Avoid elisp crash for OpenPGP User IDs with no e-mail address
* lisp/gnus/mml-sec.el (mml-secure-check-user-id): Verify that
there is an e-mail address in the current User ID before trying
to downcase it. (Bug#34121)
Eli Zaretskii [Fri, 18 Jan 2019 08:51:57 +0000 (10:51 +0200)]
Fix a typo in ELisp manual
* doc/lispref/nonascii.texi (Converting Representations): Fix
inconsistency between @defun and the argument description of
'byte-to-string'. (Bug#34119)
Alan Third [Fri, 11 Jan 2019 16:52:59 +0000 (16:52 +0000)]
Prevent redrawing if frame is garbaged
* src/nsterm.m ([EmacsView viewWillDraw]): Cancel drawing if the frame
has been garbaged.
* src/xdisp.c (expose_window_tree, expose_frame): Remove NS only
exceptions.
Eli Zaretskii [Mon, 14 Jan 2019 15:36:04 +0000 (17:36 +0200)]
Attempt to fix hangs on MS-Windows due to C-g
* src/w32uniscribe.c (uniscribe_otf_capability): Set
inhibit-quit around calls to otf_features, because the latter
cons Lisp data structures while we are in a critical section.
* src/xdisp.c (ALLOCATE_HDC) [HAVE_NTGUI]: Set inhibit-quit.
(RELEASE_HDC) [HAVE_NTGUI]: Restore inhibit-quit.
(OPTIONAL_HDC, DECLARE_HDC): Remove macros, their job is now
done by ALLOCATE_HDC and by a single #ifdef.
(draw_glyphs): Adapt to the above changes in macros.
(Bug#34059)
Eli Zaretskii [Sun, 13 Jan 2019 15:50:16 +0000 (17:50 +0200)]
Fix Calc graph output on MS-Windows
The previous code relied on "pgnuplot" executable, which is
no longer provided with Gnuplot 5.x.
* lisp/calc/calc.el (calc-gnuplot-name): Set to "pgnuplot" on
MS-Windows only if such an executable exists.
* lisp/calc/calc-graph.el (calc-graph-w32-p): New defsubst.
(calc-graph-plot, calc-graph-command, calc-gnuplot-command)
(calc-graph-init): Call calc-graph-w32-p wherever we need to
do something special for invoking gnuplot on MS-Windows,
instead of comparing against calc-gnuplot-name.
(calc-graph-plot): Set the terminal to "qt" on MS-Windows when
pgnuplot.exe is not available.
(calc-graph-kill): Delete the temporary files only after
killing the gnuplot process, otherwise the deletion might fail
on MS-Windows because the files are still in use.
Paul Eggert [Wed, 4 Apr 2018 17:59:09 +0000 (10:59 -0700)]
Improve GC+Cairo workaround
Suggested by Eli Zaretskii (Bug#20890#31).
* src/font.h (font_data_structures_may_be_ill_formed): New function.
* src/ftfont.c (ftfont_close):
* src/ftcrfont.c (ftcrfont_close): Use it.
Eli Zaretskii [Fri, 4 Jan 2019 13:40:43 +0000 (15:40 +0200)]
Improve documentation of 'server-name'
* doc/emacs/misc.texi (Emacs Server, TCP Emacs server)
(emacsclient Options):
* lisp/server.el (server-name): Document the usage of
'server-name' to specify the server file as an absolute file
name. Do not merge to master. (Bug#33934)
Eli Zaretskii [Sat, 29 Dec 2018 15:34:57 +0000 (17:34 +0200)]
Improve documentation of 'file-local-name' and related APIs
* doc/lispref/files.texi (Unique File Names)
(Magic File Names, File Name Expansion): Improve documentation
of the "local part" of a remote file name.
* doc/lispref/processes.texi (Synchronous Processes)
(Asynchronous Processes): State explicitly that program and
file names passed to functions that start remote processes
need to be relative or obtained by 'file-local-name'.
* lisp/files.el (file-local-name):
* lisp/simple.el (start-file-process, process-file): Improve
the documentation of the "local part" of a remote file name,
and its use in APIs that start remote processes.
Paul Eggert [Thu, 27 Dec 2018 20:52:45 +0000 (12:52 -0800)]
Improve accept-process-process doc
* doc/lispref/processes.texi (Accepting Output):
* src/process.c (Faccept_process_output):
Document that (accept-process-output P) can return non-nil
even after P has exited, and that it can return nil even if P
is still running (Bug#33839).
Improve process doc. with respect to handling of large input (Bug#33191)
* src/process.c (Fprocess_send_region, Fprocess_send_string):
Document that process input longer than the process input
buffer may be split into bunches. Remove an outdated
reference to a 500 character split boundary.
* doc/lispref/processes.texi (Asynchronous Processes): Remove
mention of "stray character injections" in PTY processes. See
also the comment about ICANON in src/sysdep.c, function
child_setup_tty.
Alan Mackenzie [Thu, 20 Dec 2018 12:21:16 +0000 (12:21 +0000)]
Check result from c-backward-token-2 to avoid infinite loop
This fixes bug #33784.
* lisp/progmodes/cc-fonts.el (c-get-fontification-context): While moving back
over enclosing parentheses, check that c-backward-token-2 actually moves.
Paul Eggert [Wed, 19 Dec 2018 21:42:21 +0000 (13:42 -0800)]
cl-make-random-state was not copying its arg
Problem reported by Xu Chunyang (Bug#33731).
* lisp/emacs-lisp/cl-extra.el (cl-make-random-state):
Use copy-sequence, not copy-tree, so that the record is copied.
* test/lisp/emacs-lisp/cl-extra-tests.el:
(cl-extra-test-cl-make-random-state): New test.
Glenn Morris [Wed, 19 Dec 2018 20:06:18 +0000 (15:06 -0500)]
Skip a vc-bzr test if run as root
* test/lisp/vc/vc-bzr-tests.el (vc-bzr-test-faulty-bzr-autoloads):
Skip this test when run as root. This works around a race
condition in root-specific code in vc-mode-line when deleting a file.
; Do not merge to master, instead fix vc-mode-line there.
* src/keyboard.c (read_char): Events put into `unread-command-events'
with the form (t . EVENT) should always have the t stripped when read
out.
* test/src/keyboard-tests.el: New tests for `unread-command-events'.
Glenn Morris [Wed, 19 Dec 2018 02:18:31 +0000 (21:18 -0500)]
Restrict downcasing in elisp xref tests (bug#25534)
The tests happen to not fail at the moment because find-library-name
now has an extra feature, find-library--from-load-history, which
happens to do a case-insensitive regexp match; but still it seems
better not to rely on this.
* test/lisp/progmodes/elisp-mode-tests.el (xref--case-insensitive):
New variable.
(xref-elisp-test-run, emacs-test-dir): Only downcase if the
filesystem seems to be case-insensitive.
Rob Browning [Tue, 18 Dec 2018 20:34:01 +0000 (15:34 -0500)]
Avoid test failures if directory name looks like a regexp
Taken from <https://sources.debian.org/patches/emacs/1:26.1+1-1>
* test/lisp/ibuffer-tests.el (ibuffer-filter-inclusion-3):
* test/lisp/net/tramp-tests.el (tramp-test42-remote-load-path):
Regexp-quote file names to avoid failures with directory names
of the form "build/emacs-i87jK3/emacs-26.1+1/...".
Paul Eggert [Mon, 17 Dec 2018 16:11:55 +0000 (08:11 -0800)]
More porting to GCC 8 of --enable-gcc-warnings
Backport from master.
I ran into this when building Emacs 26 with GCC 8 on Fedora 29 x86.
* lwlib/lwlib-Xaw.h (xaw_update_one_value, xaw_popup_menu):
* lwlib/lwlib-Xlw.h (xlw_update_one_value, xlw_pop_instance):
* lwlib/lwlib.h (lw_allow_resizing, lw_set_main_areas) [!USE_MOTIF]:
No longer const.
* src/emacs-module.c: Ignore -Wcast-function-type.
Eli Zaretskii [Fri, 14 Dec 2018 08:43:37 +0000 (10:43 +0200)]
Fix display of line numbers in empty lines beyond EOB
* src/xdisp.c (maybe_produce_line_number): When the current
line is at EOB, use the 'line-number-current-line' face only
on that single line, but not on the rest of empty lines beyond
EOB. (Bug#33732)
Eli Zaretskii [Tue, 11 Dec 2018 10:32:11 +0000 (12:32 +0200)]
Fix redisplay when a window's scroll bar or fringes are changed
* src/window.c (set_window_fringes, set_window_scroll_bars):
Set windows_or_buffers_changed flag to cause immediate
thorough redisplay of a window when scroll bars or fringes are
changed. (Bug#33694)
Alan Mackenzie [Mon, 10 Dec 2018 14:30:40 +0000 (14:30 +0000)]
CC Mode: stop extra parens on expression causing false fontification as type
* lisp/progmodes/cc-fonts.el (c-get-fontification-context): recognize
arithmetic operator followed by several open parentheses, not just one, as not
being an argument list.
Alan Mackenzie [Mon, 10 Dec 2018 12:12:02 +0000 (12:12 +0000)]
CC Mode: stop wrongly recognizing "func(a * 9)" as "pointer to type a"
* lisp/progmodes/cc-engine.el (c-forward-decl-or-cast-1): When testing for an
identifier after "a *", on failure additionally check for a digit, setting a
new flag variable got-number if one is found. In the test for CASE 18, check
this flag.
Raimon Grau [Sun, 9 Dec 2018 23:42:41 +0000 (01:42 +0200)]
Guard occur against an undefined orig-line
; Not to be merged to master
* lisp/replace.el (occur-engine): Avoid inserting the current line if
orig-line is nil. This happens, for example, when reverting an occur
buffer with `list-matching-lines-jump-to-current-line' set to t.
(Bug#33476)
Eli Zaretskii [Sun, 2 Dec 2018 18:04:05 +0000 (20:04 +0200)]
Revert "Revert "Fix infloop in GC mark_kboards""
This reverts commit c418c85617babbe7b63730fefb71e2c87a0141af.
This reinstates the original fix, as it had nothing to do
with the behavior reported in bug#33571, which seems to be
the expected behavior.
Eli Zaretskii [Tue, 27 Nov 2018 08:09:55 +0000 (10:09 +0200)]
Improve documentation of Ediff wordwise commands
* lisp/vc/ediff.el (ediff-windows-wordwise)
(ediff-windows-linewise, ediff-regions-wordwise): Update and
clarify the doc strings.
* doc/misc/ediff.texi (Major Entry Points): Update and clarify
the documentation of 'ediff-windows-wordwise' and
'ediff-regions-wordwise'. See the discussion starting at
https://lists.gnu.org/archive/html/help-gnu-emacs/2018-11/msg00197.html
for the details.
Eli Zaretskii [Mon, 26 Nov 2018 17:31:24 +0000 (19:31 +0200)]
Support Hunspell 1.7.0 in ispell.el
* lisp/textmodes/ispell.el
(ispell-find-hunspell-dictionaries): Invoke Hunspell with an
additional command-line argument, to work around a misfeature
in Hunspell 1.7.0 that prevents it from reporting the loaded
dictionary. (Bug#33493)