automatic post-release commit for mutt-2.2.0mutt-2-2-rel

32 files changed, 19389 insertions, 15544 deletions

diff --git a/ChangeLog b/ChangeLog
index cd16bb5c..27d3fc70 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,3 +1,1747 @@
+2022-02-12 10:49:21 -0800  Kevin McCarthy  <kevin@8t8.us> (a5175478)
+
+        * Set release date for 2.2.0 in UPDATING file.
+
+M	UPDATING
+
+2022-02-11 13:37:27 +0100  Grzegorz Szymaszek  <gszymaszek@short.pl> (cc7578a1)
+
+        * Update the Polish translation for Mutt 2.2.0
+
+M	po/pl.po
+
+2022-02-09 22:58:23 +0300  Emir SARI  <emir_sari@icloud.com> (1c47970f)
+
+        * Improve Turkish translations
+
+M	po/tr.po
+
+2022-02-06 14:53:01 -0800  Kevin McCarthy  <kevin@8t8.us> (70958893)
+
+        * Fix mutt_paddstr() to properly filter unprintable chars.
+        
+        The original version of this function had no "replacement character"
+        functionality, so it simply directly called addnstr() to display the
+        characters if everything was okay.
+        
+        Commit a080fd35 added replacement logic, similar to
+        mutt_format_string(), but forgot to change addnstr to use the
+        replacement character.
+        
+        This means garbage characters could goof up the mutt display, for
+        things such as the subject in the compose menu.
+        
+        We could add calls to wcrtomb() like in mutt_format_string(), but
+        mutt_addwch() already does this, and properly calls wcrtomb() a second
+        time to add a shift sequence if needed.
+
+M	curs_lib.c
+
+2022-02-05 14:01:39 -0800  Kevin McCarthy  <kevin@8t8.us> (fe3dd705)
+
+        * Change mailto_allow to be exact match only.
+        
+        The code was previously reusing mutt_matches_ignore(), but that allows
+        prefixes.  For mailto accepted headers we should be more picky.
+
+M	url.c
+
+2022-02-05 13:54:19 -0800  Kevin McCarthy  <kevin@8t8.us> (5c3c6e52)
+
+        * Filter headers passed via the command line.
+        
+        The values passed via '-s' and mailto urls were not sufficiently
+        sanitized, allowing an embedded newline that could be used to inject a
+        header.
+
+M	main.c
+M	parse.c
+M	protos.h
+M	url.c
+
+2022-02-05 09:22:44 -0800  Kevin McCarthy  <kevin@8t8.us> (d3b4ff88)
+
+        * Fix mbox.man asctime(3) reference.
+        
+        The section number should not be part of the BR text.
+        
+        Thanks to @hmartink and the manpage-l10n project for the bug report!
+
+M	doc/mbox.man
+
+2022-02-04 18:33:34 -0500  Aaron Schrab  <aaron@schrab.com> (7d2a53ee)
+
+        * Clarify description of $local_date_header
+        
+        When I reexamined the documentation of the $local_date_header option, I
+        wasn't very confident about what it actually does. Much of what I did
+        get from it came from a vague recollection of the discussion around when
+        it was added, and I referred back to the commit which added this to
+        solidify that understanding.
+        
+        Viewing this as a user (without looking at the internal implementation),
+        I don't think the date is being converted, to me that implies that it
+        was received from somewhere else in some other format; rather I'd view
+        it as being initially generated in the local timezone.
+        
+        The reference to the "sender's timezone" makes me think that this would
+        be operating on dates in messages received from other people; I've
+        changed that to "your" to help clarify that it's about messages created
+        locally.
+        
+        Finally, add information about what is done if this option is unset,
+        along with some reasoning on why someone might want to unset this.
+
+M	init.h
+
+2022-02-02 12:51:08 -0800  Kevin McCarthy  <kevin@8t8.us> (c2ed716b)
+
+        * Don't queue IMAP close commands.
+        
+        Commit e7df4d5a changed the close commands to queue up, on the theory
+        that it will always be followed by a SELECT or a LOGOUT.
+        
+        However, performing a <recall-message> while in the inbox will open a
+        *new* connection.  mutt_conn_find() prepends the new conection to the
+        Connections list.  The next mailbox opened will use that *new*
+        connection.
+        
+        The effect will be the queued CLOSE won't occur before that mailbox is
+        opened, and so "\Deleted" messages won't be expunged on the server
+        right away.
+        
+        If the user had $move set and reopened their $spoolfile, they would
+        find the read messages marked for delete but not yet expunged from the
+        server.
+
+M	imap/imap.c
+
+2022-02-02 06:47:45 -0800  Kevin McCarthy  <kevin@8t8.us> (92686e5d)
+
+        * Update Finnish translation.
+
+M	po/fi.po
+
+2022-01-31 06:31:43 +0100  Flammie Pirinen  <flammie@iki.fi> (b485d04f)
+
+        * update Finnish translations
+
+M	po/fi.po
+
+2022-01-30 10:15:20 -0800  Ivan Vilata i Balaguer  <ivan@selidor.net> (553f016e)
+
+        * Updated Catalan translation.
+
+M	po/ca.po
+
+2022-01-30 09:21:58 -0800  Vsevolod Volkov  <vvv@mutt.org.ua> (c7053420)
+
+        * Updated Russian translation.
+
+M	po/ru.po
+
+2022-01-30 09:19:01 -0800  Vsevolod Volkov  <vvv@mutt.org.ua> (db9ac4ad)
+
+        * Updated Ukrainian translation.
+
+M	po/uk.po
+
+2022-01-29 14:24:52 -0800  Kevin McCarthy  <kevin@8t8.us> (8a6472f0)
+
+        * Fix argc==0 handling.
+        
+        At least on Debian Linux, after invoking getopt() with argc==0, optind
+        seems to remain at the default value of 1.
+        
+        mutt_dotlock was checking for optind==argc to report a missing
+        parameter, and would end up invoking the function with the first
+        environment variable instead.
+        
+        mutt, pgpewrap, and mutt_pgpring were assuming argv[0] was non-null,
+        so just add a check and hardcode the value in that case to avoid a
+        possible segv.
+        
+        mutt_pgpring allows no arguments.  Its parameters to
+        pgpring_find_candidates() would pass the environment list and a
+        negative nhints in that case.  The code seem to handle the case, but
+        set optind=argc to be clear about it.
+
+M	dotlock.c
+M	main.c
+M	pgpewrap.c
+M	pgppubring.c
+
+2022-01-29 11:46:56 -0800  Helge Kreutzmann  <debian@helgefjell.de> (46ff6ccb)
+
+        * Update de.po.
+
+M	po/de.po
+
+2022-01-29 09:34:04 -0800  Helge Kreutzmann  <debian@helgefjell.de> (17452c87)
+
+        * Update de.po.
+
+M	po/de.po
+
+2022-01-28 12:38:00 -0800  Emir SARI  <emir_sari@icloud.com> (7ae08fd5)
+
+        * Update Turkish translations.
+
+M	po/tr.po
+
+2022-01-28 20:15:02 +0100  Petr Písař  <petr.pisar@atlas.cz> (df4eb293)
+
+        * Update Czech translation for mutt-2.2.0
+
+M	po/cs.po
+
+2022-01-28 09:34:39 -0800  Kevin McCarthy  <kevin@8t8.us> (0864ef75)
+
+        * Update translation files copyright string.
+        
+        Thanks to Vincent Lefèvre for reminding me to do this too.
+
+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/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
+
+2022-01-28 15:11:18 +0100  Vincent Lefevre  <vincent@vinc17.net> (7f5ff5f2)
+
+        * Updated French translation.
+
+M	po/fr.po
+
+2022-01-27 15:35:30 -0800  Kevin McCarthy  <kevin@8t8.us> (821803cd)
+
+        * Fix $status_format documentation of $sort_thread_groups expando.
+
+M	init.h
+
+2022-01-27 14:29:35 -0800  Kevin McCarthy  <kevin@8t8.us> (890eb88c)
+
+        * Initial draft of UPDATING file changes for 2.2.0.
+
+M	UPDATING
+
+2022-01-27 13:04:37 -0800  Kevin McCarthy  <kevin@8t8.us> (23ef6eb1)
+
+        * Update source file copyright lines based on commit history.
+        
+        This is based on the copyright-updater script.  That script only
+        updates existing copyright lines in the headers.  Other copyright is
+        stored in the commit history.
+
+M	background.c
+M	buffer.c
+M	buffer.h
+M	buffy.c
+M	hcachever.pl
+M	send.h
+M	sidebar.c
+
+2022-01-27 12:18:57 -0800  Kevin McCarthy  <kevin@8t8.us> (ed65fe18)
+
+        * Update manual and program copyright notices.
+
+M	COPYRIGHT
+M	doc/manual.xml.head
+M	doc/mutt.man
+M	main.c
+
+2022-01-27 14:18:06 +0100  Vincent Lefevre  <vincent@vinc17.net> (4f9eac94)
+
+        * Updated French translation.
+
+M	po/fr.po
+
+2021-12-31 18:30:17 -0800  Kevin McCarthy  <kevin@8t8.us> (2e8b6fea)
+
+        * Add very basic cd path completion.
+        
+        This uses existing code to partially complete, and then launch the
+        browser for selection.
+        
+        Unfortunately, completing a path argument brings up other issues, such
+        as quoting, which this currently doesn't handle.  It also might be
+        better to rotate through completions rather than launching the file
+        brower.
+        
+        Add a "select directory" mode to the browser, modeled on the IMAP
+        "enter vs select a mailbox with children" behavior: using <view-file>
+        to select the directory and <select-entry> to enter the directory.
+
+M	browser.c
+M	enter.c
+M	init.c
+M	mutt.h
+
+2021-12-31 15:24:01 -0800  Kevin McCarthy  <kevin@8t8.us> (2362c114)
+
+        * Simplify the mutt_select_file() call for MUTT_CMD completion.
+        
+        Flags never has both MUTT_CMD and MUTT_MAILBOX, so there is no need
+        for the test.
+
+M	enter.c
+
+2021-12-31 15:10:25 -0800  Kevin McCarthy  <kevin@8t8.us> (bc01beeb)
+
+        * Add memcpy on MUTT_CMD completion to invoke the file selector.
+        
+        It looks like commit 844a133f (21 years ago) accidentally removed the
+        memcpy while refactoring, but without it the mutt_select_file() won't
+        be called on the second tab keypress.
+
+M	enter.c
+
+2022-01-22 14:10:24 -0800  Kevin McCarthy  <kevin@8t8.us> (05b0dd2c)
+
+        * Enable $rfc2047_parameters by default.
+        
+        20+ years later, Mutt still gets bug reports about attachment names in
+        2047 encoded form.
+        
+        Although there is a tiny chance an attachment could legimately want to
+        be named like that, it's far far more likely the sending MUA has
+        incorrectly encoded the value.
+        
+        When a user has this problem, the "solution" is also difficult to
+        find, unless they are initimately familiar with RFC2047 vs 2231.  So I
+        think users would benefit much more from this being set by default.
+
+M	init.h
+
+2022-01-17 15:36:55 -0800  Kevin McCarthy  <kevin@8t8.us> (4015c9a0)
+
+        * Set environ for sendmail execvp() call.
+        
+        The code previously tried to use execvpe(), but that's a glibc
+        extension.  However, it's not hard to manually set environ before an
+        execvp() call to accomplish the same thing.
+
+M	sendlib.c
+
+2022-01-17 09:54:11 +0100  Vincent Lefevre  <vincent@vinc17.net> (fcd3d956)
+
+        * Updated French translation.
+
+M	po/fr.po
+
+2022-01-14 15:17:38 -0800  Kevin McCarthy  <kevin@8t8.us> (2387728b)
+
+        * Convert SMTP cyrus sasl to use buffers.
+        
+        Use the smtp_get_auth_response() added for gsasl, to allow
+        multi-response lines and arbitrary length server response lines.  The
+        rfc notes that SASL lines are not constrained by the SMTP line length
+        limits.
+        
+        Since everything was a bit tangled together before, change the output
+        and base64 conversion to use buffers too.
+
+M	smtp.c
+
+2022-01-15 12:31:16 -0800  Robert Bartel  <r.bartel@gmx.net> (44269756)
+
+        * Move hard_redraw() after sendmail invocation.
+        
+        If a curses gpg pinentry is displayed by the $sendmail program, the
+        keypad() needs to called afterwards for it to re-enable function/arrow
+        keys.
+
+M	sendlib.c
+
+2022-01-15 11:19:05 -0800  Kevin McCarthy  <kevin@8t8.us> (100d6f3b)
+
+        * Revert "Add $reply_prefix."
+        
+        This reverts commit 9c1ce59874ce1c8e97d0c5bd71847596dafb1d50.
+        
+        The change is controversial, and probably ill-advised.  Thanks to
+        everyone for the feedback.
+
+M	globals.h
+M	init.h
+M	send.c
+
+2022-01-13 14:35:20 -0800  Kevin McCarthy  <kevin@8t8.us> (5b8e4605)
+
+        * Add config.h dependency for keymap_defs.h generation.
+        
+        Ensure the header is regenerated if configure is re-run.  The $(OPS)
+        list might change as a result of re-configure, but the timestamps of
+        those files won't.  However, config.h will change so we can check that
+        as a dependency.
+
+M	Makefile.am
+
+2022-01-10 17:53:09 -0800  Kevin McCarthy  <kevin@8t8.us> (4eff6049)
+
+        * Force resort_init if $reply_regexp changes.
+        
+        Simply changing real_subj isn't enough to make sure the change is
+        reflected in the current index.  There may actually be a more subtle
+        way to achieve this, but for now I'll just dump the subj_hash and
+        force rethreading.
+
+M	init.c
+M	init.h
+
+2022-01-10 14:37:06 -0800  Kevin McCarthy  <kevin@8t8.us> (9c1ce598)
+
+        * Add $reply_prefix.
+        
+        Allow the user to modify the subject prefix used in a reply.
+        Hopefully this would be used sparingly, but nothing prevented a user
+        from manually modifying the prefix themselves before.
+
+M	globals.h
+M	init.h
+M	send.c
+
+2022-01-10 14:00:40 -0800  Kevin McCarthy  <kevin@8t8.us> (466d3b8b)
+
+        * Localize $reply_regexp.
+        
+        This will allow translators to add additional reply prefixes common to
+        their locale.
+        
+        Add more documentation to the option, to explain the parts of the
+        regular expression.  Give an example of adding more prefixes.
+        
+        Note that the result of this is stored in the header cache.  Probably
+        people rarely were affected by this before, but with the new
+        localization, people may experiment and need to be aware to turn the
+        header cache off while testing.
+
+M	doc/makedoc.pl
+M	init.c
+M	init.h
+
+2022-01-09 15:28:23 -0800  Kevin McCarthy  <kevin@8t8.us> (cd3a5c85)
+
+        * Fix $reply_regexp default value.
+        
+        The default value happened to work, but didn't make logical sense.  It
+        was using a bracket expression that *included* a backslash, and
+        left-bracket, 0-9, and again a backslash.  The first ']' closed the
+        bracket expression, and then '+' matched that 1 or more times,
+        followed by ']'.  Fortunately the whole parenthisized expression had a
+        '*' so it would repeat matching, but as far as I can tell, it was
+        working accidentally.
+        
+        Because even the fixed version is hard to understand, add some
+        explanation of the parts to the documentation.
+
+M	init.h
+
+2022-01-09 14:57:07 -0800  Kevin McCarthy  <kevin@8t8.us> (3ca96fab)
+
+        * Fix manual generation string_unescape() of backslash values.
+        
+        Since the values are shown as if double-quoted, backslash also needs
+        to be properly escaped.
+
+M	doc/makedoc.pl
+
+2021-12-30 12:58:28 -0800  Kevin McCarthy  <kevin@8t8.us> (2926cf8c)
+
+        * Merge branch 'stable'
+
+2021-12-30 12:50:49 -0800  Kevin McCarthy  <kevin@8t8.us> (31b18ae9)
+
+        * automatic post-release commit for mutt-2.1.5
+
+M	ChangeLog
+M	VERSION
+
+2021-12-30 11:09:31 +0300  Emir Sarı  <bitigchi@me.com> (da60500d)
+
+        * Improve Turkish translations
+
+M	po/tr.po
+
+2021-12-28 13:42:43 -0800  Kevin McCarthy  <kevin@8t8.us> (7855611a)
+
+        * Add pager REDRAW_FLOW redraw flag when popping a menu.
+        
+        A setting change or window resize could take place in the called menu,
+        which will require reflowing the pager when returning to it.
+        
+        When converting to the menu stack, I left in manual REDRAW_FULL redraw
+        assignments, to be cautious and because it wasn't causing an problems.
+        However, with the REDRAW_FLOW added, the pager menu needs to have
+        those assignments removed after a menu call.
+
+M	menu.c
+M	pager.c
+
+2021-12-28 13:25:54 -0800  Kevin McCarthy  <kevin@8t8.us> (5c7e2bce)
+
+        * Merge branch 'stable'
+
+2021-12-27 12:59:49 -0800  Kevin McCarthy  <kevin@8t8.us> (938f91de)
+
+        * Add config variable mentions to Forwarding and Bouncing Mail.
+        
+        Mention $forward_attribution_intro, $forward_attribution_trailer, and
+        $forward_format.
+
+M	doc/manual.xml.head
+
+2021-12-27 12:25:24 -0800  Kevin McCarthy  <kevin@8t8.us> (752a1b44)
+
+        * Merge branch 'stable'
+
+2021-12-25 14:00:02 -0800  Kevin McCarthy  <kevin@8t8.us> (89681e9e)
+
+        * Merge branch 'stable'
+
+2021-12-23 17:30:27 -0800  Kevin McCarthy  <kevin@8t8.us> (4fd19957)
+
+        * Add new gsasl files to POTFILES.in.
+
+M	po/POTFILES.in
+
+2021-12-15 17:50:00 -0800  Kevin McCarthy  <kevin@8t8.us> (68caf914)
+
+        * Add GNU SASL support for authentication.
+        
+        It turns out Cyrus SASL's license may not be compatible with GPL
+        programs, see Debian Bug 999672.
+        
+        So, add support for the GNU SASL library, using configure option
+        --with-gsasl.
+        
+        I haven't touched the Cyrus SASL code in Mutt all that much in the
+        past, but I've done my best to keep the gsasl code clean and simple.
+        There are likely mistakes to be fixed and additions to be made,
+        though.
+        
+        I queried the gsasl mailing list about the need for a socket
+        wrapper (as is done for the cyrus code), and it seems this should no
+        longer be needed.  As long as GSASL_QOP is left at the
+        default (qop-auth), the client should ask for authentication, and not
+        negotiate integrity or confidentiality.  (Thanks to Phil Pennock and
+        Simon Josefsson for their reponses - although the blame is fully
+        on *me* if this turns out to be incorrect).  Therefore there is no
+        CONNECTION wrapping in this implementation.
+        
+        Add multiline response support for SMTP authentication (which is
+        probably not actually needed).  Also add arbitrary line length for the
+        SASL server responses (the RFCs note that for SASL, the protocol line
+        lengths don't apply).
+
+M	Makefile.am
+M	ascii.h
+M	configure.ac
+M	doc/makedoc-defs.h
+M	imap/Makefile.am
+M	imap/auth.c
+M	imap/auth.h
+A	imap/auth_sasl_gnu.c
+M	main.c
+A	mutt_sasl_gnu.c
+A	mutt_sasl_gnu.h
+M	pop_auth.c
+M	smtp.c
+
+2021-12-21 18:29:26 -0800  Kevin McCarthy  <kevin@8t8.us> (e3faeb03)
+
+        * Add mutt_socket_buffer_readln().
+        
+        This will be useful for SMTP and POP SASL reading, where line of
+        arbitrary length are required to be supported.
+
+M	mutt_socket.c
+M	mutt_socket.h
+
+2021-12-20 15:37:16 -0800  Kevin McCarthy  <kevin@8t8.us> (22736484)
+
+        * Merge branch 'stable'
+
+2021-12-19 21:45:12 -0500  Mike Frysinger  <vapier@gentoo.org> (a4da1664)
+
+        * simplify envelope subject handling code a little
+        
+        Avoid excess defines.  Compiled code is the same.
+
+M	init.c
+
+2021-12-14 13:36:15 -0800  Kevin McCarthy  <kevin@8t8.us> (80ba3a49)
+
+        * Merge branch 'stable'
+
+2021-12-13 18:55:14 -0800  Kevin McCarthy  <kevin@8t8.us> (f1cc1309)
+
+        * Merge branch 'stable'
+
+2021-12-11 13:22:40 -0800  Kevin McCarthy  <kevin@8t8.us> (fcecddaa)
+
+        * Merge branch 'stable'
+
+2021-12-07 14:46:12 -0800  Kevin McCarthy  <kevin@8t8.us> (ac348e19)
+
+        * Merge branch 'stable'
+
+2021-12-05 12:02:34 +0000  Torsten Franz  <torsten.franz@ubuntu.com> (9dbde32c)
+
+        * Update de.po
+
+M	po/de.po
+
+2021-12-02 15:44:37 -0800  Kevin McCarthy  <kevin@8t8.us> (32c03d3d)
+
+        * Fix sidebar counters with $maildir_trash.
+        
+        On close, the sidebar decrements deleted messages, but should not when
+        $maildir_trash is set.
+
+M	mx.c
+
+2021-12-02 15:36:59 -0800  Kevin McCarthy  <kevin@8t8.us> (29ed0c2e)
+
+        * Improve the statusbar modified flag for $maildir_trash.
+        
+        Add a context counter for the trash status flag.  Compare that to the
+        number of deleted messages to determine modified state.
+        
+        Note that "undeleting" a trashed message will set context->changed.
+        Without that we couldn't do a simple counter comparison.  But with
+        context->changed checked first, we can assume equal trash and deleted
+        flags then means there are no "delete" modifications pending.
+
+M	init.h
+M	mutt.h
+M	mx.c
+M	status.c
+
+2021-12-01 18:12:11 -0800  Kevin McCarthy  <kevin@8t8.us> (ede3cf58)
+
+        * Ignore $delete when sync'ing with $maildir_trash set.
+        
+        When closing a mailbox, the $delete quadoption is not consulted to
+        determine whether to write out the T flags.  However, for some reason,
+        sync was doing so.
+        
+        Since we aren't "purging" the messages, the prompt doesn't make sense.
+        Make close and sync consistent by skipping the $delete check for sync.
+
+M	init.h
+M	mx.c
+
+2021-11-24 13:51:32 -0600  Aaron Poffenberger  <akp@hypernote.com> (e8a01c1a)
+
+        * Fix unused variable warning.
+
+M	mutt_ssl.c
+
+2021-11-20 14:21:29 -0800  Kevin McCarthy  <kevin@8t8.us> (04a7f1d9)
+
+        * Allow an empty To: address prompt.
+        
+        If the user has $askcc or $askbcc configured, they may want to put
+        addresses in those without filling in the To field.
+        
+        Instead check to make sure at least one of the fields is filled in
+        after all the prompt before aborting.
+
+M	send.c
+
+2021-11-11 15:05:33 -0800  Kevin McCarthy  <kevin@8t8.us> (543e1151)
+
+        * Convert COLOR_DEFAULT constant to -1.
+        
+        Previously, "mono" assigned -1 to fg/bg to indicate "unset" values.
+        
+        NCurses uses -1 to indicate "default" color (when supported), but
+        COLOR_DEFAULT was assigned the value -2 and swapped at the last
+        moment.  While it worked, I personally found this confusing.
+        
+        To make the logic clearer, create a COLOR_UNSET constant with value
+        -2, and switch COLOR_DEFAULT to -1.  Then remove the last-minute
+        translation when allocating the color.
+        
+        Change the "not set" initialization and testing to use COLOR_UNSET for
+        clarity.
+        
+        Because map_getvaluebyname() returns -1 for a missing value,
+        create a new lookup function that returns the matching mapping_t or
+        NULL if not found.
+
+M	color.c
+M	init.c
+M	mapping.h
+
+2021-11-08 20:45:03 +0100  Charalampos Kardaris  <ckardaris@outlook.com> (f210e9a0)
+
+        * [Fix] poll to use PollFdsCount and no PollFdsLen
+        
+        PollFdsLen tracks the maximum elements of the pollfd struct before
+        needing reallocation, whereas PollFdsCount tracks the number of actual
+        elements in it. Polling has to be done only on the actual number of
+        elements.
+
+M	monitor.c
+
+2021-10-26 14:42:17 -0700  Kevin McCarthy  <kevin@8t8.us> (623eeb26)
+
+        * Add a few mutt variable configure dependency comments.
+        
+        Add dependency comments to a few variables where it might not be
+        obvious are dependent on the feature being enabled.
+        
+        While reviewing the dependencies, I noticed a couple #if could be
+        indented for more uniformity in the file.
+
+M	init.h
+
+2021-10-22 18:08:48 -0700  Kevin McCarthy  <kevin@8t8.us> (75be20d6)
+
+        * Fix color object case insensitive comparisons.
+        
+        In most of the color code, mutt_getvaluebyname() is used to compare
+        against valid tables, which is (ascii) case insensitive.
+        
+        However, 'uncolor' and 'color quoted' for some reason used case
+        sensitive comparison.
+        
+        The uncolor code can use object comparison instead, as the matching
+        Fields value was just retrieve.  For 'color quoted', switch to using
+        ascii_strncasecmp() instead.
+
+M	color.c
+
+2021-10-20 14:30:48 -0700  Kevin McCarthy  <kevin@8t8.us> (9c1c9182)
+
+        * Correct multipart/alternative manual section slightly.
+        
+        Step three only selects from enriched, plain, and then html; not _any_
+        text type.
+        
+        Step four considers any text type as one that Mutt can handle
+        internally.
+
+M	doc/manual.xml.head
+
+2021-10-19 18:48:47 -0700  Kevin McCarthy  <kevin@8t8.us> (60ab5f11)
+
+        * Add internal mutt_ctime() implementation.
+        
+        ctime() is marked obsolescent in the POSIX guide, so we ought to stop
+        using it to ensure future portability.
+
+M	from.c
+M	mx.c
+M	protos.h
+
+2021-10-12 13:31:13 -0700  Kevin McCarthy  <kevin@8t8.us> (534f3ae0)
+
+        * Improve pattern compiler whitespace skipping.
+        
+        Commit 564b515c fixed a problem with nested patterns, where trailing
+        whitespace could cause a compilation error.  That fix added missing
+        SKIPWS() calls in a couple branches.
+        
+        However, the real problem is that the loop doesn't ensure whitespace
+        is skipped *before* checking for *ps.dptr.
+        
+        Rather than relying on all the case branches (and "eater" functions)
+        to skip the whitespace after processing, rearrange the logic so the
+        SKIPWS() is after the switch statement.
+
+M	pattern.c
+
+2021-10-09 19:55:00 +0200  Vincent Lefevre  <vincent@vinc17.net> (6270278f)
+
+        * Updated French translation.
+
+M	po/fr.po
+
+2021-10-07 17:40:50 -0700  Kevin McCarthy  <kevin@8t8.us> (986e9e74)
+
+        * Remove trailing slashes when opening maildir/mh mailboxes.
+        
+        IMAP cleans this up already (and file-based mailboxes don't allow it).
+        
+        It's easy to have this happen by tab-completing a maildir mailbox at
+        the change folder prompt.  Removing the slash makes the display more
+        uniform, and prevents some problems such as $trash folder comparison.
+
+M	mh.c
+
+2021-10-07 13:02:29 -0700  Kevin McCarthy  <kevin@8t8.us> (14618717)
+
+        * Note that IMAP mailboxes polling doesn't support flag counts.
+        
+        The IMAP protocol only supports unread and total counts, but the
+        manual had no mention of this.
+
+M	doc/manual.xml.head
+M	init.h
+
+2021-09-28 15:02:34 -0700  Kevin McCarthy  <kevin@8t8.us> (466432b5)
+
+        * Add $attach_save_charset_convert quadoption.
+        
+        This prompts to allow charset conversion of received text-type
+        attachments.
+        
+        Mutt automatically converts when sending to a value from
+        $send_charset.  If an attachment were automatically converted from
+        utf-8 to iso-8859-1 on the sending side, then on the receiving
+        side (also 99.9% likely to be utf-8 nowadays) it may not be helpful to
+        save the the originally utf-8 attachment as iso-8859-1.  This can be
+        especially destructive for attached patches or translation updates.
+        
+        Add the quadoption, defaulting to ask-yes so people know about it.
+        With a quadoption, it can be easily completely turned off or made the
+        default if the user desires.
+        
+        Add code to skip the prompt if converting from ascii to extended
+        ascii (e.g. utf-8 or iso-8859-*).  There may be others too but this
+        should cover almost all cases nowadays.