02-07 19:41:21 -0800 Kevin McCarthy (564b515c)
* Fix pattern compilation error for ~( !~>(~P) )
The problem is a nested recursive call followed by trailing
whitespace. Other codepaths ensure trailng whitespaces are processed
when processing the argument, but the parenthesis calls did not.
This meant the outer loop while check:
while (*ps.dptr)
{
SKIPWS (ps.dptr);
switch (*ps.dptr)
{
/* process (~P), for instance */
}
}
would return true for the trailing space(s), the SKIPWS would skip
over them, and then switch would try to evaluate '\0'.
I think a better solution would be to rearrange the calls like:
SKIPWS (ps.dptr);
while (*ps.dptr)
{
switch (*ps.dptr)
{
}
SKIPWS(ps.dptr);
}
to ensure any trailing whitespace is removed.
However, that's a bit risky for a stable branch commit. So instead,
just explicitly add the SKIPWS for the ~() thread ops and for the ()
groupings after processing the inside expression.
M pattern.c
2021-02-02 12:47:24 -0800 Kevin McCarthy (ca272c39)
* Fix contrib/markdown2html crash.
The script crashes if the last line in the message ends with a colon.
Thanks to Aitor Soroa (@asoroa) for the bug report and patch!
M contrib/markdown2html
2021-01-27 14:55:35 -0800 Kevin McCarthy (6df1891b)
* Fix another yesorno prompt, for autocrypt account creation.
I seem to have trouble remembering that ctrl-g is one of the
choices. :-(
There is yet another autocrypt prompt for "enabling prefer-encryption"
that *could* abort on ctrl-g. But, since it's not really an "action"
prompt, I'm keeping that one as a "no on ctrl-g".
M autocrypt/autocrypt_gpgme.c
2021-01-27 09:31:58 +0100 Christopher Zimmermann (2342b4d1)
* Correctly handle CTRL-G on remaining background_edit sessions prompt
M curs_lib.c
M curs_main.c
2021-01-21 12:31:43 -0800 Kevin McCarthy (da5e3282)
* automatic post-release commit for mutt-2.0.5
M ChangeLog
M VERSION
2021-01-21 12:27:50 -0800 Kevin McCarthy (dd66f4ae)
* Update UPDATING file for 2.0.5.
M UPDATING
2021-01-17 10:40:37 -0800 Kevin McCarthy (4a2becbd)
* Fix memory leak parsing group addresses without a display name.
When there was a group address terminator with no previous
addresses (including the group display-name), an address would be
allocated but not attached to the address list.
Change this to only allocate when last exists.
It would be more correct to not allocate at all unless we are inside a
group list, but I will address that in a separate commit to master.
M rfc822.c
2020-12-01 14:21:31 -0800 Kevin McCarthy (464a9bc6)
* Fix memory leak in imap_copy_messages().
mx.mbox (allocated by imap_parse_path) was not always freed before
return.
The sync_cmd and cmd buffers were also not always freed. One case was
on retrying after creating the mailbox, which would overwrite the
allocated pointers.
As long as I'm touching the buffers, convert them to use the buffer
pool. I think the mutt_buffer_clear() at the beginning of the retry
loop isn't necessary, but will keep it to make it clear any existing
values won't be reused a second time through the loop.
M imap/message.c
2021-01-13 19:06:49 -0800 Kevin McCarthy (b40c7f6c)
* Fix imap memory leaks.
The mx.mbox allocated by imap_parse_path() was not freed on error in a
couple places.
There is also a leak in imap_copy_message(), but this was fixed in
master in commit 1ec642e2. I'll cherry pick that into stable.
M imap/imap.c
2021-01-08 19:04:18 -0800 Kevin McCarthy (b67ae9ce)
* Fix color overlay when HAVE_COLOR is unset.
mutt_merge_colors() and mutt_attrset_cursor() were not defined in that
case, but are called in the menu and sidebar.
Change mutt_merge_colors() to just combine attributes in that case.
M color.c
2020-12-30 14:23:18 -0800 Kevin McCarthy (26f41dd1)
* automatic post-release commit for mutt-2.0.4
M ChangeLog
M UPDATING
M VERSION
2020-12-30 14:15:08 -0800 Kevin McCarthy (5cc62090)
* Bump copyright notices.
I'll run the update tool on source files in master a bit later.
M COPYRIGHT
M doc/manual.xml.head
M doc/mutt.man
M main.c
2020-12-29 13:13:09 -0800 Kevin McCarthy (b23f900b)
* Fix smtp debug segfault on invalid SmtpAuthenticators list.
sasl_client_start() sets the mech parameter only on success. dprint()
the whole mechlist on failure.
M smtp.c
2020-12-24 15:33:57 -0800 Kevin McCarthy (48d08860)
* Fix header cache BDB version checking.
Update BDB versions up to 6.2.
The scheme used here is, as Oswald nicely puts it, "totally insane",
and seems to come straight from the initial header cache commit in
c11667eaa3.
For a stable release, I'm just going to add a few more Jenga pieces to
the top.
However, going forward this is unmaintainable. Oswald reports using a
simple check for "db.h" and "-ldb" for about two decades with no
issues. I will make that change in master later this week.
M configure.ac
2020-12-21 15:30:01 -0800 Kevin McCarthy (11b18027)
* Fix offset to use LOFF_T in a couple places.
The BODY->hdr_offset was incorrectly of type long, which could result
in corrupted >2gb mbox files in some circumstances.
The uses in mh_rewrite_message() are not as serious, but they should
still be of type LOFF_T.
I found both of these fixes in a patch file,
bug-676388-largefile.patch, in the openSUSE mutt src rpm. It looks
like Harald Koenig was the original author of a larger patch in that
openSUSE ticket, which was reduced over time as fixes were made to
Mutt.
Note that patch file also incorrectly adjusted old_hdr_lines in mh.c.
I've removed that part, as HEADER->lines is type int.
Unfortunately, the BODY->hdr_offset type change will result in a
header cache change (i.e. invalidation). I'm not enthused about doing
that during a stable release, but the change is important enough to
merit it.
M mh.c
M mutt.h
2020-12-07 14:47:22 -0800 Kevin McCarthy (a7b839e5)
* Ensure idata->check_status is cleared on mailbox close.
I don't think this would cause any issues, but it should be cleared
here in any case.
M imap/imap.c
2020-12-07 13:03:41 -0800 Kevin McCarthy (b5ef1155)
* Abort IMAP open if condstore/qresync updates fetch fails.
An error in imap_cmd_step() was not being properly returned to the
caller.
M imap/message.c
2020-12-04 10:54:21 -0800 Kevin McCarthy (a51f058f)
* automatic post-release commit for mutt-2.0.3
M ChangeLog
M VERSION
2020-12-04 10:48:04 -0800 Kevin McCarthy (c6f114f8)
* Update UPDATING file for 2.0.3 release.
M UPDATING
2020-12-02 13:03:06 -0800 Kevin McCarthy (3ba8dac0)
* Fix pager dropped input on SigWinch flag handling.
The code to process the SigWinch flag ocurred after, and would drop an
input character when the flag was set outside of, km_dokey(). This
could happen, for instance, when a pipe operation was invoked.
Thanks to Vincent Lefèvre for uncovering the problem.
M pager.c
2020-11-30 15:53:49 -0800 Kevin McCarthy (f420be68)
* Ensure mutt_extract_token() never returns a NULL dest->data.
Commit e5a32a61 removed a 'mutt_buffer_addch (dest, 0)' at the end of
the function. Most callers had been converted to use the buffer pool,
and the call was strange since buffers self-terminate.
However, this line covered up logic errors in some of the callers,
which assumed the buffer->data could not be NULL afterwards.
I will try to fix up callers with the logic errors in master. This is
to fix the problem in stable, and also ensure future callers don't
make the same mistake.
M init.c
2020-11-29 13:44:30 -0800 Kevin McCarthy (cfdcfa7f)
* Fix REPLY_TO environment variable handling.
Commit 4e153adf changed this code to reuse the function buffer
variable, but forgot to rewind the buffer for parsing in
parse_my_hdr().
Additionally commit e5a32a61 removed an extra "null termination"
mutt_buffer_addch() at the end of mutt_extract_token(). This caused a
NULL value to be passed to the strpbrk() in parse_my_hdr(), causing a
segv. Change to use a buffer pool token parameter instead.
I actually think, like with the previous IMAP mailbox handling, this
method of adding a my_hdr is dangerous. I'll look into refactoring it
in master instead.
Thanks to Paul Nevai for reporting the problem and tracking down the
backtrace.
M init.c
2020-11-29 10:48:22 -0800 Kevin McCarthy (dbdf481c)
* Fix undefined NULL pointer arithmetic.
clang 10 is giving a warning about arithmetic with a NULL pointer. To
prevent any problems, add checks in the BUFFER increase-size handling
functions.
M buffer.c
2020-11-25 13:46:47 -0800 Kevin McCarthy (9109eff8)
* Fix exact-address recording of last value.
If the last address was also terminated by a comma (e.g: foo@local,
bar@local,) the exact-address was incorrectly overwriting the recorded
value.
M rfc822.c
2020-11-24 12:54:00 -0800 Kevin McCarthy (d73a83f7)
* Fix exact-address handling when addr->personal is set.
The exact-address compile-time option takes an exact copy of an
address when it is parsed, and prints that out when outputting the
address. The idea is to preserve older "user@host (Name)" syntax.
Unfortunately, when code tries to "update" the personal/name field, it
needs to clear the exact-address copy for it to have any effect.
An object-oriented design encapsulating the setting would help prevent
this problem. It might be desirable to create a C function instead,
but callers would have to remember it, and such a thing isn't common
in the Mutt codebase.
Another patch (I'm still debating applying) changes the address parser
to discard the exact-address copy when it exactly matches the
mailbox. However, that still won't fix every case (and it makes the
parser even more difficult to follow.)
So this commit (to stable) takes the straightforward approach.
It fixes bugs in:
* Alias creation. The "personal name" prompt was ignored.
* Autocrypt initialization from address setting with $realname.
* $pgp_getkeys_command handling.
* Query menu results.
* "unset $reverse_realname" handling.
* $from handling of $realname.
* Bounce Resent-From handling of $realname.
M alias.c
M autocrypt/autocrypt.c
M pgpinvoke.c
M query.c
M send.c
M sendlib.c
2020-11-20 09:23:29 -0800 Kevin McCarthy (d9268908)
* automatic post-release commit for mutt-2.0.2
M ChangeLog
M VERSION
2020-11-20 09:20:01 -0800 Kevin McCarthy (e4fd9247)
* Update UPDATING file for 2.0.2.
M UPDATING
2020-11-16 10:20:21 -0800 Kevin McCarthy (04b06aaa)
* Ensure IMAP connection is closed after a connection error.
During connection, if the server provided an illegal initial response,
Mutt "bailed", but did not actually close the connection. The calling
code unfortunately relied on the connection status to decide to
continue with authentication, instead of checking the "bail" return
value.
This could result in authentication credentials being sent over an
unencrypted connection, without $ssl_force_tls being consulted.
Fix this by strictly closing the connection on any invalid response
during connection. The fix is intentionally small, to ease
backporting. A better fix would include removing the 'err_close_conn'
label, and perhaps adding return value checking in the caller (though
this change obviates the need for that).
This addresses CVE-2020-28896. Thanks to Gabriel Salles-Loustau for
reporting the problem, and providing test cases to reproduce.
M imap/imap.c
2020-11-19 15:06:51 -0800 Keld Simonsen (d4c97068)
* Updated Danish translation.
M po/da.po
2020-11-14 13:16:03 -0800 Kevin McCarthy (42e08237)
* automatic post-release commit for mutt-2.0.1
M ChangeLog
M VERSION
2020-11-14 13:10:45 -0800 Kevin McCarthy (78fe7d4e)
* Update UPDATING file for 2.0.1.
M UPDATING
2020-11-12 09:42:28 -0800 Kevin McCarthy (894a49f6)
* Clarify pattern completion uses .
Some users may have bound tab to another function, so be specific.
M UPDATING
2020-11-09 10:59:44 +0100 Remco Rijnders (86d64caa)
* Consistently use uint32_t (closes #294)
M hcache.c
M mutt_random.c
2020-11-07 12:18:03 -0800 Kevin McCarthy (3d08634b)
* automatic post-release commit for mutt-2.0.0
M ChangeLog
M VERSION
M po/bg.po
M po/ca.po
M po/cs.po
M po/da.po
M po/de.po
M po/el.po
M po/eo.po
M po/es.po
M po/et.po
M po/eu.po
M po/fi.po
M po/fr.po
M po/ga.po
M po/gl.po
M po/hu.po
M po/id.po
M po/it.po
M po/ja.po
M po/ko.po
M po/lt.po
M po/nl.po
M po/pl.po
M po/pt_BR.po
M po/ru.po
M po/sk.po
M po/sv.po
M po/tr.po
M po/uk.po
M po/zh_CN.po
M po/zh_TW.po
2020-11-07 11:30:13 -0800 Kevin McCarthy (0c8f4357)
* Add missing new header files to EXTRA_mutt_SOURCES.
I forgot to add the mutt_lisp.h and mutt_random.h files there.
M Makefile.am
2020-11-07 11:01:47 -0800 Kevin McCarthy (a374cea8)
* Update UPDATING file for 2.0.
M UPDATING
2020-11-06 14:31:57 -0800 Vsevolod Volkov (3e9c605d)
* Updated Russian translation.
M po/ru.po
2020-11-06 14:29:11 -0800 Vsevolod Volkov (3317883a)
* Updated Ukrainian translation.
M po/uk.po
2020-11-06 14:26:09 -0800 Ivan Vilata i Balaguer (32fcad9f)
* Updated Catalan translation.
M po/ca.po
2020-11-06 14:02:49 -0800 Kevin McCarthy (f5aa9382)
* Disable normalization after expand_path().
The algorithm used was incorrect, for '..' expansion with symlinks
involved. Furthermore, mutt_pretty_mailbox() takes care of this for
us.
To be conservative, just before the release, I'm leaving the function
but converting it to a noop. I'll pull the function out after the
release.
Thanks to Oswald Buddenhagen for pointing out the bug!
M muttlib.c
2020-11-06 13:18:27 -0800 Kevin McCarthy (6704caf4)
* Don't relative-path expand for fcc-hook and save-hook.
"fcc-hook ~x. \\^" used to work, because mutt_addr_hook() calls
mutt_make_string(), which performs backslash expansion.
The same would happen for save-hook.
Many thanks to Oswald Buddenhagen for reporting the issue.
M hook.c
M muttlib.c
M protos.h
2020-11-04 23:01:55 +0100 Grzegorz Szymaszek (533c38f4)
* Fix the Polish translation of “Attachments”
I have misunderstood the “Atts” abbreviation for “Attributes”, while it
actually means “Attachments”.
M po/pl.po
2020-11-04 20:15:52 +0100 Grzegorz Szymaszek (cdc99643)
* Update the Polish translation for Mutt 2.0
M po/pl.po
2020-11-03 16:50:23 +0100 Flammie Pirinen (f26731fa)
* updated finnish translation
M po/fi.po
2020-11-03 07:11:00 -0800 Kevin McCarthy (e677c196)
* Minor fix to UPDATING file.
M UPDATING
2020-10-31 15:35:53 +0100 Petr Písař (d752a64c)
* Czech translation updated for 2.0
M po/cs.po
2020-10-29 14:46:30 -0400 Remco Rijnders (f6fb0ca9)
* Adjust Makefile.am to exclude BEWARE file
M Makefile.am
2020-10-29 14:24:08 -0400 Remco Rijnders (37a1950a)
* Move contents from BEWARE to devel-notes.txt
D BEWARE
M doc/devel-notes.txt
2020-10-29 13:59:13 -0400 Remco Rijnders (4fa19ba4)
* Change instructions to subscribe to dev mail list
M doc/devel-notes.txt
2020-10-25 15:07:55 -0700 Kevin McCarthy (9258922a)
* Add a more explicit mention of ^G in the manual.
It's a FAQ, so I think is worth emphasizing in the "getting started"
section.
M doc/manual.xml.head
2020-10-25 20:49:22 +0100 Christopher Zimmermann (cafe0fb5)
* Allow to abort on question about multipart/alternative
M send.c
2020-10-24 09:13:19 -0400 Remco Rijnders (63e8ba68)
* Updated Dutch translation.
M po/nl.po
2020-10-23 15:50:06 -0700 Emir Sarı (d8bbd22c)
* Update Turkish translations.
M po/tr.po
2020-10-23 19:10:58 +0200 Olaf Hering (18a07d14)
* refresh de.po
Signed-off-by: Olaf Hering
M po/de.po
2020-10-22 19:41:14 -0700 Kevin McCarthy (e1089b5e)
* Remove obsolete stamp-h.in.
This file was used a long time ago (pre-1.0) when Makefile.in was
checked in.
We now generate Makefile.in from Makefile.am, and automake appears to
use a different mechanism of time-stamping, using stamp-h + $counter.
D stamp-h.in
2020-10-17 18:55:27 -0700 Kevin McCarthy (1036f0ed)
* Add an initial change list for 2.0.0 to the UPDATING file.
M UPDATING
2020-10-17 18:54:52 -0700 Kevin McCarthy (47fa1503)
* Add the mailboxes history category to the manual.
M doc/manual.xml.head
2020-10-14 14:29:27 +0200 Philipp Klaus Krause (36640a88)
* Since the string from strerror should never be modified, use const.
M curs_lib.c
2020-10-12 15:34:44 -0700 Kevin McCarthy (1061dcbc)
* Create $attach_save_dir.
This will be used when saving attachments via
mutt_save_attachment_list().
Try to create the directory if it doesn't exist. If we're unable to
chdir or create the directory just continue on, using cwd.
M globals.h
M init.h
M recvattach.c
2020-10-09 09:20:11 -0700 Kevin McCarthy (f8263764)
* Fix mutt_oauth.py.README example.
$imap_authenticators should be colon delimited.
M contrib/mutt_oauth2.py.README
2020-10-06 13:39:55 +1100 Cameron Simpson (0241b030)
* doc/manual.xml.head: Most common mail sending keys: replace "compose" with "mail", incorrect function name
M doc/manual.xml.head
2020-10-03 15:43:06 -0700 Kevin McCarthy (489ade5c)
* Move MuttLisp example descriptions before the code.
M doc/manual.xml.head
2020-10-01 13:33:55 -0700 Kevin McCarthy (ed9303a8)
* More ansi/special cleanup.
Relocate the checks for a->attr and special around the blocks that use
them.
M pager.c
2020-09-30 09:49:17 -0700 Kevin McCarthy (65b1ceab)
* Don't do ansi coloring on a search result.
M pager.c
2020-09-29 18:46:27 -0700 Kevin McCarthy (73a7bf2c)
* Don't free and reuse ansi colors.
You can't reallocate a color-pair to another color while the previous
one is still on the screen.
Since there are at most 64 ansi-color combinations, just let them
accumulate.
M pager.c
2020-09-29 14:20:24 -0700 Kevin McCarthy (a3a4d12f)
* Fix $allow_ansi end-of-line handling for attachments.
Attachment viewing doesn't set MUTT_SHOWCOLOR, but for proper display
we need to call resolve_color() before clearing to end-of-line.
M pager.c
2020-09-29 14:17:27 -0700 Kevin McCarthy (15372907)
* Separate special color setting from $allow_ansi colors.
The ColorDefs for MT_COLOR_BOLD and MT_COLOR_UNDERLINE should only be
applied for the special formatting. They shouldn't apply when
$allow_ansi has a bold/underline sequence.
Change the exclusive or operator to a regular or. Exclusive or
implies we are doing some kind of toggling, which we are not in this
case. It just confuses things.
Don't make the ansi underline, reverse, and blink mutually exclusive.
M pager.c
2020-09-23 11:08:03 -0700 Kevin McCarthy (b0ccf259)
* Delay $hostname setting until after the muttrc is evaluated.
Commit 5c5c34f2 made back in 1.6 changed Fqdn setting to use
gethostname() and getaddrinfo() to get the canonical domain. This is
more accurate but can cause startup delays for systems where the DNS
resolution is not set up properly.
Because this occurred before muttrc reading, there was no workaround
except to "fix DNS".
Change Fqdn ($hostname) setting to occur after the muttrc and '-e'
argument processing occur.
This is a possible breaking change if users rely on $hostname inside
their muttrc, for example 'source "muttrc.$hostname"'. The workaround
would be to put something like (depending on the system type): 'set
hostname = `hostname --fqdn`' in the muttrc above that invocation.
Also note that we still set Hostname (the internal variable) early,
because it is used in more places than Fqdn, such as tempfiles.
M init.c
M init.h
2020-09-22 11:17:07 -0700 Kevin McCarthy (8a2fc801)
* Note that $cursor_overlay affects tree colors too.
Change the font on "default" to indicate that actual value allows for
color overlays between the layers.
M init.h
2020-09-20 18:46:48 -0700 Kevin McCarthy (9204b24e)
* Reenable $ssl_force_tls.
The next release will be 2.0, providing more justification for
(reasonable) breaking changes. In 2020, this should be the default.
The documentation seems to already make appropriate warnings about
$tunnel_is_secure. Since that option is introduced this release,
there is no need to worry about existing users with it unset.
However, there were a couple stray double-quotes in the documentation
for that, so clean those up in this commit.
M doc/manual.xml.head
M init.h
2020-09-19 15:31:36 -0700 Kevin McCarthy (2ce2fb7b)
* Add refresh option to smime_keys man page.
M doc/smime_keys.man
2020-09-18 09:39:18 -0700 Kevin McCarthy (34ba89ec)
* Add missing full stops in smime_keys.man page.
Thanks to hmartink and the manpage-l10n project for pointing out the
issues.
M doc/smime_keys.man
2020-09-16 09:44:07 -0700 Kevin McCarthy (e91313b8)
* Remove casts for mutt_random_bytes() argument.
Unneeded casts can hide issues later on, so take them out.
M mutt_random.c
M muttlib.c
M sendlib.c
2020-09-15 14:32:19 -0700 Kevin McCarthy (ee2b9d1d)
* Remove message-id security leaks section of manual.
The message-id generator has been changed to use a combination of time
and random components for the left side.
M doc/manual.xml.head
2020-09-08 14:05:49 -0400 Remco Rijnders (9da4e6e1)
* Change Message-ID to be more unique and leak less information
A Message-ID should be globally unique. Currently mutt generates this ID
based on the current date and time, followed by ".G", followed by a letter
A to Z (A for the 1st and 27th email sent, Z for the 26th, etc.), followed
by the pid of the active mutt process, followed by "@" and the configured
fqdn.
This can lead to information being leaked as to an users email habits and
activities, which might be undesirable.
By replacing everything left of the "@" in the Message-ID with a Base64
encoded timestamp and 64 bits of randomness, we no longer include this
information.
M sendlib.c
2020-08-31 13:48:51 -0400 Remco Rijnders (b48233f7)
* Use PRIu64 macro as format when printing uint64_t values
M muttlib.c
2020-08-31 13:10:25 -0400 Remco Rijnders (8ccd96db)
* Implement LFRS113 PRNG functions
- Instead of relying on random() implementations which can be of
questionable quality or relying on the presence of /dev/urandom, we
implement our own PRNG implementation that uses the LFRS113 PRNG algorithm
by Pierre L'Ecuyer.
We seed this PRNG with values based on time, pid and ppid. It is OK if not
all seeds are of the highest quality as all four seeds would have to be
known to predict the numbers generated. In addition to this, we also use
/dev/urandom values (if available) that we mix into our four seeds.
In case we are reseeding we will also reuse our existing state information
for setting the new seed values.
- Add a function to Base64 encode 96 random bits
M Makefile.am
M init.c
A mutt_random.c
A mutt_random.h
M muttlib.c
M sendlib.c
2020-09-13 00:05:22 +0100 isdtor (56de020a)
* Add note about $smime_default_key to use with GPGME.
When using GPGME for S/MIME, the key id used should be the id
displayed by "gpgsm --list-keys".
M init.h
2020-09-06 15:39:07 -0700 Kevin McCarthy (4219d5f9)
* Block signals during mbox-append operation.
I first noticed this back in 2017, but no one replied to my query to
mutt-dev and I forgot about it too.
Thanks to Oswald Buddenhagen for the followup when he was going
through the mailing list archives, confirming the mistake.
I've also reviewed MUTT_APPEND and MUTT_NEWFOLDER uses to make sure
the context is properly closed (and the signals restored).
M mbox.c
2020-09-05 15:15:40 -0700 Kevin McCarthy (7157d009)
* Removed unused mutt_expand_path()
All usages now call mutt_buffer_expand_path().
M muttlib.c
M protos.h
2020-09-05 15:11:59 -0700 Kevin McCarthy (d76aebd2)
* Convert _mutt_enter_string() to use buffer expand_path function.
This is the last usage of mutt_expand_path(). Converting all of
_mutt_enter_string() is a large project, so instead just localize to
use a BUFFER for the mutt_expand_path() call.
M enter.c
2020-09-05 14:10:18 -0700 Kevin McCarthy (4a2f1067)
* Remove unused mutt_quote_filename().
All callers now call the buffer function instead.
M lib.c
M lib.h
2020-09-04 19:46:02 -0700 Kevin McCarthy (bf68f089)
* Convert mutt_create_alias and helpers to use buffer pool.
This is working towards removing a few of the partially converted path
functions, such as mutt_expand_path().
M alias.c
M protos.h
2020-09-06 23:21:15 +0200 Olaf Hering (2d9c05c9)
* Fix typos in de.po
Signed-off-by: Olaf Hering
M po/de.po
2020-09-04 14:16:21 -0700 Kevin McCarthy (c4fe5624)
* Use muttmua ubuntu image for shellcheck.
M .gitlab-ci.yml
2020-09-04 14:11:55 -0700 Kevin McCarthy (0f97dd3a)
* Turn off shellcheck unused var warning for doc/instdoc.sh.
M doc/instdoc.sh.in
2020-08-30 13:26:29 -0700 Kevin McCarthy (7ae6d9aa)
* Disable relative expansion for signature and source pipes.
Relative expansion shouldn't be performed when those values are to be
executed as a pipe for the output.
I believe this issue only affects $signature and the source command,
so for now I'm disabling relative expansion for those two cases. If
there is too much more breakage, I will consider backing out the
relative expansion and disabling the :cd command.
Thanks to Aaron Schrab for reporting this issue.
M init.c
2020-08-28 15:51:55 -0700 Kevin McCarthy (2ba1d11f)
* Normalize expanded paths with '.' or '..' in them.
This usage is not too likely, but to prevent strange looking mailbox
paths, make sure the path is normalized after being expanded.
M muttlib.c
2020-07-29 13:44:17 -0700 Kevin McCarthy (6fc79323)
* Remove relative path expansion for some cases.
Some paths, such as the folder browser, or certain hook values,
should be expanded for shortcuts, but not for relative paths.
Configuration variables that hold commands that search by PATH should
also not be expanded.
Create a separate call, and a separate data type for config vars.
M browser.c
M commands.c
M doc/makedoc.pl
M hook.c
M init.c
M init.h
M muttlib.c
M protos.h
M recvattach.c
2020-07-26 19:10:57 -0700 Kevin McCarthy (0f455d51)
* Change expand_path() to expand relative paths.
The introduction of the "cd" command can make relative paths
unreliable. Change expand_path() to expand relative paths.
Change pretty_mailbox() to contract relative paths, but only when the
folder is outside of homedir, or when cwd is underneath the homedir.
To try and keep the sidebar from changing, use the pretty_mailbox code
to contract relative paths only.
M muttlib.c
M sidebar.c
2020-08-29 12:30:18 -0700 Kevin McCarthy (092bc204)
* Merge branch 'stable' into master
2020-08-29 12:29:50 -0700 Kevin McCarthy (f34d0909)
* automatic post-release commit for mutt-1.14.7
M ChangeLog
M VERSION
2020-08-27 15:11:25 -0700 Kevin McCarthy (55c09cfb)
* Merge branch 'stable' into master
2020-08-25 12:29:14 -0400 Remco Rijnders (20165b4e)
* Remove always true conditional (#if 1)
M main.c
2020-08-25 12:23:37 -0400 Remco Rijnders (9ef96f1f)
* Remove always true conditional (#if 1)
M strcasestr.c
2020-08-25 12:16:52 -0400 Remco Rijnders (ae3014e9)
* Make returns after mx_fastclose_mailbox calls more consistent and cleaner
M mx.c
2020-08-25 12:14:33 -0400 Remco Rijnders (97398c4f)
* Remove always true conditional (#if 1)
M imap/imap.c
2020-08-22 07:43:15 -0700 William Yardley (3f856d11)
* refactor: update shell scripts for shellcheck warnings
This updates shell scripts in the project to resolve shellcheck lint
warnings.
A couple of warnings are ignored instead, at least for now.
M check_sec.sh
M contrib/bgedit-detectgui.sh
M contrib/iconv/make.sh
M gen_defs
M txt2c.sh
2020-08-24 12:12:39 -0700 William Yardley (6f4c0f7a)
* gen_defs: switch from $(()) to expr
For greater compatibility with Solaris's Bourne shell, switch to "expr" for
numeric evaluation
M gen_defs
2020-08-21 21:43:23 -0700 William Yardley (18dc7be4)
* ci: enable shellcheck
M .gitlab-ci.yml
M Makefile.am
2020-08-25 09:00:26 -0700 Kevin McCarthy (2d1da9c0)
* Merge branch 'stable' into master
2020-08-21 22:50:55 -0700 William Yardley (2f5a9496)
* mutt_oauth2: update for pylint / flake8 warnings
* Update style to resolve some pylint / flake8 warnings
* Run pylint / flake8 in CI
M .gitlab-ci.yml
A .pylintrc
M contrib/mutt_oauth2.py
2020-08-21 15:22:50 -0700 Alexander Perlis (c0218ade)
* Updates to contrib/mutt_oauth2.py and README.
The newer version of the mutt_oauth2.py script incorporates the
following changes:
- Uses /usr/bin/env at top
- Many formatting changes to appease pylint
- Improvement to POP test error message output
Also attached is a README that has instructions for both Microsoft and Google.
M contrib/mutt_oauth2.py
M contrib/mutt_oauth2.py.README
2020-07-23 19:15:11 +0300 Maxim Tarasov (a563ce85)
* Change hardcoded subject of replies
This affects prefilled subject of a reply to an email with an empty
subject.
M send.c
2020-08-13 13:00:01 -0700 Kevin McCarthy (2bfb7fed)
* Merge branch 'stable' into master
2020-08-13 19:07:10 +0200 Vincent Lefevre (c994f365)
* Updated French translation.
M po/fr.po
2020-08-10 11:19:46 -0700 Kevin McCarthy (a84e8823)
* Improve clarity of help format_line() splitting.
Thanks to Zero King for the initial patch. He says the original line
confused some static analysis tools.
I don't see anything wrong with the operator precedence, but as long
as we are cleaning up the line, split assignment and conditional
evaluation to make it even clearer.
M help.c
2020-08-06 15:50:53 -0700 Kevin McCarthy (17df35f0)
* Merge branch 'stable'
2020-08-04 11:03:08 -0700 Kevin McCarthy (8e8c586d)
* Fix manual typo.
M doc/manual.xml.head
2020-08-03 16:14:29 -0700 Kevin McCarthy (2fbc29ec)
* Allow my_var assignments to reference the $my_var in the value.
This is important for MuttLisp, where the current value could be used
to determine the new assignment.
M init.c
2020-08-01 19:55:26 -0700 Kevin McCarthy (312e9bd9)
* Fix MuttLisp extract_token() call to remove trailing whitespace.
MoreArgs() doesn't look for whitespace, and so assumes
mutt_extract_token() will remove trailing whitespace too.
M init.c
2020-07-27 08:56:10 +0200 Olaf Hering (7fe1aaad)
* refresh de.po
Signed-off-by: Olaf Hering
M po/de.po
2020-07-26 03:04:33 +0300 Maxim Tarasov (0b3fc03c)
* Fix man section in reference to mutt_dotlock
M init.h
2020-07-26 03:00:54 +0300 Maxim Tarasov (661bd665)
* Update mutt.1 manpage
* Change the wording for some options to be imperative.
* Rename parameters to be descriptive ("draft" instead of "file").
* Mention possible values for -m parameter.
* Change "file [...]" to "file ..." as ellipsis already implies it's
optional.
* Format entries in the SYNOPSIS to avoid wrapping inside square
brackets.
* Space out examples in -a description.
* Use consistent formatting everywhere: options using B, paths and
option values using I, manual references using BR.
* Group together descriptions of EDITOR and VISUAL environment
variables since they're closely related. Mention the default editor.
* Add references to these environment variables: EGDSOCKET, RANDFILE,
LC_ALL, LC_CTYPE, LANG, and TEXTDOMAINDIR.
* Clarify the way TMPDIR is used.
* Add ~/.muttdebug0 to the FILES section.
* Add missing manual pages to the SEE ALSO section. Sort the list.
* Add line breaks after sentences (this is recommended in roff manual
to enable sentence processing; sentences are double spaced
automatically).
M doc/mutt.man
2020-07-26 13:47:06 +0000 Asif Talybov (be906781)
* Update russian translation - add a missing character and fix typo
M po/ru.po
2020-07-25 16:08:03 -0700 Kevin McCarthy (317abd8f)
* Fix a few typos in the manual.
M doc/manual.xml.head
M init.h
2020-07-25 16:07:14 -0700 Kevin McCarthy (9ab12e1e)
* Add missing commands in the muttrc.man page.
M doc/muttrc.man.head
2020-07-25 14:25:43 -0700 Kevin McCarthy (692a6069)
* Add empty-request lines to muttrc.man for spacing.
Adding them makes it a bit easier to see where the different commands
stop and start.
M doc/muttrc.man.head
2020-07-25 13:03:09 -0700 Kevin McCarthy (9edbbac1)
* Change POP3 oauth to not use initial response.
The Microsoft documents indicate their implementation requires a
> AUTH XOAUTH2
< +
> token
sequence.
I don't have their services configured to test, but it's safer to
assume not all implementations will support initial response.
I did test Gmail's and they are fine with the second round trip.
M pop_auth.c
2020-07-25 09:29:21 -0700 Kevin McCarthy (e050c314)
* Improve w3m invocation for manual.txt generation.
This fix is also from Ambrose Li (@gniw). The invocation makes sure
w3m won't attempt a remote connection if manual.html doesn't exist and
a proxy is configured, by using manual.txt as stdin.
It also gives a non-zero exit status in that case, and makes sure
input is processed as utf-8.
M doc/Makefile.am
2020-07-25 04:18:03 +0300 Maxim Tarasov (daab98e3)
* Change \fC to \fB during muttrc.man generation
\fC is not actually a valid escape sequence. Using it results in
misplaced underlined formatting.
M doc/makedoc.pl
2020-07-24 16:02:32 -0700 Kevin McCarthy (a21c9fc1)
* Merge branch 'stable'
2020-07-23 17:11:47 +0300 Maxim Tarasov (6479ec5c)
* Mention $XDG_CONFIG_HOME in the manpage
Mutt supported processing $XDG_CONFIG_HOME/mutt/muttrc for a while
now, but mutt(1) man page hasn't been updated to reflect that.
This will also replace .IP with .TP to use one line per file name and
highlight file names with .I which is common practice in manual
formatting.
M doc/mutt.man
2020-07-22 18:46:51 -0700 Kevin McCarthy (3c2501e5)
* Fix redraw_motion() redraw bug.
Commit e2a28006 moved the cur_color assignment so that $arrow_cursor
could also use the value for overlays.
Unfortunately, it at the same time moved the call *after* the call to
mutt_window_move(), which would introduce a drawing bug if
menu->color() went across IMAP.
The previous commit removed overlays for $arrow_cursor, also removing
the need for cur_color to be assigned for both. Move it back to where
it was before.
M menu.c
2020-07-22 18:22:03 -0700 Maxim Tarasov (12ff55e2)
* Add color overlay to print_enriched_string().
It is called by main menu drawing routines and contains its own
conditional coloring of a tree component. As is, everything in there
gets overwritten by the indicator. Furthermore, do_color parameter
specifically disables any coloring in order for indicator highlight to
overwrite everything else.
We need to merge three colors here:
1) index color;
2) tree color, that is drawn within menu using its own color;
3) indicator color, that, if active, is drawn over both of them.
Combining these colors is implemented in that order.
After some testing, I think it doesn't make sense to merge colors when
$arrow_cursor is on. This part is reverted to the original behavior.
Make print_enriched_string() coloring behavior self-contained, instead
of assuming the proper base_color is already set. This makes the
callers simpler and allows them to only handle $arrow_cursor logic.
M menu.c
2020-07-21 14:40:50 -0700 Maxim Tarasov (aa8e6d4c)
* Fix cursor overlay logic.
Attributes should always be merged, not only when indicator background
or foreground is set.
M color.c
2020-07-20 19:43:58 -0700 Kevin McCarthy (e12028bd)
* Don't increment color refcount for overlays.
Add a ref parameter, setting to 0 for the overlays.
The pager also allocates temporary colors, but seems to take the
trouble to try and free them too.
M color.c
M pager.c
M protos.h
2020-07-20 18:22:41 -0700 Kevin McCarthy (f2e96586)
* Add $cursor_overlay, default unset, to control overlaying.
Although I think the new behavior is desirable, experience has shown
that changes need to be controlled by an option and usually defaulting
the same as before.
M color.c
M init.h
M menu.c
M mutt.h
M protos.h
M sidebar.c
2020-07-20 22:50:10 +0300 Maxim Tarasov (e2a28006)
* Add cursor overlay capability.
Currently, when the indicator line is over an unread message, full
indicator color definition is used, and the line is no longer
displayed bold. It's harder to notice when using default reverse style
for the indicator, but it becomes much more prominent when you have
black on white display and only subtle change in the indicator
background.
We could improve this by merging indicator color definition with color
definition of the underlying menu item. Here is how it's done:
1. First color definitions for the indicator and underlying menu item
are taken and split back to foreground, background and attributes
(meaning bold, underline, etc.).
2. If indicator does not specify foreground or background color, new
color pair is created by using underlying menu item as base. Else
indicator's own foreground and background are used (current behavior).
3. Then attributes of both colors are combined together. For example,
if indicator has 'underline' and menu item has 'bold' result will be
highlighted as 'bold underline'.
4. The resulting color is used in place of MT_COLOR_INDICATOR when
highlighting things in menu.c.
Combining attributes this way makes sense to me, because indicator is
meant to _indicate_, meaning increasing information density. It also
looks better when moving through the list containing differently
styled items.
This patch affects indicator, sidebar_highlight, and sidebar_indicator
color objects.
M color.c
M menu.c
M protos.h
M sidebar.c
2020-07-15 19:44:52 -0700 Kevin McCarthy (40ce20c8)
* Add first version of MuttLisp.
This is somewhat simple enhancement to muttrc processing, not a
full-blown embedded scripting language. There are no variables,
functions, types, or abilities that compete with macros.
MuttLisp can be invoked using the "run" command. If
$muttlisp_inline_eval is set, it can also be invoked with a bare
parenthesis expression as a command argument.
$muttlisp_inline_eval defaults unset, to avoid breaking existing
configurations, which might have bare parenthesis arguments for
regexps.
M Makefile.am
M doc/manual.xml.head
M init.c
M init.h
M keymap.c
M mutt.h
A mutt_lisp.c
A mutt_lisp.h
M muttlib.c
M po/POTFILES.in
2020-07-19 13:44:47 -0700 Kevin McCarthy (c5c448d4)
* Merge branch 'stable'
2020-07-14 14:42:30 -0700 Kevin McCarthy (c037a5a9)
* Add mutt_oauth2.py and README file.
This was contributed by Alexander Perlis to the mutt-dev mailing list.
mutt_oauth2.py.README contains the contents of his introduction email
for the script.
M contrib/Makefile.am
A contrib/mutt_oauth2.py
A contrib/mutt_oauth2.py.README
M doc/manual.xml.head
2020-07-13 12:43:39 -0700 Kevin McCarthy (35d63818)
* Improve GPGME inline processing.
Properly deal with multiple blocks:
* Extract the inline block to armored_data, instead of the entire
message. Use the code from pgp.c to find the block boundaries.
* The copy_clearsigned() function, probably originally ported from the
classic code too, assumes the first blank line starts the clear text.
But this is only true if the passed in data is just the block.
* Use the Charset armor header if found for encrypted content
conversion, as pgp.d does.
* Reset all states in each loop.
It looks like the pgp.c code wasn't resetting pgp_keyblock and
gpgcharset, so reset those in pgp.c too.
M crypt-gpgme.c
M pgp.c
2020-07-13 08:45:01 +0200 Olaf Hering (04490360)
* update de.po
Signed-off-by: Olaf Hering
M po/de.po
2020-07-12 10:50:51 -0700 Kevin McCarthy (69afc2f0)
* Try to improve "fcc to an imap mailbox" warning message.
Add a "Warning: prefix", and print a followup message indicating the
mailbox is being skipped.
Previously, the message might leave the impression the send was
aborted too.
M send.c
2020-07-11 12:59:26 -0700 Kevin McCarthy (0e826090)
* Merge branch 'stable'
2020-07-10 13:41:53 -0700 Kevin McCarthy (c94f43f9)
* Fix mutt_print_attachment_list() to unstuff format=flowed.
Several of the cases are covered by the pipe fixes in the last commit.
The last case is similar to the mutt_view_attachment case. Since the
send-mode file might be modified, change it to always save to a
tempfile and then unstuff the tempfile.
M attach.c
2020-07-09 19:31:48 -0700 Kevin McCarthy (b8217c6b)
* Fix mutt_pipe_attachment_list() to unstuff format=flowed.
The pipe routines are not used in other places (except printing
attachments - to be handled next), so just directly modify them.
Take care to space-stuff the filter output, since it replaces the
space-stuffed compose part.
M attach.c
M recvattach.c
2020-07-08 19:07:55 -0700 Kevin McCarthy (104768bf)
* Fix mutt_save_attachment_list() to unstuff format=flowed.
Directly modifying mutt_save_attachment was awkward for several
reasons. So instead implement a proxy function. It's ugly but
straight-forward.
M recvattach.c
2020-07-06 15:18:42 -0700 Kevin McCarthy (0600ebae)
* Fix mutt_view_attachment() to unstuff format=flowed.
This is used in receive mode for and .
I don't think those are accessible from the compose menu, but the code
allows it. So fix up the send-case too.
Since we may need to unstuff the send-mode file, make a copy instead
of symlinking. Also clean up the code to just sanitize the filename
regardless.
M attach.c
2020-07-06 10:39:06 -0700 Kevin McCarthy (e8436a01)
* Factor out stuff/unstuff routines for attachment handling.
M rfc3676.c
M rfc3676.h
2020-07-09 19:25:52 +0000 Emir (00838646)
* Update Turkish translations.
M po/tr.po
2020-07-08 12:59:36 -0700 Kevin McCarthy (b108d1b7)
* Merge branch 'stable'
2020-07-06 20:17:59 -0700 Kevin McCarthy (b48f8e1e)
* Directly add/remove mailboxes for IMAP.
The LSUB processor along with and would
construct a mailboxes command and send it to the muttrc parser.
Since the poll/label refactor it's much easier to just directly call
the functions, and is much more secure. So do that instead.
M buffy.c
M buffy.h
M imap/command.c
M imap/imap.c
2020-06-30 15:31:19 -0700 Kevin McCarthy (25f82cf6)
* Create $copy_decode_weed, $pipe_decode_weed, $print_decode_weed.
$weed covers a bit too much functionality. It makes sense for it to
control the display of messages in the pager, along with forwarding
and replying. But (in my opinion) piping and copy/save-decode are
something users might like to set differently without affecting the
rest of Mutt's behavior.
$copy_decode_weed affects the and
functions. I've defaulted this *unset*, despite that it breaks
backward compatibility, because I believe the header weeding is
surprising to users and loses important information when copy/saving a
message.
$pipe_decode_weed affects when $pipe_decode is set.
I've default this *set* for backward compatibility.
$print_decode_weed affects when $print_decode is set.
I've default this *set* for backward compatibility.
M commands.c
M doc/manual.xml.head
M init.h
M mutt.h
2020-06-27 15:04:45 -0700 Kevin McCarthy (3cacfb12)
* Consider $reply_to before $reply_self.
This is based on Derek Martin's patch in Trac ticket 2304.
I've modified it to:
* default to the value set by OPTREPLYSELF if the
$ignore_list_reply_to short-circuit fires.
* only apply the "from==reply_to" short circuit when $reply_self would
indicate replying to from.
* improve the "from==reply_to" short circuit to also fire when the
display names are equal.
The discussion in the trac ticket was somewhat contentious. This
isn't particularly an area in the code I'm eager to touch. However,
there are three points worth noting:
1) Mail-Followup-To is considered before $reply_self, so it makes some
sense for Reply-To to also be.
2) $reply_to is a quadoption, allowing the header to be declined and
then using $reply_self if appropriate.
3) $reply_self is a boolean. Previously, the only way to evaluate
reply_to in these cases is to set $reply_self. Most users probably
don't want that set by default, making that a poor workaround.
If this becomes sore point, I'll revert, but I don't believe the
change will be controversial in practice.
M send.c
2020-06-28 11:55:31 -0700 Kevin McCarthy (2311519a)
* Disable hook "command" argument expansion of config variables.
From my email to mutt-dev:
I've been thinking about this a *lot* more, and digging into past
commits. It looks like the escaping has pretty consistently been
intended for value completion and queries. See commits 3c00eb7a and
032dcbeb.
I don't believe the escaping is useful for variable expansion as
part of mutt_extract_token() - even for the hook "command" argument
case. The escaping is targeted only for an expression surrounded by
double quotes (as is generated by completion and queries). It
doesn't escape single quotes or spaces, for instance.
Thanks also to Oswald Buddenhagen and Cameron Simpson for their
feedback.
M hook.c
M init.c
2020-06-26 12:31:22 -0700 Kevin McCarthy (e53ae9ed)
* Escape config vars when expanding hook "command" argument.
The "command" parameter of folder, send, send2, account, reply, and
message hooks is run through mutt_extract_token() twice. Once when
parsing the hook command initially, and once via mutt_parse_rc_line()
when the actual hook fires.
In theory, Mutt users should be aware of this, and should place all
their command strings with configuration variables in single-quotes,
to delay expansion until execution.
But in reality Mutt has been escaping configuration variables as part
of mutt_extract_token() for many years.
The previous commit turned the escaping off, because it damages
"backup" assignments to a $my_ variable, and I believe is confusing
behavior.
But for hooks, users now expect the escaping behavior (for
double-quoted or unquoted command strings).
Note that mutt_parse_hook() also passed the MUTT_TOKEN_SPACE flag, but
neglected to do so for message-hook. Since all other "command"
argument hooks were included, and were added to the list over the
years, I believe this was an accidental oversight.
M hook.c
M init.c
M mutt.h
2020-06-25 15:13:48 -0700 Kevin McCarthy (e5a32a61)
* Convert var_to_string() to use a buffer, and disable auto-escaping.
Inside mutt_extract_token(), expanding a configuration variable used a
fixed size LONG_STRING. This would truncate long values, such as
oauth_refresh, which may have a large token embedded in them.
Convert var_to_string() to use a BUFFER parameter instead.
var_to_string() also performed escaping, which is not always helpful.
Disable the escaping inside. Add it back in
mutt_var_value_complete(), via the pretty_var() call.
The escaping may also be useful for hook "command" parameters. Since
Mutt has had the escaping behavior there for years, removing it will
likely break configurations. This will be added back in the next
commit.
M init.c
2020-06-24 18:44:18 -0700 Kevin McCarthy (110562b3)
* Add tab-completion menu for patterns.
This is based upon the patch from TAKAHASHI Tamotsu, maintained by
Vincent Lefèvre.
The push/pop menu operations take care of most redraw operations, so
remove those from the patch. Change post-search to '|= REDRAW_MOTION'
instead of assigning so it doesn't overwrite a full redraw.
Add L10N for the new help strings.
Add a format string and configuration variable for the menu.
Add the multi-character thread patterns.
Doing the above required some structural changes to the menu, so I
rewrote it in a more common style used elsewhere in Mutt.
M curs_main.c
M enter.c
M globals.h
M init.h
M pattern.c
M protos.h
2020-06-24 02:03:40 +0200 Vincent Lefevre (bb407ec3)
* Add a missing comma in the manual.
M doc/manual.xml.head
2020-06-24 01:48:29 +0200 Vincent Lefevre (5b3b7a80)
* Correct wording in a comment.
M mutt_tunnel.c
2020-06-22 18:31:47 -0700 Kevin McCarthy (c9b58e3d)
* Add $tunnel_is_secure config, defaulting set.
The config variable is to resolve an ambiguity in Mutt about whether
using $tunnel is secure.
On the one hand, the examples in the manual show using ssh or a direct
pipe to a program. Many users do this to connect to an IMAP server
with PREAUTH configured, relying on the tunnel to be secured by ssh or
by the fact that it's a local pipe.
On the other hand, the Mutt connection code still respects
$ssl_starttls and $ssl_force_tls, as if the $tunnel connection were
not already secured.
After some discussion on mutt-dev, it seemed the best idea to assume
the connection is secure by default, in order to not break IMAP
PREAUTH connections, but to provide a configuration variable in case
there are situations where it is not.
Thanks to Aaron Schrab for the original idea of setting conn->ssf for
$tunnel in his patch to ticket 250.
M doc/manual.xml.head
M imap/imap.c
M init.h
M mutt.h
M mutt_socket.h
M mutt_tunnel.c
2020-06-23 10:45:01 -0700 Kevin McCarthy (0e428dd5)
* Merge branch 'stable'
2020-06-22 12:46:08 -0700 Kevin McCarthy (22d7a1b3)
* Merge branch 'stable'
2020-06-21 15:17:04 +0200 Vincent Lefevre (e807360d)
* Updated French translation.
M po/fr.po
2020-06-20 14:12:43 -0700 Kevin McCarthy (43184600)
* Improve warning message for possible STARTTLS response injection.
Thanks to Vincent Lefèvre for the improved wording, and to Cameron
Simpson for his suggestions too.
M mutt_ssl.c
M mutt_ssl_gnutls.c
2020-06-20 22:30:20 +0200 Olaf Hering (db656f46)
* Update de.po
Signed-off-by: Olaf Hering
M po/de.po
2020-06-20 06:55:47 -0700 Kevin McCarthy (4adb274a)
* Merge branch 'stable'
2020-06-19 15:00:53 -0700 Kevin McCarthy (73b70169)
* Add recommendation to use $ssl_force_tls.
M init.h
2020-06-19 14:39:33 -0700 Kevin McCarthy (a400fc3f)
* Add L10N comment for unencrypted PREAUTH warning.
I forgot to add one while trying to get the fix out.
M imap/imap.c
2020-06-18 14:18:48 -0700 Kevin McCarthy (fee31356)
* Merge branch 'stable'
2020-06-15 12:27:24 +0200 Vincent Lefevre (75be2b07)
* Updated French translation.
M po/fr.po
2020-06-14 14:18:49 -0700 Kevin McCarthy (64c1d9ac)
* Merge branch 'stable'
2020-06-11 16:11:36 -0700 Kevin McCarthy (c7a872d1)
* Add basic XOAUTH2 support.
This still relies on an external script to obtain the resource access
token. Since XOAUTH2 should be slowly going away, use the same
refresh_commands as with OAUTHBEARER.
Unlike OAUTHBEARER, XOAUTH2 must be explicitly added to the
$imap/smtp/pop_authenticators list.
To keep the shared functions simpler, convert them to use buffers.
RFC 7628 indicates that upon authentication failure the clients should
be sending an BASE 64 encoded '^a' ("AQ=="), to terminate the SASL
session, so change all the handlers to do that and read the following
response. The RFC doesn't comment about a line terminator being
required, but I assume it is, so add that too.
M account.c
M account.h
M doc/manual.xml.head
M imap/auth.c
M imap/auth.h
M imap/auth_oauth.c
M imap/command.c
M imap/imap_private.h
M pop_auth.c
M smtp.c
2020-06-08 12:57:43 -0700 Kevin McCarthy (5b844328)
* Perform small cleanups in gnutls.c.
These are some cleanups I started making during the stable branch
fixes committed yesterday. It seemed a better idea to perform this
amount of change in master rather than in stable, so they are in this
separate commit.
Fix the function-call spacing style.
Change the #define CERTERR values to match the pattern used elsewhere
in mutt.
Use safe_calloc() instead of calloc.
Add a few error checks in tls_get_client_cert(), and change to
directly extract the CN using gnutls_x509_crt_get_dn_by_oid().
Use a do-while around the gnutls_handshake() instead of repeating it.
M mutt_ssl_gnutls.c
2020-06-08 23:24:18 +0200 Vincent Lefevre (4938c97a)
* Updated French translation.
M po/fr.po
2020-06-07 12:30:01 -0700 Kevin McCarthy (8799adcf)
* Merge branch 'stable'
2020-06-06 16:17:07 +0200 Vincent Lefevre (dcc59c63)
* French translation: correction and improvement.
M po/fr.po
2020-06-06 15:57:03 +0200 Vincent Lefevre (d4cb383e)
* Updated French translation.
Note: I initially translated "forward" as "faire suivre", but this is
no longer possible now that the past participle "forwarded" is used.
The translated term is now "transférer" (like in Thunderbird).
M po/fr.po
2020-06-02 15:59:12 -0700 Kevin McCarthy (14709d7d)
* Change postpone mode to write Date header too.
This will ensure that postponed protected headers will match, in case
Mutt later adds checks about that.
I can't see any reason to not include it for the two other
cases (searching and edit draft).
M sendlib.c
2020-06-02 15:48:59 -0700 Kevin McCarthy (82b5c697)
* Add other headers to written Protected Headers.
In order to avoid $edit_headers, , or other places
setting env->date and inadvertantly generating a wrong Date header,
store the "protected headers" generated date header in the sctx, in
addition to mime_headers.
I initially removed env->date setting from mutt_parse_rfc822_line(),
thinking it a more elegant solution, but unfortunately protected
headers reading needs it to be there. We don't currently print or
compare the headers, but might wish to in the future.
Explicitly pass in a date parameters to mutt_write_rfc822_header() to
make each caller conscious of where the date is coming from.
M crypt.c
M headers.c
M main.c
M pattern.c
M protos.h
M send.c
M send.h
M sendlib.c
2020-06-01 18:48:41 -0700 Kevin McCarthy (b0faa8d2)
* Modify mutt_make_date() to append to a buffer.
Remove the "Date: " prefix and the trailing newline, and convert
callers to add those.
This will be needed for protected headers Date support, where we will
have to generate and store the date inside mutt_protect().
M copy.c
M protos.h
M sendlib.c
2020-06-01 13:50:41 -0700 Kevin McCarthy (82e80422)
* Move protected-headers parameter cleanup out to send.c cleanup.
Remove it in the same place that the mime-headers are removed. This
is so that the parameter doesn't stay in there if we loop back to the
compose menu on failure and the sender toggle encrption off.
M crypt.c
M send.c
2020-06-01 13:40:25 -0700 Kevin McCarthy (f07fa1a1)
* Update the terminology and link for protected headers.
M init.h
2020-06-01 13:27:38 -0700 Kevin McCarthy (e7ed170c)
* Change default of $crypt_protected_headers_subject to "...".
The most recent protected headers draft now specifies the obscured
subject be "...". See https://github.com/autocrypt/protected-headers
and
https://datatracker.ietf.org/doc/draft-autocrypt-lamps-protected-headers/
M init.h
2020-06-03 12:34:51 -0700 Kevin McCarthy (c5441904)
* Add check for sqlite3_prepare_v3() in configure.ac
I didn't realize the function was added so recently: sqlite
3.20.0 (2017-08-01).
Thanks to isdtor for pointing out the problem.
M configure.ac
2020-05-29 10:54:56 -0700 Kevin McCarthy (44711a24)
* Try to automatically reconnect to an open IMAP mailbox on error.
For the Context, change cmd_handle_fatal() to flag idata->status on
the error, but not close the mailbox and disconnect.
Then have imap_check_mailbox() reconnect when IMAP_REOPEN_ALLOW is
set (so the index is prepared to rebuild with all new headers). Try
to merge flag, envelope, and attach_del changes back in. Replace the
existing Context pointer with the new values, and swap out the idata
passed in on success, so we don't continue to work with the wrong
idata.
The imap_check_mailbox() changes are purposely coded loosely to try
and catch all sorts of errors - not just polling and imap_idle
errors.
Create a new check_mailbox() return code to indicate a reconnect.
Display a more appropriate message in the index, but otherwise treat
it the same as a REOPEN.
M curs_main.c
M imap/command.c
M imap/imap.c
M imap/imap_private.h
M mailbox.h
M mutt.h
2020-05-31 09:09:14 +0200 Grzegorz Szymaszek (878e261a)
* Improve Polish translation of “undeleted” messages
M po/pl.po
2020-05-29 15:46:29 -0400 Remco Rijnders (4091cbcd)
* Remove support for OpenSSL <0.9.5
The current code provides a workaround for the absence of RAND_status in
OpenSSL versions before 0.9.5. The comments in the code indicate these
versions have to be supported, but as these versions are now more than
20 years old, this no longer applies. Removing this support simplifies the
code and on the small chance that anyone is still using such old versions,
breaking their build will actually be doing them a favor given the known
issues with these OpenSSL versions.
M configure.ac
M mutt_ssl.c
2020-05-27 15:25:13 -0700 Kevin McCarthy (e2d27aac)
* Add myself to the contributor list in the manual.
I didn't realize previous maintainers were in the list, so I guess I
should add myself. :-)
M doc/manual.xml.tail
2020-05-27 14:51:04 -0700 Kevin McCarthy (9723c62c)
* Allow tagging string config vars as L10N.
Translate them once in mutt_set_default().
Modify makedoc.pl to remove the N_() tag, and to also note the type as
"string (localized)".
The makedoc.pl already will ignore L10N comment so add a basic comment
above each config var in init.h.
I'm not tagging $crypt_protected_headers_subject because the spec now
says this should be "...". I plan on implementing that change along
with other fixes before the next release.
M doc/makedoc.pl
M init.c
M init.h
M po/POTFILES.in
2020-05-27 14:39:51 -0700 Kevin McCarthy (4aedee85)
* Merge branch 'stable'
2020-05-26 13:57:18 -0700 Kevin McCarthy (d3fd97b4)
* Merge branch 'stable'
2020-05-25 13:38:46 -0700 Kevin McCarthy (52f5694b)
* Merge branch 'stable'
2020-05-25 12:59:50 -0700 Kevin McCarthy (cb9c683d)
* Merge branch 'stable'
2020-05-24 13:34:33 -0700 Kevin McCarthy (cd3b529f)
* Merge branch 'stable'
2020-05-23 14:41:50 -0700 Kevin McCarthy (0ac6b609)
* Refactor mutt_buffer_strip_formatting() inside of pager.c.
Rather than duplicate the logic, move the function back inside
pager.c. Add a parameter to optionally remove attachment markers.
Inside fill_buffer(), use a stack-based buffer and directly assign the
buffer.data back to fmt, to keep it as fast as before.
M handler.c
M muttlib.c
M pager.c
M pager.h
M protos.h
2020-05-22 16:13:38 -0700 Kevin McCarthy (c44f0ce5)
* Add cd command to change the current working directory.
Original Author: Christoph Berg
Use cases are saving attachments to some other place and others.
Modified by Kevin McCarthy:
- Fix up documentation to use cmdsynopsis/command/arg.
- Add muttrc.man entry.
- Fix compilation error.
- Clean up the parse_cd() function to use buffers and remove new
translation strings.
M doc/manual.xml.head
M doc/muttrc.man.head
M init.c
M init.h
2020-05-19 13:14:48 -0700 Kevin McCarthy (b3ab8ca9)
* Merge branch 'stable'
2020-05-18 05:35:32 +0200 Cyril Roelandt (5bbc6be5)
* Add default keybindings for common operations.
This adds the following bindinds:
- ^P for "history up"
- ^N for "history down"
M functions.h
2020-05-17 13:01:35 -0700 Kevin McCarthy (6150d285)
* Clean up Editor Menu interface inside Mutt.
enter_filename() api:
* Rename enter_fname() to _enter_fname()
* Rename mutt_buffer_enter_mailbox() to mutt_enter_mailbox()
* Rename mutt_buffer_enter_filenames() to mutt_enter_filenames()
* Remove 'multiple' and 'fname buffer' parameters from
mutt_enter_filenames(). Create a temp buffer, and enable multiple
automatically. This removes the possibility of mistakenly looking
at an unpopulated 'fname' parameter for the result.
* Create mutt_enter_filename(). This isn't used currently, but it
makes sense to have a "single filename" api already set up with
correct parameters.
get_field() api:
* Rename _mutt_buffer_get_field() to a static _get_field() function.
* Remove _mutt_get_field() and make mutt_get_field() an actual
function.
* Remove the multiple/files/numfiles parameters, since this
should only be done through mutt_enter_filenames().
_mutt_enter_string() api:
Add comments to mutt_enter_string() and _mutt_enter_string() headers.
Noting that they should generally not be called directly. List the
functions that should be used inside Mutt. Note the behavior of the
multiple parameter.
M autocrypt/autocrypt.c
M commands.c
M compose.c
M curs_lib.c
M curs_main.c
M enter.c
M protos.h
M send.c
2020-05-17 12:37:24 -0700 Kevin McCarthy (95cf171e)
* Fix attach-file tag operation to work when quitting browser.
There was a disconnect between the compose menu and the behavior when
selecting multiple files. The select_file() operation populates the
files and numfiles when hitting 'quit'. The compose menu was
incorrectly looking at the fname parameter, which won't be populated
unless the user presses enter.
The enter_string() menu makes the same mistake. Change it to behave
differently depending on whether multiple is set.
The logic needs to be made clearer by changing the api. I will do
that in the next commit.
M browser.c
M compose.c
M enter.c
2020-05-16 14:39:00 -0700 Kevin McCarthy (1f20684a)
* Merge branch 'stable'
2020-05-16 11:15:37 -0700 Kevin McCarthy (a0bbc02e)
* Merge branch 'stable'
2020-05-16 14:41:58 +0800 Martin Michlmayr (5d4ec51a)
* Fix typos
M doc/makedoc.pl
M doc/manual.xml.head
M doc/muttrc.man.head
2020-05-15 18:23:32 -0400 Remco Rijnders (a8824b55)
* Fix sidebar_sort_method usage in documentation
M init.h
2020-05-15 09:27:31 +0200 Vincent Lefevre (e78d2b4f)
* Updated French translation.
M po/fr.po
2020-05-14 15:42:38 -0700 Kevin McCarthy (d1199f76)
* Add Autocrypt error messages when trying to force send.
If the account doesn't exist, or isn't enabled, display a message back
in the compose menu.
M autocrypt/autocrypt.c
2020-05-14 14:02:20 -0700 Kevin McCarthy (e3e0a0ba)
* Change mailto handling to parse in-reply-to loosely.
To fix ticket 241 without loosening in-reply-to email header parsing,
add a direct call to mutt_parse_references(). Change the second
parameter, which was not used, to indicate "allow nonbracketed" mode.
M parse.c
M protos.h
M url.c
2020-05-14 12:01:21 -0700 Kevin McCarthy (194a60ae)
* Adjust message-id extraction to retry more permissively on failure.
It turns out some illegal message-id headers have neither angle
brackets nor even an '@' in them. So adopt a different approach.
First try parsing the Message-ID header in a strict(ish) mode
requiring angle brackets. If that returns nothing, then try again in
loose mode, which will basically grab the first non-comment token.
Change References and In-Reply-To to keep the strict(ish) mode. We
might loose some parent references, but that won't cause much loss of
functionality in Mutt, and also helps prevents parsing in "garbage"
and passing that back in replies.
Thanks to Oswald Buddenhagen for the suggested approach.
M parse.c
M protos.h
M send.c
2020-05-13 15:52:30 -0700 Kevin McCarthy (99ba609f)
* Loosen message-id parser to parse ids without angle brackets.
To do this, we need to enhance the loop to skip over comments. It
does this using the rfc822 address parse_comment() routine.
Keep parsing inside angle brackets non-strict, because of the history
of trac tickets (1116, 1935, 3090) demonstrating message-ids with
spaces inside the angle brackets, double-'@'s, or even completely
missing '@'.
However, in order to reduce the risk of extracting garbage, insist
that message-id's outside of angle brackets at least have an '@'.
Automatically add missing angle brackets so that Mutt will generate
valid references/in-reply-to headers.
M parse.c
M rfc822.c
M rfc822.h
2020-05-13 19:55:48 +0200 Vincent Lefevre (ba2430af)
* Update section "Terminal Keybindings" in the manual.
Also list ^Q, ^S and ^Z.
Note that not all bindings output by "stty -a" affect Mutt.
M doc/manual.xml.head
2020-05-12 20:05:12 -0700 Kevin McCarthy (48a0e90e)
* Add a brief section on stty to the manual.
Place it at the bottom of the keybindings section. It's by no means
complete, but may be helpful to point some beginners in the right
direction.
M doc/manual.xml.head
2020-05-11 14:15:27 -0700 Kevin McCarthy (b0570d76)
* Separate mailbox vs buffy-incoming for the "enter_fname" functions.
Create mutt_buffer_enter_mailbox(), which turns on the MUTT_MAILBOX
flag (renamed from MUTT_EFILE). This uses the mailbox history file
category, and turns on MUTT_SEL_FOLDER in the select_file browser.
Add a do_incoming paramter and a new separate flag MUTT_INCOMING, so
that buffy-completion can be enabled optionally for mailboxes. It
only makes sense for the change-folder operation, so disable it in
other contexts (such as saving/copying a message, prompting for an Fcc
mailbox, or an autocrypt scan mailbox).
Change saving/copying a message to use the mailbox function, so that
it shares the history file with other mailbox operations. It would
previously use the "file" history category.
Create mutt_buffer_enter_filenames() to replace the enter_fname()
prompt for files.
Since nothing directly uses the _mutt_buffer_enter_fname() function
anymore, rename and make it static.
M autocrypt/autocrypt.c
M commands.c
M compose.c
M curs_lib.c
M curs_main.c
M enter.c
M mutt.h
M protos.h
M send.c
2020-05-11 13:12:23 -0700 Kevin McCarthy (d338650f)
* Remove unused mutt_enter_fname() functions.
All the callers use the buffer version.
M curs_lib.c
M protos.h
2020-05-09 18:43:55 -0700 Kevin McCarthy (64d67279)
* Merge branch 'stable'
2020-05-09 21:21:20 +0200 Vincent Lefevre (254e5eb5)
* Updated French translation.
M po/fr.po
2020-05-08 15:57:18 -0700 Kevin McCarthy (78ab1cca)
* Add utime() loop for _maildir_commit_message().
While fixing up mutt_decrease_mtime(), I thought about creating a
helper function to loop for EINTR. However, there is only one other
place in the code that cares enough about utime succeeding to check
the return value.
maildir_commit_message() is used fairly often, so it's possible I
might be writing useless error handling. However there are
reports (easily found by internet search) of some systems returning
EINTR, so might as well fix this place too.
M mh.c
2020-05-08 13:47:41 -0700 Kevin McCarthy (1aa61c2e)
* Improve mutt_decrease_mtime() error handling.
Add a check for utime failure. Some non-posix implementations return
EINTR. Since editing messages is a very common operation in Mutt,
take care to retry in that case so as not to introduce message
composition failures.
Make sure all callers check mutt_decrease_mtime()'s return value,
since stat could also fail.
M editmsg.c
M headers.c
M muttlib.c
M send.c
2020-05-08 13:05:44 -0700 Kevin McCarthy (08294769)
* Merge branch 'stable'
2020-05-07 19:41:43 -0400 Remco Rijnders (4717b041)
* Remove commented out mutt_message_hook calls
M recvcmd.c
2020-05-07 19:17:42 -0400 Remco Rijnders (ad5dcf17)
* Clarify CH_WEED debug message
M copy.c
2020-05-07 16:32:46 -0700 Kevin McCarthy (a9fdeeab)
* Fix sed invocation in gen_defs to work with non-GNU sed.
\n isn't supported on FreeBSD's version, so instead directly embed the
newline.
M gen_defs
2020-05-07 16:09:24 -0700 Kevin McCarthy (3745c59e)
* Turn off check_sec.sh warning.
The script is stupid, but still useful, so just add the tag to turn
off the warning. :-)
M signal.c
2020-05-07 16:01:46 -0700 Kevin McCarthy (b70676a2)
* Add L10N strings to exit_handler().
Since gettext() is not safe in the handler, cache them in advance.
M main.c
M protos.h
M signal.c
2020-05-07 15:26:52 -0700 Kevin McCarthy (6c21b111)
* Merge branch 'stable'
2020-05-07 15:10:44 +0200 Vincent Lefevre (52751db2)
* exit_handler: add explanation about the missing l10n translations.
Details in commit 5835f4c0594ff94a2aa62ad122c4dc2871ee8492.
M signal.c
2020-04-25 18:39:38 -0700 Kevin McCarthy (27439a7e)
* Support domain-literal values in email adddresses.
RFC 5322 defines the domain-literal:
domain = dot-atom / domain-literal / obs-domain
domain-literal = [CFWS] "[" *([FWS] dtext) [FWS] "]" [CFWS]
dtext = %d33-90 / ; Printable US-ASCII
%d94-126 / ; characters not including
obs-dtext ; "[", "]", or "\"
RFC 5321 defines possible literal values in section 4.1.3.
I think this was unnoticed (and possibly unimplemented) because it's a
rare use-case. However, Stuart Gathman @sdgathman in the
ticket (#226) noted:
The use case is replacing alpine with mutt for this fully
decentralized usage:
https://fedoramagazine.org/decentralize-common-fedora-apps-cjdns/
The only requirements are to support IPv6 and IPv6 literals (to
avoid use of DNS).
Adding this slightly complicates the parser, because we can't wait to
decide whether it's a domain-literal until parse_mailboxdomain(), but
we also need to accommodate the CFWS.
Break out a parse_domain() function, with partial duplication of
parse_mailboxdomain()'s logic.
Also, add parse_literal() to the outer parser, because we need to
handle both name-addr and top level addr-spec.
The RFC822Errors[] are currently unused (and broken), but add a new
error code and message, in case those get fixed up and introduced to
the code in the future.
Disable IDNA encoding for domain-literals.
M mutt_idna.c
M rfc822.c
M rfc822.h
2020-05-06 21:54:02 +0200 Grzegorz Szymaszek (a48328bd)
* Fix the Polish translation of a message
The “delete the current entry” message refers to any entries/items
(“element”), not just to letters/messages (“list”).
M po/pl.po
2020-04-25 01:28:30 +0900 Tamotsu TAKAHASHI (f11a6b1f)
* L10N: add more helpful comments on OPS entries.
Now translators can see exactly where the entries came from.
Also, OPS.* can have L10N comments. Add "/* L10N: blabla */" to OPS files.
This commit would help translators a lot.
For example, the msgid "delete the current account" once surprised me.
I said, "What? Can mutt delete my account???"
With this commit, I can see that "the account" is clearly the "autocrypt" account.
```
#. L10N: Help screen description for OP_AUTOCRYPT_DELETE_ACCT
#. #: ../OPS:8
#: ../keymap_alldefs.h:29
msgid "delete the current account"
```
M doc/gen-map-doc
M gen_defs
2020-05-05 13:32:59 +0200 Vincent Lefevre (3b2c4b7c)
* Updated French translation.
M po/fr.po
2020-04-19 16:28:02 -0400 Remco Rijnders (c5b8dba7)
* olen length requirement was too conservative
When converting binary values to Base 64 encoding, we go from a 8 bit
representation to a 6 bit representation. As a result, we need extra space
to store the result compared to the input value being converted.
In the current implementation, the requirements on the size of the output
buffer are too conservative; We only have to check if the remaining space
is larger than 4 (3 input bytes get converted into 4 Base64 encoded output
bytes, plus allowing for a terminating '\0' character) rather than 10.
M base64.c
2020-05-04 11:06:23 -0700 Kevin McCarthy (e6679dfa)
* Change pattern "operator" to "modifier" in doc and L10N strings.
Thanks to TAKAHASHI Tamotsu for pointing out the inconsistency.
M UPDATING
M doc/muttrc.man.head
M init.h
M pattern.c
2020-05-03 17:29:55 +0200 Vincent Lefevre (f5a8419d)
* Add missing blank lines in ChangeLog due to buggy update-changelog.
M ChangeLog
2020-05-03 17:18:28 +0200 Vincent Lefevre (986f6263)
* Fix update-changelog rule (missing blank line in generated ChangeLog).
M Makefile.am
2020-08-29 12:25:28 -0700 Kevin McCarthy (5fb17352)
* Update UPDATING file for 1.14.7.
M UPDATING
2020-08-27 15:03:08 -0700 Kevin McCarthy