* lisp/gnus/message.el (message-signature-separator): New face
(bug#17757).
(message-font-lock-keywords): Add it to the signature, and ensure
that the trailing space isn't removed when hitting RET.
(message--match-signature): New helper function.
* lisp/hilit-chg.el (hilit-chg-set-face-on-change): Remove
highlighting from characters that are just highlighted because of
something that has been previously deleted (bug#17784).
* src/image.c (UINT32) [TIFFLIB_VERSION >= 20210416]: Define to
use stdint.h type for recent libtiff versions. Reported by Andy
Moreton <andrewjmoreton@gmail.com>.
* lisp/apropos.el (apropos--preamble): Factor out (bug#17832)...
(apropos-print-doc): ... from here.
(apropos-library): Use it to display the apropos buffer even if it
has no symbols of its own.
* lisp/jka-cmpr-hook.el (jka-compr-info-uncompress-function): New
function (bug#18823).
(jka-compr-compression-info-list): Expand info with decompression
function.
* lisp/jka-compr.el (jka-compr-insert-file-contents): Fall back on
internal decompression function if external doesn't exist.
* lisp/isearch.el (isearch-search): Move (= (match-beginning 0) (match-end 0))
and (bobp)/(eobp) outside the call to isearch-filter-predicate.
Use forward-char 1/-1 on empty matches only when going to retry search.
(isearch-lazy-highlight-search): Remove (= (point) bound),
but leave (= (match-beginning 0) (match-end 0)) since empty matches
make no sense in lazy-highlighting.
* doc/emacs/dired.texi (Dired Visiting): Document it.
* lisp/dired.el (dired-kill-when-opening-new-dired-buffer): New
user option (bug#20598).
(dired-up-directory, dired-find-file): Use it.
(dired--find-possibly-alternative-file): New convenience command
to respect the user option.
If a TLS handshake is attempted before the completion of an
asynchronous TCP connection has been ascertained, our local state will
not be set up correctly for further progress and the sentinel "open"
event will never be sent. This can occur if sufficient time passes
after the initiation of an async TCP connection so that by the time
`wait_reading_process_output` is called, the connection has already
been established on the TCP level.
This somewhat timing-sensitive bug has plagued HTTPS connections on
some platforms, notably macOS, for a long time (bug#49449).
* src/process.c (wait_reading_process_output): Gate the TLS handshake
by the NON_BLOCKING_CONNECT_FD flag. The flag will be cleared as soon
as the TCP socket is found to be writable.
* test/src/process-tests.el (process-async-https-with-delay):
New test.
* src/sysdep.c (struct speed_struct): New struct.
(speeds): New static array of struct speed_struct.
(convert_speed): New static function to convert a numerical baud
rate (e.g., 9600) to a Bnnn constant defined in termios.h (e.g.,
B9600).
(serial_configure): Use convert_speed to make the call to cfsetspeed
compliant with its advertised API. (Bug#49524)
* test/Makefile.in (REPLACE_FREE, FREE_SOURCE_0, FREE_SOURCE_1):
New macros.
($(test_module)): Improve accuracy of test as to whether free.c
should be compiled; glibc 2.33 does not need it compiled and the
compilation breaks if you try, if you build with
--enable-gcc-warnings.
* lib-src/etags.c (regexp): Omit member force_explicit_name,
since it’s always true. All uses removed. This lets us
remove calls to strlen (name) where GCC isn’t smart enough
to deduce that name must be nonnull.
* lib-src/movemail.c (main): Fix bug that could cause
link (tempname, NULL) to be called.
* src/emacs.c (argmatch): Break check into two ‘if’s,
since GCC doesn’t seem to be smart enough to check the single ‘if’.
* src/gtkutil.c (xg_update_menu_item): Fix bug where strcmp
could be given a NULL arg.
* src/xfont.c (xfont_list_family): Use nonnull value for dummy
initial value.
* lisp/gnus/gnus-search.el (gnus-search-query-expand-key): It's
possible that KEY could be partially completed (ie no longer string=
to COMP), but not all the way. Use a more accurate test. Add
docstring.
Prevent some false-positives in conservative GC marking.
This doesn’t fix any correctness bugs; it’s merely to
reclaim some memory instead of keeping it unnecessarily.
* src/alloc.c (mark_maybe_pointer): New arg SYMBOL_ONLY.
All callers changed. Check that the pointer’s tag, if any,
matches the pdumper-reported type.
* src/alloc.c (mark_maybe_pointer): Also mark pointers
to pdumper objects, even when the pointers are tagged.
Add a FIXME saying why this isn’t enough.
