Eli Zaretskii [Mon, 30 Mar 2020 13:57:43 +0000 (16:57 +0300)]
Avoid assertion violation at startup in pdumped Emacs
* src/buffer.c (init_buffer) [USE_MMAP_FOR_BUFFERS]: Remove the
assertion that ensured all buffers in pdumped Emacs have non-NULL
pointer to buffer text. That was false when Emacs was pdumped
with killed buffer(s) in the all_buffers linked list. See
https://lists.gnu.org/archive/html/emacs-devel/2020-03/msg00800.html
for more details.
Juri Linkov [Sun, 29 Mar 2020 22:34:47 +0000 (01:34 +0300)]
Support state changing VC operations on directories in Dired (bug#34949)
* lisp/dired-aux.el (dired-vc-next-action): New command.
(dired-vc-deduce-fileset): Rename from vc-dired-deduce-fileset in vc.el.
* lisp/dired.el (dired-mode-map): Remap vc-next-action to
dired-vc-next-action.
* lisp/vc/vc-dir.el (vc-dir-mark-files): New function.
(vc-dir-refresh): Run hook vc-dir-refresh-hook.
* lisp/vc/vc.el (vc-deduce-fileset): Rename arg 'observer' to
'not-state-changing' and document it in docstring.
(vc-dired-deduce-fileset): Rename to dired-vc-deduce-fileset in dired-aux.el.
* lisp/cedet/ede.el (ede-turn-on-hook, ede-minor-mode):
* lisp/desktop.el (desktop-minor-mode-table): Rename the long ago
obsolete vc-dired-mode to vc-dir-mode.
Juri Linkov [Sat, 28 Mar 2020 23:41:29 +0000 (01:41 +0200)]
Switch to literal mode with message when regexp is too big in char-fold search
* lisp/char-fold.el (char-fold-to-regexp): Don't use regexp-quote
when the length of regexp reaches 5000. (Bug#40216)
* lisp/isearch.el (isearch-search): On big regexp in char-fold mode
gracefully fall back to literal mode, try to search again and display
momentary-message about switching to literal mode.
(isearch--momentary-message): Add optional arg SECONDS.
Paul Eggert [Sat, 28 Mar 2020 20:58:44 +0000 (13:58 -0700)]
Stop using newly-deprecated dosname Gnulib module
Code is supposed to use the filename module now.
* admin/merge-gnulib (GNULIB_MODULES): Replace dosname with filename.
* lib/dosname.h: Remove this forwarding stub.
* lib/gnulib.mk.in, m4/gnulib-comp.m4: Regenerate.
* lib-src/emacsclient.c, src/fileio.c:
Include filename.h instead of dosname.h.
Paul Eggert [Sat, 28 Mar 2020 20:50:02 +0000 (13:50 -0700)]
Update from Gnulib
This incorporates:
2020-03-28 Use module 'filename' instead of module 'dosname'
2020-03-28 dosname: Redirect to 'filename'
* lib/at-func.c, lib/canonicalize-lgpl.c, lib/dosname.h:
Copy from Gnulib.
* lib/gnulib.mk.in, m4/gnulib-comp.m4: Regenerate.
* lib/filename.h: New file, copied from Gnulib.
Stefan Monnier [Sat, 28 Mar 2020 14:16:58 +0000 (10:16 -0400)]
* lisp/jit-lock.el (jit-lock-mode): Pass `local` to add-hook
The old code used local=nil knowing that add-hook would affect the
local part only anyway. Remove this hideous assumption.
Remove redundant `:group` args while we're at it.
Used to be called `ebrowse-tags-loop-form` and passed to `eval`.
Now it's passed to `apply` instead, which is better for karma.
(ebrowse-tags-loop-continue, ebrowse-tags-search)
(ebrowse-tags-query-replace, ebrowse-tags-search-member-use):
Adjust accordingly.
(ebrowse-electric-position-mode-map): Move init into declaration.
(ebrowse-electric-position-mode): Derive from special.
Michael Albinus [Sat, 28 Mar 2020 11:59:03 +0000 (12:59 +0100)]
Tramp cache fixes
* lisp/net/tramp-adb.el (tramp-adb-handle-write-region):
* lisp/net/tramp-smb.el (tramp-smb-handle-write-region): Flush the
cache after the file has been written.
Stefan Monnier [Fri, 27 Mar 2020 20:38:52 +0000 (16:38 -0400)]
* lisp/gnus/gnus-registry.el: Use lexical-binding
(gnus-registry-install-shortcuts): Use a closure (with dynamic :documentation)
(gnus-registry-user-format-function-M): Use define-obsolete-function-alias.
(gnus-registry-article-marks-to-names): η-reduce.
Stefan Monnier [Fri, 27 Mar 2020 16:54:52 +0000 (12:54 -0400)]
* lisp/textmodes/tex-mode.el: Replace double-definition hack with an advice
(tex-verbatim-environments): Add "Verbatim".
(tex--guess-mode): Rename from tex-guess-mode and return the mode
rather than calling it.
(tex-mode): Replace second definition with an advice.
Stefan Monnier [Fri, 27 Mar 2020 16:24:19 +0000 (12:24 -0400)]
* lisp/progmodes/ebrowse.el: Prefer hash-tables to obarrays
Remove redundant :group args.
Use `defvar-local` and `setq-local` where possible.
(ebrowse-some): Use seq-some instead.
(ebrowse-every): Use seq-every-p instead.
(ebrowse-position): Use seq-position.
(ebrowse--tree-table): Rename from `ebrowse--tree-obarray`.
Change all users to use a hash-table rather than an obarray.
(ebrowse-for-all-trees): Adjust to the table being a hash-table.
(ebrowse-tree-table-as-alist): Rename from `ebrowse-tree-obarray-as-alist`.
(ebrowse-build-tree-obarray): Rename from `ebrowse-build-tree-obarray`.
(ebrowse-tree-mode): Remove redundant setting of `ebrowse--tree-obarray`.
(ebrowse-set-tree-indentation, ebrowse-view-file-other-frame)
(ebrowse-last-completion-table): Rename from
ebrowse-last-completion-obarray.
(ebrowse-position): Make it a proper struct.
Eli Zaretskii [Fri, 27 Mar 2020 12:43:20 +0000 (15:43 +0300)]
Port the 'module/async-pipe' test to MS-Windows
These changes let the code compile and produce a valid DLL, but the
test hangs. It looks like the hang is in Fdelete_process, when it
closes one of the descriptors of the pipe process.
In addition, this use of the pipe process cannot currently work
on MS-Windows, since make-pipe-process doesn't set up the reader
thread to read from the Emacs's side of the pipe, so the select
emulation doesn't know there's stuff to read from that pipe.
* test/data/emacs-module/mod-test.c [WINDOWSNT]: Include
windows.h.
(ALIGN_STACK) [!__x86_64__]: Define for 32-bit builds.
(sleep_for_half_second): New function.
(write_to_pipe): Declare return type differently for WINDOWSNT.
Call sleep_for_half_second.
(Fmod_test_async_pipe) [WINDOWSNT]: Use _beginthread as substitute
for pthread_create.
(invalid_finalizer): Replace non_ASCII character in a comment.
* test/src/emacs-module-tests.el (module/async-pipe): Skip on
MS-Windows, as the test fails and then hangs.
Paul Eggert [Fri, 27 Mar 2020 07:58:31 +0000 (00:58 -0700)]
Treat out-of-range positions consistently
If a position argument to get-byte etc. is an out-of-range integer,
treat it the same regardless of whether it is a fixnum or a bignum.
* src/buffer.c (fix_position): New function.
* src/buffer.c (validate_region):
* src/character.c (Fget_byte):
* src/coding.c (Ffind_coding_systems_region_internal)
(Fcheck_coding_systems_region):
* src/composite.c (Ffind_composition_internal):
* src/editfns.c (Fposition_bytes, Fchar_after, Fchar_before)
(Finsert_buffer_substring, Fcompare_buffer_substrings)
(Fnarrow_to_region):
* src/fns.c (Fsecure_hash_algorithms):
* src/font.c (Finternal_char_font, Ffont_at):
* src/fringe.c (Ffringe_bitmaps_at_pos):
* src/search.c (search_command):
* src/textprop.c (get_char_property_and_overlay):
* src/window.c (Fpos_visible_in_window_p):
* src/xdisp.c (Fwindow_text_pixel_size):
Use it instead of CHECK_FIXNUM_COERCE_MARKER, so that
the code is simpler and treats bignums consistently with fixnums.
* src/buffer.h (CHECK_FIXNUM_COERCE_MARKER): Define here
rather than in lisp.h, and reimplement in terms of fix_position
so that it treats bignums consistently with fixnums.
* src/lisp.h (CHECK_FIXNUM_COERCE_MARKER): Move to buffer.h.
* src/textprop.c (validate_interval_range): Signal with original
bounds rather than modified ones.
Paul Eggert [Thu, 26 Mar 2020 23:12:21 +0000 (16:12 -0700)]
Refactor and fix typo in CHECK_*_COERCE_MARKER
* src/data.c (check_integer_coerce_marker)
(check_number_coerce_marker): New functions.
Also, fix a typo in the former, by having it use
Qinteger_or_marker_p not Qnumber_or_marker_p.
(arithcompare, floatop_arith_driver, bignum_arith_driver)
(arith_driver, Fplus, Fminus, Ftimes, Fquo, Frem, Fmod)
(minmax_driver, Flogand, Flogior, Flogxor, Fadd1, Fsub1):
Use them in place of the similarly-named macros.
* src/lisp.h (CHECK_NUMBER_COERCE_MARKER)
(CHECK_INTEGER_COERCE_MARKER): Remove; no longer used.
Philipp Stephani [Thu, 26 Mar 2020 16:22:25 +0000 (17:22 +0100)]
Add a module function to open a file descriptor connected to a pipe.
A common complaint about the module API is that modules can't
communicate asynchronously with Emacs. While it isn't possible to
call arbitrary Emacs functions asynchronously, writing to a pipe
should always be fine and is a pretty low-hanging fruit.
This patch implements a function that adapts an existing pipe
process. That way, users can use familiar tools like process filters
or 'accept-process-output'.
* src/emacs-module.c (module_open_channel): Provide definition for
'open_channel'.
(initialize_environment): Use it.
* src/process.c (open_channel_for_module): New helper function.
(syms_of_process): Define necessary symbol.
* test/src/emacs-module-tests.el (module/async-pipe): New unit test.
* test/data/emacs-module/mod-test.c (signal_system_error): New helper
function.
(signal_errno): Use it.
(write_to_pipe): New function running in the background.
(Fmod_test_async_pipe): New test module function.
(emacs_module_init): Export it.
* doc/lispref/internals.texi (Module Misc): Document new module
function.
* doc/lispref/processes.texi (Asynchronous Processes): New anchor
for pipe processes.
Paul Eggert [Thu, 26 Mar 2020 00:40:57 +0000 (17:40 -0700)]
Fix integer overflow in forward-point
* lisp/subr.el (forward-point): Rewrite in Lisp and move here ...
* src/cmds.c (Fforward_point): ... from here. This fixes an
integer overflow bug with (forward-point most-positive-fixnum).
Make compilation-mode regexp matching case-sensitive (bug#40119)
The number of regexps is large, they are written independently of one
another, and they frequently intersect. Using case-sensitive matching
improves separation and performance, and is probably what everyone
have being assuming was used by compilation-mode all along.
* lisp/progmodes/compile.el (compilation-error-case-fold-search): New.
(compilation-parse-errors): Bind case-fold-search to
compilation-error-case-fold-search during matching.
* etc/NEWS: Announce.
Paul Eggert [Wed, 25 Mar 2020 20:39:21 +0000 (13:39 -0700)]
Update from gnulib
This incorporates:
2020-03-25 getopt-posix: port __GETOPT_PREFIX to macOS
2020-03-22 acl-permissions: Improve autoconf macro
* lib/getopt-pfx-core.h, m4/acl.m4: Copy from Gnulib.
Stefan Monnier [Wed, 25 Mar 2020 18:09:48 +0000 (14:09 -0400)]
* lisp/textmodes/conf-mode.el (conf-mode): Fix last change
`delay-mode-hooks` cannot be tested from within `define-derived-mode`
because it's always non-nil in there, so arrange to test it before we
enter the body.
Eli Zaretskii [Wed, 25 Mar 2020 14:18:37 +0000 (16:18 +0200)]
Improve the UI of 'list-timers'
* lisp/emacs-lisp/timer-list.el (list-timers): Display both "Next"
and "Repeat" in units of seconds, for consistency.
(timer-list-mode): Add help-echo to column headers.
Juri Linkov [Tue, 24 Mar 2020 21:58:01 +0000 (23:58 +0200)]
Rename dired-mark-region choices and ignore empty region.
* lisp/dired.el (dired-mark-region): Rename choices
'exclusive' to 'file', and 'inclusive' to 'line'.
(dired-mark-if, dired-mark): Check for non-empty region explicitly
instead of using use-region-p to ignore non-nil value of
use-empty-active-region. (Bug#39902)
Philip K [Tue, 17 Mar 2020 14:29:53 +0000 (15:29 +0100)]
Add support for multiple Gravatar services
Now supports Libravatar and Unicornify, next to Gravatar (Bug#39965).
* lisp/image/gravatar.el (gravatar-base-url): Remove constant.
(gravatar-service-alist): List supported services.
(gravatar-service): Add user option to specify service, defaults to
Libravatar.
(gravatar--service-libravatar): New function, libravatar image host
resolver implementation.
(gravatar-build-url): Use alist gravatar-service-alist instead of
gravatar-base-url.
* etc/NEWS: Mention new gravatar service option.
Don't add repeated xlmns:xlink declarations in svg-create
* lisp/svg.el (svg-create): Fix previous unconditional addition of
the xmlns:xlink declaration -- callers may already add one, and
having it twice is something most svg libraries doesn't like.
Don't add repeated xlmns:xlink declarations in svg-create
* lisp/svg.el (svg-create): Fix previous unconditional addition of
the xmlns:xlink declaration -- callers may already add one, and
having it twice is something most svg libraries doesn't like.
Eli Zaretskii [Mon, 23 Mar 2020 14:37:27 +0000 (16:37 +0200)]
Don't build the Gnulib 'utimens' module on MinGW
* nt/gnulib-cfg.mk (OMIT_GNULIB_MODULE_utimens): Omit the
'utimens' module in the MinGW build: 'utimens' is not used by
Emacs, and 'fdutimens' is implemented in w32.c.
Noam Postavsky [Fri, 20 Mar 2020 08:07:39 +0000 (04:07 -0400)]
Don't signal during backtrace unrewind (Bug#40088)
backtrace_eval_unrewind is used to temporarily reverse
let-bindings (it's called with a positive argument to reverse
bindings, and then a negative argument to re-apply them) by
backtrace--locals and backtrace-eval. For the SPECPDL_LET_DEFAULT and
SPECPDL_LET_LOCAL cases (which occur for let-bindings on buffer-local
variables), the code calls Fdefault_value and Fbuffer_local_value on
the symbol.
For symbols which are unbound at top-level, the first (with positive
argument) call to backtrace_eval_unrewind will set the symbol's value
to unbound (putting the current value in the specpdl's "old value"
slot). On the second (with negative argument) call,
backtrace_eval_unrewind attempts to retrieve the symbol's value with
Fdefault_value or Fbuffer_local_value, but that raises a void-variable
signal. This interrupts the restoration of the let-bindings, so any
other variables more recent on the stack will now have the wrong
value.
* src/data.c (default_value): Make non-static.
* src/lisp.h: Declare it.
* src/eval.c (backtrace_eval_unrewind): Replace the calls to
Fdefault_value and Fbuffer_local_value with default_value and
buffer_local_value, respectively. The latter do exactly the same as
the former, except if the symbol's value is Qunbound they just return
it instead of signaling void-variable.
Noam Postavsky [Fri, 20 Mar 2020 10:00:11 +0000 (06:00 -0400)]
Fix a couple of problems in changelog generating functions
* lisp/vc/diff-mode.el (diff-add-log-current-defuns): If there is a
scan-error when calling end-of-defun, go to end of hunk. This can
easily happen since we are calling end-of-defun on a partial code
fragment from a diff.
* lisp/vc/log-edit.el (log-edit-generate-changelog-from-diff): Bind
display-buffer-overriding-action around the log-edit-show-diff call
only. Otherwise, it can affect, for example, debugger windows
triggered by the diff-add-log-current-defuns call.
Noam Postavsky [Sun, 22 Mar 2020 11:48:14 +0000 (07:48 -0400)]
Fix cl-concatenate (Bug#40180)
* lisp/emacs-lisp/cl-extra.el (cl-concatenate): Use apply, to avoid
adding extra nesting of args.
* test/lisp/emacs-lisp/cl-extra-tests.el (cl-concatenate): New test.
Eli Zaretskii [Sat, 21 Mar 2020 11:26:19 +0000 (13:26 +0200)]
Improve documentation of project.el commands
* lisp/progmodes/project.el (project-find-regexp): Require 'grep'
to be able to call 'grep-read-files'.
(project-search, project-query-replace-regexp): Doc fixes.
* doc/emacs/maintaining.texi (Projects): New section.
* doc/emacs/emacs.texi (Top): Add "Projects" to the detailed menu.
* lisp/vc/ediff-init.el (ediff-temp-file-prefix): Redefine as
obsolete variable alias for 'temporary-file-directory'.
* lisp/vc/ediff-util.el (ediff-make-temp-file): Don't use obsolete
variable name.
Stefan Kangas [Sat, 21 Mar 2020 01:47:08 +0000 (02:47 +0100)]
Declare some <package>-version variables obsolete
These are not used for anything these days and can therefore be
removed. Package developers should check the Emacs version instead.
Ref: https://lists.gnu.org/r/emacs-devel/2020-03/msg00080.html
Juri Linkov [Thu, 19 Mar 2020 23:21:57 +0000 (01:21 +0200)]
* lisp/tab-line.el (tab-line-new-button-show): New defcustom.
* lisp/tab-line.el (tab-line-format-template): Use tab-line-new-button-show.
(tab-line-new-tab-choice): Remove choice 'nil' with "No button".
(tab-line-separator): Add docstring.
* lisp/image/exif.el (exif--direct-ascii-value): New function
(bug#40127).
(exif--parse-directory): Use it to get the correct values for
in-directory (i.e., shorter than 4 octets) strings.
* lisp/image/exif.el (exif--direct-ascii-value): New function
(bug#40127).
(exif--parse-directory): Use it to get the correct values for
in-directory (i.e., shorter than 4 octets) strings.
Eli Zaretskii [Thu, 19 Mar 2020 14:57:35 +0000 (16:57 +0200)]
Fix display of :box face when overlay strings are around
* src/xdisp.c (reset_box_start_end_flags): New function.
(handle_face_prop): Only set the start_of_box_run_p flag, don't
reset it.
(pop_it): Set the face_box_p flag, if the popped face requires
that, when continuing iteration over buffer text.
(get_next_display_element, next_element_from_display_vector): Only
set the end_of_box_run_p flag, never reset it here.
(set_iterator_to_next): Don't reset the start_of_box_run_p and
end_of_box_run_p flags here. They are now reset as side effect of
PRODUCE_GLYPHS.
(append_space_for_newline): Restore the end_of_box_run_p flag
after PRODUCE_GLYPHS where we previously didn't reset it.
* src/dispextern.h (PRODUCE_GLYPHS): Call
reset_box_start_end_flags after producing glyphs.
(Bug#40124)
Paul Eggert [Wed, 18 Mar 2020 21:48:52 +0000 (14:48 -0700)]
Omit timestamp optimization invalid on 387 FPU
* src/timefns.c (frac_to_double): Omit optimization that is
invalid on machines with excess precision (e.g., gcc x86 with 387
FPU), because it double-rounds. Found via ‘gcc -m32’ on x86-64.
Glenn Morris [Wed, 18 Mar 2020 14:50:37 +0000 (07:50 -0700)]
Merge from origin/emacs-27
e92b8e535a (origin/emacs-27) Remove raw carriage return characters fr... 5747a59a88 Recalculate default font when switching font backend a2dd8c4234 * lisp/tab-line.el: Fix tab-line-format and tab-line-forma... a7b8291b6c * etc/NEWS: Make the `--eval` example slightly more precise f8254aad14 * lisp/image/image-converter.el: Support more ImageMagick ... 5beb269505 Support Unicode 13.0 3a671ad7ed Fix regression in wisent-total-conflicts
Don't generate useless range table entries for ASCII chars
In multibyte regexps, each ASCII char or range in a character
alternative produces a nonsense range table entry in addition to the
correct bits in the ASCII bitmap. Those entries do not match anything
but waste space and time.
Robert Pluim [Mon, 16 Mar 2020 16:21:02 +0000 (17:21 +0100)]
Recalculate default font when switching font backend
This is an updated version of the patch by Dmitry Antipov
<dmantipov@yandex.ru> in
<https://debbugs.gnu.org/cgi/bugreport.cgi?bug=23386#43>.
Fixes Bug#23386
* src/dispextern.h (struct redisplay_interface): New member
default_font_parameter.
* src/xterm.h: Add prototype for x_default_font_parameter.
* src/xterm.c (x_redisplay_interface): Initialize
default_font_parameter member.
* src/xfns.c (x_default_font_parameter): Make non-static.
* src/w32term.h: Add prototype for w32_default_font_parameter
* src/w32fns.c (w32_default_font_parameter): Make non-static.
* src/w32term.c (w32_redisplay_interface): Initialize
default_font_parameter member.
* src/nsterm.m (ns_redisplay_interface): Add dummy
ns_default_font_parameter (there is currently only one possible font
backend on macOS). Initialize default_font_parameter member.
* src/frame.c (gui_set_font_backend): Recalculate default font using
RIF default_font_parameter to avoid crash when changing font backend.
Juri Linkov [Mon, 16 Mar 2020 23:11:27 +0000 (01:11 +0200)]
* lisp/tab-line.el: Fix tab-line-format and tab-line-format-template.
* lisp/tab-line.el (tab-line-format): Use buffer-name in the cache key
instead of just buffer object to invalidate the cache on buffer renaming.
(tab-line-format-template): Always leave the separator after the last tab
like tab-bar.el already does.
Yuan Fu [Sun, 15 Mar 2020 15:47:43 +0000 (16:47 +0100)]
Add store/restore window configuration feature for gdb-mi
Add a feature that allows a user to save a gdb window
configuration (window layout) to a file with
'gdb-save-window-configuration' and load it back with
'gdb-load-window-configuration'. Set a default window configuration
by setting 'gdb-default-window-configuration-file'.
Add an option to make gdb preserve the window configuration
that the user had before starting gdb. In window.el, add
'with-window-non-dedicated'.
* lisp/progmodes/gdb-mi.el (top/level): Require 'pcase' and 'cl-seq'.
(gdb--window-configuration-before): New variable.
(gdb-restore-window-configuration-after-quit): New option.
(gdb-window-configuration-directory,
gdb-default-window-configuration-file): New variables.
(gdb): Save configuration on startup.
(gud-menu-map): Add "Load Layout" and "Save Layout" to menu. Add
"Restore Layout After Quit" button to menu. Rename "Restore Window
Layout" to "Restore Default Layout", add some help echo, and move it
from "GDB-MI" menu to "GDB-WINDOWs" menu.
(gdb-toggle-restore-window-configuration): New function.
(gdb-get-source-buffer): New function, extracted out of
'gdb-restore-window'.
(gdb-setup-windows): Add a condition branch that loads default window
configuration when available. Fix docstring.
(gdb-buffer-p, gdb-function-buffer-p, gdb--buffer-type,
gdb-save-window-configuration, gdb-load-window-configuration): New
functions.
(gdb-restore-windows): Edit docstring to mention
'gdb-default-window-configuration-file'.
(gdb-reset): Restore window configuration after quit.
* lisp/window.el (with-window-non-dedicated): New macro.
Eli Zaretskii [Sun, 15 Mar 2020 14:47:54 +0000 (16:47 +0200)]
Reverse the meaning of 2nd arg to 'live_buffer_holding'
* src/alloc.c (live_buffer_holding): Rename ALL_BUFFERS ti
IGNORE_KILLED, and reverse the condition for returning killed
buffers.
(live_buffer_p): Add commentary.
(live_buffer_p, mark_maybe_object, mark_maybe_pointer): Reverse
the 2nd argument to live_buffer_holding. (Bug#39962)
Juri Linkov [Sat, 14 Mar 2020 23:58:58 +0000 (01:58 +0200)]
* lisp/image/image-converter.el: Support more ImageMagick versions (bug#39994)
* lisp/image/image-converter.el (image-converter--probe): Match
a possible additional second column in some ImageMagick versions
of "convert -list format".
projects / emacs.git / log