2021-06-12 10:40:18 -0700  Kevin McCarthy  <kevin@8t8.us> (bfccf9c6)

        * Update UPDATING file for 2.1.0 release.

M	UPDATING

2021-06-07 10:15:24 -0700  Vsevolod Volkov  <vvv@mutt.org.ua> (0b2be236)

        * Updated Russian translation.

M	po/ru.po

2021-06-07 10:14:14 -0700  Vsevolod Volkov  <vvv@mutt.org.ua> (6b458ebd)

        * Updated Ukrainian translation.

M	po/uk.po

2021-06-05 13:49:57 -0700  Kevin McCarthy  <kevin@8t8.us> (e8792708)

        * Minor UPDATING file fixes.

M	UPDATING

2021-06-05 13:48:05 -0700  Kevin McCarthy  <kevin@8t8.us> (bab86bf4)

        * Mention <browse-mailboxes> in the new-mail documentation.
        
        Add to the sentence mentioning '-y'.

M	doc/manual.xml.head

2021-06-03 10:40:22 +0200  Grzegorz Szymaszek  <gszymaszek@short.pl> (6a7f32a6)

        * Update the Polish translation for Mutt 2.1

M	po/pl.po

2021-05-31 16:55:39 +0200  Petr Písař  <petr.pisar@atlas.cz> (732efe2e)

        * Update Czech translatation for mutt-2.1.0

M	po/cs.po

2021-05-30 10:20:50 -0700  Helge Kreutzmann  <debian@helgefjell.de> (f0a64564)

        * Update de.po.

M	po/de.po

2021-05-29 12:39:25 -0700  Helge Kreutzmann  <debian@helgefjell.de> (b37d5aa1)

        * Update de.po.

M	po/de.po

2021-05-29 12:38:04 -0700  Kevin McCarthy  <kevin@8t8.us> (c728976c)

        * Revert "Update de.po."
        
        This reverts commit e450615f5b0a77539916d9e8542c9a64bca73c09.
        
        Author was not properly attributed.  Will reapply with author set.

M	po/de.po

2021-05-29 12:14:30 -0700  Kevin McCarthy  <kevin@8t8.us> (e450615f)

        * Update de.po.

M	po/de.po

2021-05-28 21:21:32 +0200  Flammie Pirinen  <flammie@iki.fi> (3d24855d)

        * Update Finnish translation

M	po/fi.po

2021-05-28 20:48:58 +0200  Eike Rathke  <erack@erack.de> (36fb911f)

        * Do not set From reverse name on postponed or to be resend messages
        
        Messages to be resend or resumed from postponed already had
        reverse names and alternates evaluated the first time they entered
        the compose menu and send-hook were already executed then. The
        user may also had edited the From to any other content before
        postponing or sending the message. Upon resuming, do not overwrite
        with a reverse_name lookup and do not remove the real name if
        reverse_realname=no is set.
        
        This restores the old behaviour and fixes the regression
        introduced with
        
            commit 961cf4bed679d9495cc3682d18496a53061d4581
            CommitDate: Mon Feb 1 14:00:56 2021 -0800
        
                Apply $reverse_name when tag-reply/forwarding too.

M	send.c

2021-05-28 20:29:03 +0300  Emir Sarı  <bitigchi@me.com> (ddd292b8)

        * Update Turkish translations

M	po/tr.po

2021-05-28 14:23:37 +0200  Greg Kroah-Hartman  <greg@kroah.com> (88845a41)

        * fix gcc-11 build warnings for H_TO_INTL()
        
        The macro H_TO_INTL() in mutt_idna.c, causes a lot of build warnings
        about the indentation of the if () line to try to warn the developer
        that it looks like this might be a bug:
        
        mutt_idna.c: In function ‘mutt_env_to_intl’:
        mutt_idna.c:327:5: warning: this ‘if’ clause does not guard... [-Wmisleading-indentation]
          327 |     if (tag) *tag = #a; e = 1; err = NULL;              \
              |     ^~
        mutt_idna.c:333:3: note: in expansion of macro ‘H_TO_INTL’
          333 |   H_TO_INTL(return_path);
              |   ^~~~~~~~~
        mutt_idna.c:327:25: note: ...this statement, but the latter is misleadingly indented as if it were guarded by the ‘if’
          327 |     if (tag) *tag = #a; e = 1; err = NULL;              \
              |                         ^
        mutt_idna.c:333:3: note: in expansion of macro ‘H_TO_INTL’
          333 |   H_TO_INTL(return_path);
              |   ^~~~~~~~~
        
        This goes on for many lines.
        
        Fix this up by properly indenting the lines so that gcc "knows" we mean
        well here.

M	mutt_idna.c

2021-05-28 14:38:41 +0300  Maxim Tarasov  <mu@magi.net.ru> (76276257)

        * Trim the ToC for the option list in the manual
        
        Currently, every variable is shown in the table of contents of the
        reference manual. Variable list is over 400 items long. This makes the
        ToC difficult to work with.
        
        Removing is done by adding an overwriting template for the parent of
        variables section (that is sect1 tag with id="variables"). In the
        DocBook source matching template calls another template named subtoc
        with two parameters: context (self element) and nodes (child
        sections). Normally a list of descendant nodes is passed here, but we
        don't want that and instead call subtoc with no parameters.
        
        This approach results in semantically correct code being generated in
        all modes.
        
        Note, that the ID (variables) will have to be kept in sync with XML
        generating code, but that is already hardcoded elsewhere.

M	doc/TODO
M	doc/mutt.xsl

2021-05-26 15:41:40 -0700  Kevin McCarthy  <kevin@8t8.us> (f60882fc)

        * Add debug statements around $hostname setting.
        
        This will make slowdowns due to DNS issues more obvious.

M	init.c

2021-05-26 15:30:20 -0700  Kevin McCarthy  <kevin@8t8.us> (30aea067)

        * Add draft 2.1.0 notes to UPDATING file.

M	UPDATING

2021-05-23 11:20:54 -0700  Kevin McCarthy  <kevin@8t8.us> (ff7c68cc)

        * Merge branch 'stable'

2021-05-23 11:19:02 -0700  Kevin McCarthy  <kevin@8t8.us> (9f40cc39)

        * Update irc channel to libera.chat.

M	doc/manual.xml.head

2021-05-14 18:55:00 -0600  Gregory Anders  <greg@gpanders.com> (4c786d87)

        * Add local_date_header option
        
        Add an option to convert the date used in the Date header into the local
        (sender's) timezone. This is the current behavior and the option
        defaults to true. Unsetting this option causes the date in the Date
        header to be formatted using the GMT timezone.
        
        This option is useful for privacy-sensitive users who may not wish to
        divulge their sending timezone.

M	init.h
M	mutt.h
M	sendlib.c

2021-05-04 11:03:32 -0700  Kevin McCarthy  <kevin@8t8.us> (791fc1c6)

        * Merge branch 'stable'

2021-05-04 10:55:15 -0700  Kevin McCarthy  <kevin@8t8.us> (481f3800)

        * automatic post-release commit for mutt-2.0.7

M	ChangeLog
M	VERSION

2021-05-04 10:38:24 -0700  Kevin McCarthy  <kevin@8t8.us> (bd29a9a0)

        * Merge branch 'stable'

2021-04-28 12:57:53 -0700  Kevin McCarthy  <kevin@8t8.us> (b24a1a25)

        * Add comments to the 2231 decode blocks.

M	rfc2231.c

2021-04-30 18:55:37 -0700  Kevin McCarthy  <kevin@8t8.us> (a10d0bb5)

        * Add sanitize call in print mailcap function.
        
        The mutt_rfc1524_expand_filename() function calls mutt_adv_mktemp(),
        which also calls the sanitizer.
        
        However, if the recv-mode suggested filename has a slash, it's better
        to sanitize before calling expand_filename() too, to preserve the
        entire filename (albeit with the slashes sanitized out).
        
        For example (ignoring name templates) if the file were called
        "either/or.pdf", pre-sanitizing would generate "either_or.pdf", while
        not doing so would generate "or.pdf".

M	attach.c

2021-04-27 13:37:21 -0700  Kevin McCarthy  <kevin@8t8.us> (98f0ca6b)

        * Add allow_8bit flag to sanitize_filename functions.
        
        This will allow "view" mailcap invocations and autoview invocations to
        display something reasonable for international file names.  Also add
        the flag to mutt_adv_mktemp(), which the mailcap viewer calls.
        
        Convert the 'slash parameter' to an ALLOW_SLASH flag.  This makes the
        flags consistent (both "allow" something not permitted by default),
        and invocations more readable.

M	attach.c
M	handler.c
M	lib.c
M	lib.h
M	muttlib.c
M	pop.c
M	protos.h
M	rfc1524.c

2021-04-29 12:54:41 -0700  Kevin McCarthy  <kevin@8t8.us> (fdee288a)

        * Modify makedoc.pl to read sort maps for the default value.
        
        Use the map tables to print out a correct default value for sort
        types.
        
        Remove the unneeded SORT_ALPHA type, for $sort_browser documentation,
        since we can do the lookup now.

M	doc/makedoc.pl
M	init.h
M	sort.h

2021-04-28 13:28:14 -0700  Kevin McCarthy  <kevin@8t8.us> (23b77a20)

        * Arrange sort maps by value instead of name.
        
        The value-to-name mapping uses the first match.  Grouping by value
        makes it more obvious when there are duplicates and which one will be
        printed out, e.g. when querying the value of a variable.

M	init.h

2021-04-28 13:21:17 -0700  Kevin McCarthy  <kevin@8t8.us> (bee3474f)

        * Move sort maps before vars in init.h.
        
        This is so makedoc.pl can read in the maps and use them to properly
        print out default values for the sort options.

M	init.h

2021-04-28 11:16:50 -0700  Kevin McCarthy  <kevin@8t8.us> (acb2bdd4)

        * Merge branch 'stable'

2021-04-25 15:01:07 -0700  Kevin McCarthy  <kevin@8t8.us> (515a315c)

        * Mention ~B in $thorough_search documentation.
        
        It affects ~b, ~B, and ~h.

M	init.h

2021-04-25 10:47:32 -0700  Kevin McCarthy  <kevin@8t8.us> (b2125bcd)

        * Keep file browser '..' entries at the top.
        
        This makes it easier to navigate up a directory when the file browser
        is sorted by a value other than "alpha".

M	browser.c

2021-04-24 15:10:22 -0700  Kevin McCarthy  <kevin@8t8.us> (2333015f)

        * Change browser sort "unsorted" to preserve the original order.
        
        Record entry numbers as they are added to the list, so re-sorting by
        "unsorted" can return to the original order.
        
        This is most useful for the mailboxes case, where "unsorted" should
        mean the order listed in the muttrc.

M	browser.c
M	browser.h
M	imap/browse.c

2021-04-24 14:56:39 -0700  Kevin McCarthy  <kevin@8t8.us> (f8984468)

        * Fix imap_add_folder() reallocation check.
        
        entrymax contains the number of slots allocated.  entrylen contains
        the current number of slots used.
        
        Note: there isn't a memory access bug here - it was merely
        reallocating too early.

M	imap/browse.c

2021-04-24 14:41:20 -0700  Kevin McCarthy  <kevin@8t8.us> (a877a29e)

        * Add $sort_browser_mailboxes configuration variable.
        
        This allows to control file browser sorting and mailbox browser
        sorting separately.

M	browser.c
M	browser.h
M	init.h
M	sort.h

2021-04-24 15:23:49 -0700  Kevin McCarthy  <kevin@8t8.us> (8b7dd25b)

        * Merge branch 'stable'

2021-04-23 18:35:30 -0700  Kevin McCarthy  <kevin@8t8.us> (30e657d3)

        * Improve sidebar drawing in small windows.
        
        Ensure the index window is given at least one column during reflow,
        truncating the sidebar width as necessary.
        
        Change the sidebar to use MuttSidebarWindow->cols instead of directly
        referencing SidebarWidth.

M	curs_lib.c
M	sidebar.c

2021-04-22 15:42:14 -0700  Kevin McCarthy  <kevin@8t8.us> (fe866a19)

        * Disable markers when wrap is at column 1.
        
        This can occur when the user resizes the window to a single column, or
        if they 'set wrap=1'.
        
        In that case, Mutt was outputting the '+' and setting the initial
        output to the second column.  However, if there is only one column,
        nothing would then be printed.  When trying to search or jump to end,
        Mutt would enter an infinite loop because no forward progress would be
        made in the file with each line.
        
        Disable the markers if there is only a single column, so at least one
        character of the message will get printed on each line.

M	pager.c

2021-04-19 18:53:30 -0700  Kevin McCarthy  <kevin@8t8.us> (8162c129)

        * Improve body_pattern and quoting when skipping down.
        
        The previous commit added the MUTT_TYPES flag to the search operation.
        However, adding the flag caused body_pattern and classify_quote
        matching to occur, as part of searching.
        
        Separate those two to only be done when MUTT_SHOWCOLOR is passed.  Keep
        the classify_quote() call in resolve_types(), since otherwise during a
        normal display we would end up matching the regexps twice.
        
        Add a flag so the computation only occurs once, instead of each time
        the line is shown.
        
        For continuation lines, we need to scan from the start of line.
        Otherwise body_patterns and proper quote level coloring won't appear
        if we jump down to the middle of a continuation line.
        
        As part of this commit, I noticed is_cont_hdr was not reset when
        handling a REDRAW_FLOW.  I believe this also fixes the old trac ticket
        issue 3744.

M	pager.c

2021-04-19 14:49:52 -0700  Kevin McCarthy  <kevin@8t8.us> (7bce3182)

        * Add MUTT_TYPES flag for pager search operation.
        
        This allows the display_line() search scanner to also determine the
        type of each line.  Without that, searching down would not be able to
        properly highlight header lines or quote lines.

M	pager.c

2021-04-19 14:48:01 -0700  Kevin McCarthy  <kevin@8t8.us> (f8e405b3)

        * Revert "Add type resolution when searching in the pager."
        
        This reverts commit aa308b67a983918eb59f1d7b5af543817c2cfd04.
        
        This may be what I want to push to master eventually, but this commit
        was accidental.  I need to test it more on a branch and add a better
        commit message before commiting to master.

M	pager.c

2021-04-19 14:45:01 -0700  Kevin McCarthy  <kevin@8t8.us> (aa308b67)

        * Add type resolution when searching in the pager.

M	pager.c

2021-04-19 14:20:47 -0700  Kevin McCarthy  <kevin@8t8.us> (4b6cd846)

        * Add a note about crypto operations being available interactively.
        
        They aren't available in batch mode.  This was mentioned in the "batch
        compose flow" section, but should be mentioned a bit more prominently
        in the manual.

M	doc/manual.xml.head

2021-04-18 14:54:02 -0700  Kevin McCarthy  <kevin@8t8.us> (10c66a53)

        * Mention 8-bit ansi color support in manual.

M	doc/manual.xml.head

2021-04-17 09:34:16 -0700  Kevin McCarthy  <kevin@8t8.us> (36a704cf)

        * Add cc, in-reply-to, and references to default mailto_allow list.
        
        Thanks to Gregory Anders for the ticket and suggested patch, which
        this commit is based upon.

M	doc/manual.xml.head
M	init.c

2021-04-13 18:02:31 -0700  Kevin McCarthy  <kevin@8t8.us> (0684a051)

        * Add a check to make sure pair doesn't overflow.
        
        It's not clear if COLOR_PAIRS can be larger than SHRT_MAX + 1, so make
        sure with a check.

M	color.c

2021-04-11 15:56:42 -0700  Kevin McCarthy  <kevin@8t8.us> (2d683201)

        * Fix ansi colors to use default bg/fg when not set.
        
        Note the ColorList stores COLOR_DEFAULT (-2) not -1, so we need to
        translate those values in mutt_alloc_ansi_color().  If the curses
        backend doesn't support default colors, we abort and just return color
        pair 0.

M	color.c
M	pager.c

2021-04-11 14:19:33 -0700  Kevin McCarthy  <kevin@8t8.us> (c20a5169)

        * Use bkgrndset() as ATTRSET() backend if available.
        
        This function allows more usable color pairs, because it passes the
        pair parameter separately.  The COLOR_PAIR() macro only returns 8-bits
        for a pair, limiting the total pairs used to 256 despite many systems
        having much larger COLOR_PAIRS values.
        
        In order to do this, separate out the pair and attrs values, instead
        of combining attr + COLOR_PAIR(pair) into a single value and storing
        that inside Mutt.  For the older calls, bkgdset() and attrset(), we
        call COLOR_PAIR() at the last minute.
        
        Add COLOR_ATTR to header cache sum calculation, since it is used in
        the HEADER now.

M	Makefile.am
M	color.c
A	color.h
M	commands.c
M	configure.ac
M	curs_main.c
M	hcache.c
M	hcachever.pl
M	menu.c
M	mutt.h
M	mutt_curses.h
M	mutt_menu.h
M	pager.c
M	protos.h
M	score.c
M	sidebar.c
M	thread.c

2021-04-10 11:33:58 +0200  Vincent Lefevre  <vincent@vinc17.net> (869c6cc4)

        * Updated French translation.

M	po/fr.po

2021-04-09 17:51:12 -0700  Kevin McCarthy  <kevin@8t8.us> (6a2e3dea)

        * Clarify -H usage in batch mode is not a "pass through" option.
        
        Mutt runs the file through mutt_prepare_template() in both cases,
        which will decrypt, remove signatures, strip multipart/alternatives.
        
        The documentation wasn't clear about this, and it might be
        misunderstood that -H is a "pass through" option.

M	doc/manual.xml.head
M	doc/mutt.man

2021-04-09 15:12:11 -0700  Kevin McCarthy  <kevin@8t8.us> (6c510773)

        * Add user-friendly error message if append to $trash fails.
        
        For the maildir case, this will then be displayed instead of a raw
        stat error message, in the case the mailbox disappears.

M	mx.c

2021-03-29 04:57:28 -0300  Marco Ippolito  <maroloccio@gmail.com> (9566a95c)

        * Fix documentation typo

M	doc/manual.xml.head

2021-03-27 19:14:10 -0700  Kevin McCarthy  <kevin@8t8.us> (4e01ccdb)

        * Stop batch mode emails with no argument or recipients.
        
        Closes #345.  There are checks in send.c to guard against a
        no-recipient email being sent, but there might be a my_hdr or other
        setting in the muttrc.  In that case, it would be surprising for an
        email to be sent.

M	main.c

2021-03-20 18:33:01 +0100  Vincent Lefevre  <vincent@vinc17.net> (8ac730a1)

        * Updated French translation.

M	po/fr.po

2021-03-19 14:35:53 -0700  Kevin McCarthy  <kevin@8t8.us> (5ee27795)

        * Merge branch 'stable'

2021-03-16 14:13:38 -0700  Kevin McCarthy  <kevin@8t8.us> (9b8cc85e)

        * Add SigInt handler for pattern functions.
        
        Allow, for example, a ~B against a huge mailbox to be interruped via
        ctrl-c.  This commit follows the same behavior as
        mutt_search_command() - not prompting and just interrupting.  We could
        add a prompt if it's really deemed necessary, but I think if they hit
        ctrl-c they want to stop the pattern operation now.
        
        Since it's not practical to roll back, leave the Context in the "in
        the middle" state too.  This may actually be desired.

M	pattern.c

2021-03-15 13:11:07 -0700  Kevin McCarthy  <kevin@8t8.us> (50f22244)

        * Turn off ~h header-only download when $message_cachedir is set.
        
        The initial pattern match will be slow, but then the body will be in
        the message cache.  This is the same as previous behavior, so a
        message cache user should not notice any change.
        
        Those not using a message cache will see improvements from the
        header-only download, but of course they will have to download the
        full headers each time.

M	pattern.c

2021-03-13 18:51:34 +0100  Vincent Lefevre  <vincent@vinc17.net> (5eb9dec8)

        * Updated French translation.

M	po/fr.po

2021-03-12 13:23:42 -0800  Kevin McCarthy  <kevin@8t8.us> (5cf52cd1)

        * Minor fixes to match_body_patterns().
        
        Change the end condition to compare against buflen.  I prefer this
        because we are not directly incrementing offset ourselves, but relying
        on the rational return values from regexec.  If the regex library
        (incorrectly) returned a nul-match at the '\0' terminator for
        instance, the new nul-match logic would assign offset just past the
        terminator.
        
        Remove the i=0 initialization during definition, because it doesn't
        use or need it.

M	pager.c

2021-03-12 13:08:29 -0800  Kevin McCarthy  <kevin@8t8.us> (64dba506)

        * Factor out match_body_patterns to make resolve_types() simpler.
        
        The body pattern matching part has been getting more complex, and is
        now too big to have its details sit amongst the resolve_types()
        logic.  Factor it out into a separate procedure.

M	pager.c

2021-03-10 18:10:39 -0800  Kevin McCarthy  <kevin@8t8.us> (63d29881)

        * Fix body color null_rx handling.
        
        The existing code did not work properly.  If the null_rx occurred
        before another matching entry in the list, it would be ignored.  If it
        occurred afterward, it could cause the entries to be created
        out-of-order which would affect the binary search later on when
        rendering.
        
        Instead, use the normal "match assignment" code, but if the best
        result is a null_rx match, simply discard it and increment the offset
        past the end of the null match.

M	pager.c

2021-03-10 15:09:49 -0800  Kevin McCarthy  <kevin@8t8.us> (53ffdb93)

        * Improve body color matching speed by caching future matches.
        
        On a *very* long body (or header_partial) lines with multiple color
        lines that match, performance can be degraded.  For instance if there
        were moderately expensive regexp matches against "A" "B" and "C".  A
        line with:
        
          A A A A A B A A C A A A B A A A C
        
        The B and C regexps were scanned again after every "A" match, despite
        that the result would be discarded again for the next A match.
        
        Change the color matching to cache the location of each color line
        match.  Discard the cache once the match offset passes that point.

M	mutt_curses.h
M	pager.c

2021-03-11 14:02:03 -0800  Kevin McCarthy  <kevin@8t8.us> (17774247)

        * Merge branch 'stable'

2021-03-09 18:19:21 -0800  Kevin McCarthy  <kevin@8t8.us> (847ce203)

        * Add <view-alt-pager> to compose menu.
        
        This mirrors the <view-pager> in the attachment menu (and just added
        to the compose menu), but adds it to the compose menu for alternative
        generation.

M	OPS
M	compose.c
M	doc/manual.xml.head
M	functions.h

2021-03-09 18:05:17 -0800  Kevin McCarthy  <kevin@8t8.us> (5a325ba4)

        * Add view-mailcap, view-pager, view-text to compose menu.
        
        These are less commonly needed, so don't bind them to a key.  But for
        those who need it, make them be available.

M	OPS
M	compose.c
M	functions.h

2021-03-10 18:58:22 +0100  Vincent Lefevre  <vincent@vinc17.net> (db80f61a)

        * Updated French translation.

M	po/fr.po

2021-03-05 18:52:48 -0800  Kevin McCarthy  <kevin@8t8.us> (191b0513)

        * Fix $ssl_client_cert usage with SMTP.
        
        The ssl and gnutls client-cert setup code was calling
        mutt_account_getuser().  This caused two problems.  First, it's not
        necessarily the case that there will be a username.  Second,
        populating the user would cause smtp_open() to check for AUTH
        capabilities and call smtp_auth - even if the user is already
        authenticated by the cert.  The server won't advertize AUTH if they
        already authenticated, causing a connection abort.
        
        Remove prompt for mutt_account_getuser() in the ssl and gnutls client
        certificate connection code.  The SASL code has callbacks, so I don't
        understand why it would need this.  Let's take it out and see if
        anyone screams 8-P.  If necessary, we can add a mutt_account_getuser()
        call to the very beginning of imap_auth_sasl().
        
        Revamp the openssl ssl_passwd_cb() prompt.  From the man pages, it
        appears to be used for the cert decryption.  There's no need to call
        mutt_account_getuser() and use the generic mutt_account_getpass() just
        to read a password in.  Instead create a callback function version to
        customize the prompt for a client cert with just the host.
        
        Change the SMTP authentication test to check if the AUTH capabilities
        are set, instead of if the user field is set before calling
        smtp_auth().

M	account.c
M	account.h
M	mutt_ssl.c
M	mutt_ssl_gnutls.c
M	smtp.c

2021-03-07 09:02:23 +0100  Olaf Hering  <olaf@aepfle.de> (fcc62991)

        * refresh de.po
        
        align with result of make update-po
        
        Signed-off-by: Olaf Hering <olaf@aepfle.de>

M	po/de.po

2021-03-06 13:59:54 -0800  Kevin McCarthy  <kevin@8t8.us> (2054ccf2)

        * Relocate po OPS files to bottom of POTFILES.in.
        
        This matches the pre-0.21 migration order, and will reduce the size of
        updates for the next release.

M	po/POTFILES.in

2021-03-06 22:26:21 +0100  Olaf Hering  <olaf@aepfle.de> (1267c1ca)

        * refresh de.po
        
        Signed-off-by: Olaf Hering <olaf@aepfle.de>

M	po/de.po

2021-03-06 11:22:19 -0800  Kevin McCarthy  <kevin@8t8.us> (8940dbe0)

        * Merge branch 'stable'

2021-03-04 12:24:42 -0800  Kevin McCarthy  <kevin@8t8.us> (a796a1bd)

        * Add -nonumbers to lynx options.
        
        This will remove the useless link and field numbers from output in the
        manual.txt.

M	doc/Makefile.am

2021-03-04 12:18:30 -0800  Kevin McCarthy  <kevin@8t8.us> (22bab8ff)

        * Don't install an empty manual.txt file.
        
        Closes #334.
        
        If all the html to txt programs we use (lynx, elinks, w3m) are missing
        from the system, the shell redirection willl create an empty file.
        
        Change the install step to use -s "FILE exists and has a size greater
        than zero".
        
        Thanks to Mitchell Blank Jr for the ticket and suggested fix.

M	doc/Makefile.am

2021-03-03 23:17:55 +0100  Vincent Lefevre  <vincent@vinc17.net> (9e338a19)

        * Updated French translation.

M	po/fr.po

2021-03-03 23:13:09 +0100  Vincent Lefevre  <vincent@vinc17.net> (fb5f052f)

        * Updated French translation.

M	po/fr.po

2021-03-03 12:18:51 -0800  Kevin McCarthy  <kevin@8t8.us> (1d13381f)

        * Don't leak fp when aborting mutt_make_message_attach().
        
        I forgot to close up the fp before returning.

M	sendlib.c

2021-03-01 18:19:08 -0800  Kevin McCarthy  <kevin@8t8.us> (16c381d8)

        * Prompt for retry on decrypt failure for mime message attachments.
        
        Closes #331.
        
        Change $forward_decyrpt to a quadoption, defaulting 'yes' for backward
        compatibility.  This will allow users to choose on a per-message
        attachment basis if desired.
        
        If decoding from $mime_forward_decode fails or decrypting from
        $forward_decrypt, notify the user and prompt to retry attaching
        without the decode/decrypt.
        
        $forward_decrypt directly invokes decryption and returns -1 on error,
        while $mime_forward_decode uses the handler rendering, which returns a
        non-fatal 1 (to allow message display of failing encrypted message).
        So prompt on either case.
        
        If they decline to retry, or the retry fails, return NULL only on a
        hard-fail (-1) return code.  They've been warned and can peek at the
        final result.
        
        The send generate_body() code was not checking for a NULL result.
        This seems like a bad idea, so add a check and error message in that
        case.

M	init.h
M	mutt.h
M	send.c
M	sendlib.c

2021-02-26 14:34:10 -0800  Kevin McCarthy  <kevin@8t8.us> (ad14641e)

        * Merge branch 'stable'

2021-02-24 14:59:19 -0800  Kevin McCarthy  <kevin@8t8.us> (e0d08c59)

        * Add a note about the '^' mailbox shortcut.
        
        Emphasize that it's not always set, for example in command-line mode,
        or when evaluating the muttrc at startup.

M	doc/manual.xml.head

2021-02-23 14:41:47 -0800  Kevin McCarthy  <kevin@8t8.us> (b412d83b)

        * Merge branch 'stable'

2021-02-20 13:08:54 -0800  Kevin McCarthy  <kevin@8t8.us> (7fd331fd)

        * Scan OPS files directly for gettext translation strings.
        
        Previously, gettext's po/Makefile.in.in was modified to separately
        scan and merge a generated file: keymap_alldefs.h.  Gettext doesn't
        recommend nor provide support for scanning generated files.
        
        However, there is no reason we have to scan the generated file.
        Instead move the N_() markers and L10N comments into the OPS files.
        Then simply add the OPS files to po/POTFILES.in.
        
        Add a '--language=C' option to xgettext to avoid a warning about the
        OPS files not have any known language suffix.
        
        Modify gen_defs and doc/gen-map-doc to deal with the new format of the
        OPS files.

M	OPS
M	OPS.CRYPT
M	OPS.MIX
M	OPS.PGP
M	OPS.SIDEBAR
M	OPS.SMIME
M	doc/gen-map-doc
M	gen_defs
M	po/Makevars
M	po/POTFILES.in

2021-02-18 15:40:39 -0800  Kevin McCarthy  <kevin@8t8.us> (d4e13e4b)

        * Fix gettext update problems.
        
        The keymap_alldefs.h file was previously generated and scanned inside
        po/Makefile.in.in as an additional POTFILES files build dependency and
        step.
        
        The file can't be scanned by just adding it to po/POTFILES.in because
        it's a generated file that will be in the build directory.  The next
        commit will instead add the OPS* files directly to POTFILES.in to be
        scanned.
        
        For this commit, remove the generation of keymap_alldefs.h.
        
        Remove the redundant ALL_LINGUAS setting from configure.ac.  Reformat
        the LINGUAS file to be on one line, as the documentation shows.  I
        think it was okay before, but just to be safe.
        
        Fix L10N scanning by adding '--add-comments=L10N' to XGETTEXT_OPTIONS
        in po/Makevars.
        
        Set MSGID_BUGS_ADDRESS to silence a warning.
        
        Add gettext.h to the Makefile.am EXTRA_DIST list.
        
        Remove -Iintl flags from all Makefile.am files, since the intl
        directory is no longer bundled in Mutt.

M	Makefile.am
M	autocrypt/Makefile.am
M	configure.ac
M	imap/Makefile.am
M	po/LINGUAS
M	po/Makevars

2021-02-19 00:20:42 +0100  Vincent Lefevre  <vincent@vinc17.net> (8d94928e)

        * Remove the AM_GNU_GETTEXT_VERSION line as it makes autoreconf fail with gettext < 0.21 because autopoint complains that gettext is too old.
        
        Using this line with a version smaller than 0.21 makes autopoint itself
        fail, thus does not solve the problem.

M	configure.ac

2021-02-19 00:02:15 +0100  Vincent Lefevre  <vincent@vinc17.net> (33fe78af)

        * Add AM_GNU_GETTEXT_VERSION line to avoid a warning from autoreconf.

M	configure.ac

2021-02-07 18:27:47 -0800  Kevin McCarthy  <kevin@8t8.us> (ccc18061)

        * Update gettext to version 0.21.
        
        Run 'gettextize -f --no-changelog' to update to 0.21.
        
        This removes the entire intl directory, and relies on an external
        libintl only.
        
        $(INTLLIBS) should now be referred to as $(LIBINTL), so change all
        references.
        
        Unfortunately, some of the gettext m4 files were modified and "mutt
        customized" in the past.  I'm trying to keep them in a pristine state
        after this upgrade.  glibc21.m4, lcmessage.m4, codeset.m4 have also
        been removed.
        
        Manually call AM_ICONV in configure.ac.  This was previously called
        in the modified "MUTT_AM_GNU_GETTEXT" but the result $am_cv_func_iconv
        is tested just below.
        
        locale.h is a standard library defined by C99.  The gpgme and pgp code
        wrapped those includes in HAVE_LOCALE_H checks, but that check was
        performed inside of gettext.m4.  Remove the check inside those
        files (locale.h is already included unconditionally elsewhere in
        Mutt).
        
        Copy the sample gettext.h file and include that instead, as suggested
        by the gettext documentation.
        
        unistd.h is also a standard posix header, so remove HAVE_UNISTD_H
        checks inside mutt too.
        
        Create po/LINGUAS file listing available languages.
        
        Create po/Makevars from the template and customize for Mutt.
        
        Set PO_DEPENDS_ON_POT to no to prevent po file rebuilds.  According to
        the gettext manual, with this set, "PO files can be accidentally
        updated even if a POT file is not updated" when using git.

M	.gitignore
M	ABOUT-NLS
M	INSTALL
M	Makefile.am
A	config.rpath
M	configure.ac
M	crypt-gpgme.c
M	crypt.c
A	gettext.h
D	intl/COPYING.LIB-2
D	intl/COPYING.LIB-2.1
D	intl/ChangeLog
D	intl/Makefile.in
D	intl/VERSION
