Navigation and use of diff buffers had several annoying corner cases
that this patch fixes. These corner cases were largely due to
inconsistent treatment of file headers. Say you have a diff such as
this:
The file headers here are the '---' and '+++' lines. With the point on
such a line, hunk operations would sometimes refer to the next hunk and
sometimes to the previous hunk. Most of the time it would be the
previous hunk, which is not what the user would expect. This patch
consistently treats such headers as the next hunk. So with this patch,
if the point is on the '--- ccc' line, the point is seen as referring to
hunk3.
Specific behaviors this fixes are:
1. It should be possible to place the point in the middle of a diff
buffer, and press M-k repeatedly to kill hunks in the order they appear
in the buffer. With the point on hunk1, M-k M-k would kill hunk1 then
hunk2. With the point on hunk3, it would kill hunk3 then hunk4; this is
fine. However, with the point on hunk2, it'd kill hunk2 then hunk1.
This is fixed by this patch.
2. Similarly, it should be possible to apply hunks in order. Previously
with the point at the start, C-c C-a would apply the hunk1, then move
the point to the first @@ header, and thus C-c C-a would try to apply
the same hunk again.
* lisp/vc/diff-mode.el (diff--wrap-navigation): New function to add better
navigation logic to diff-{hunk,file}-{next,prev}.
(diff-hunk-next, diff-hunk-prev):
(diff-file-next, diff-file-prev): Better navigation logic if
skip-hunk-start is true, which happens when called interactively.
(diff-bounds-of-hunk, diff-find-source-location):
(diff-apply-hunk, diff-current-defun, diff-refine-hunk): Small tweaks to
improve hunk navigation.
Noam Postavsky [Fri, 18 Nov 2016 21:26:53 +0000 (16:26 -0500)]
Upcase Path and ComSpec in process-environment
Since 2016-07-18 "Keep w32 environment settings internal only", the
upcasing of environment variables "Path" and "ComSpec" occured after
initializing process-environment. This meant that Lisp code trying to
override "PATH" environment had no effect (Bug #24956).
* src/w32.c (init_environment): Upcase the "Path" and "ComSpec" entries
in Vprocess_environment.
Philipp Stephani [Tue, 22 Nov 2016 20:32:11 +0000 (21:32 +0100)]
Guard terminal parameter in XTerm mouse mode
It has been observed (in the HTerm terminal emulator) that the
event stored in the 'xterm-mouse-last-down' terminal parameter gets
overwritten during a mouse drag operation, causing Emacs to attempt to
synthesize the non-existing <drag-mouse-0> event. Copy the event into
the terminal parameter to protect against such modifications.
* lisp/xt-mouse.el (xterm-mouse-translate-1): Guard against modification
of input event list.
Reuse already existing lisp symbols for ignore_event (bug#19547).
* lisp/subr.el (while-no-input-ignore-events): Use them instead.
* src/keyboard.c (kbd_buffer_store_buffered_event):
Use help-echo for HELP_EVENT, iconify-frame for ICONIFY_EVENT,
and make-frame-visible for DEICONIFY_EVENT.
(syms_of_keyboard): Remove unneeded symbols.
Allow configuring which event throw-on-input should ignore (bug#19547).
* src/keyboard.c (kbd_buffer_store_buffered_event):
Translate event to corresponding symbol from `while-no-input-ignore-events`
and check them with Fmemq.
(syms_of_keyboard): Declare new lisp variable `while-no-input-ignore-events`
and its symbols.
Paul Eggert [Sat, 26 Nov 2016 08:19:08 +0000 (00:19 -0800)]
Don't access pointers to freed storage in regex.c
Remove __BOUNDED_POINTERS__ code, which does not work with
-fcheck-pointer-bound and which has undefined behavior anyway.
Problem found when trying to port to gcc -fcheck-pointer-bounds.
(This code was removed from glibc and gnulib regex.c many years ago.)
* src/regex.c (ELSE_EXTEND_BUFFER_HIGH_BOUND): Remove.
(EXTEND_BUFFER): Use a more-portable approach that avoids
undefined behavior due to inspecting pointers to freed storage.
Paul Eggert [Sat, 26 Nov 2016 05:24:28 +0000 (21:24 -0800)]
Port build to gcc -fcheck-pointer-bounds
This does not let Emacs run, just build.
* lib-src/etags.c (main):
* lib-src/profile.c (main):
Use return, not exit.
* src/bytecode.c (BYTE_CODE_THREADED) [__CHKP__]:
Do not define, as -fcheck-pointer-bounds is incompatible with taking
addresses of labels.
* src/menu.c (Fx_popup_dialog): Use eassume, not eassert,
to pacify gcc -fcheck-pointer-bounds -Wnull-dereference.
Hong Xu [Fri, 25 Nov 2016 10:51:22 +0000 (12:51 +0200)]
Allow user control of progress messages in cpp.el
* progmodes/cpp.el (cpp-message-min-time-interval): New defcustom.
(cpp-progress-time): Use 'cpp-message-min-time-interval'. Improve
the doc string.
(cpp-highlight-buffer): Use 'cpp-progress-message' instead of
'message' to print messages. (Bug#24861)
Alan Mackenzie [Thu, 24 Nov 2016 20:34:42 +0000 (20:34 +0000)]
Handle correctly an (undocumented) bare mode in hack-local-variables.
lisp/files.el (hack-local-variables-prop-line): When a file's first line
contains only a mode specification without the string "mode:", return the mode
symbol only when `handle-mode' is t.
Dima Kogan [Thu, 24 Nov 2016 02:04:21 +0000 (18:04 -0800)]
Clarify ediff-directories prompt
* lisp/vc/ediff-mult.el (ediff-filegroup-action):
* lisp/vc/ediff.el (ediff-directories,ediff-directory-revisions,
ediff-directories3, ediff-merge-directories,
ediff-merge-directories-with-ancestor, ediff-merge-directory-revisions,
ediff-merge-directory-revisions-with-ancestor): Clarify prompt message for
filename filter in interactive ediff. The new message makes it clear what is
being filtered
Mark Oteiza [Tue, 22 Nov 2016 07:42:47 +0000 (02:42 -0500)]
Make sure elided long buffer names have ellipses added (Bug#24972)
* lisp/ibuffer.el (ibuffer-compile-make-eliding-form): Restore the
string concat, and chop "strvar" less the width of the ellipsis.
(ibuffer-compile-make-substring-form): Add space as padding, to fix
off-by-one in alignment.
Tino Calancha [Tue, 22 Nov 2016 06:23:50 +0000 (15:23 +0900)]
buff-menu: Add command to unmark all buffers
Bind 'U' in buff-menu, bs and electric-buff-menu to commands
to unmark all buffers (Bug#24880).
* lisp/emacs-lisp/tabulated-list.el (tabulated-list-header-overlay-p):
New predicate; return non-nil if tabulated-list has a fake header.
* lisp/buff-menu.el (Buffer-menu-unmark-all-buffers):
New command; remove all flags that use a particular mark from all the lines.
Bind it to 'M-DEL'.
(Buffer-menu-unmark-all):
New command; remove all flags from all the lines. Bind it to 'U'.
(Buffer-menu-marker-char, Buffer-menu-del-char): New variables.
(Buffer-menu-delete, Buffer-menu-mark): Use them.
(Buffer-menu-mode-map): Update menus.
(Buffer-menu-mode): Update mode doc.
* lisp/bs.el (bs-unmark-all, bs-unmark-previous): New commands.
(bs-mode-map): Bind them to 'U' and '<backspace>' respectively.
(bs-mode): Update mode doc.
* lisp/ebuff-menu.el (electric-buffer-menu-mode-map):
Bind Buffer-menu-unmark-all to 'U' and Buffer-menu-unmark-all-buffers
to 'M-DEL'.
(bs--down, bs-down, bs--up, bs-up, bs-unmark-current, bs-mark-current):
Use point instead of cursor in doc string.
(electric-buffer-list): Update mode doc.
* doc/emacs/buffers.texi (Several Buffers): Mention Buffer-menu-unmark-all
and Buffer-menu-unmark-all-buffers.
; * etc/NEWS: Add an entry per each new feature.
Paul Eggert [Mon, 21 Nov 2016 16:42:12 +0000 (08:42 -0800)]
Fix another CANNOT_DUMP problem
Reported by Robert Pluim in:
http://lists.gnu.org/archive/html/emacs-devel/2016-11/msg00468.html
* src/emacs.c (might_dump) [CANNOT_DUMP]: Move enum decl from here ...
* src/lisp.h: ... to here.
Katsumi Yamaoka [Mon, 21 Nov 2016 08:21:27 +0000 (08:21 +0000)]
Don't collect strings existing out of <tr>...</tr>
* lisp/net/shr.el (shr-collect-extra-strings-in-table):
Don't collect strings existing out of <tr>...</tr> to avoid
duplication with what `shr-tag-table' renders.
Mike Kupfer [Mon, 21 Nov 2016 06:10:13 +0000 (06:10 +0000)]
Add a variable to pass additional options to rcvstore
* lisp/gnus/gnus-mh.el (gnus-rcvstore-options): New variable.
(gnus-summary-save-in-folder): Include gnus-rcvstore-options in
the arguments that are passed to rcvstore.
cf. <nntp://news.gmane.org/gmane.emacs.gnus.general/87263>
and followups, i.e., ding mailing list.
Paul Eggert [Mon, 21 Nov 2016 04:55:35 +0000 (20:55 -0800)]
Fix undefined refs on some GNU/Linux hosts
Problem reported by Ken Raeburn in:
http://lists.gnu.org/archive/html/emacs-devel/2016-11/msg00463.html
* src/emacs.c (heap_bss_diff) [CANNOT_DUMP]: Remove, as this is
not needed in the CANNOT_UNDUMP case. All uses removed. This
removes unwanted references to my_endbss and my_endbss_static,
which are not optimized away on some platforms.
Paul Eggert [Mon, 21 Nov 2016 00:57:17 +0000 (16:57 -0800)]
Make CANNOT_DUMP work better on GNU/Linux
Clean up some of the bitrot affecting the CANNOT_DUMP code. This
lets the build succeed again, and fixes the testing framework so
that most test cases now pass. About twenty test cases still
fail, though, and we still have Bug#24974.
* configure.ac (CANNOT_DUMP): Now empty if CANNOT_DUMP.
(SYSTEM_MALLOC): Now true if CANNOT_DUMP. There should no longer
be any point to messing with a private memory allocator unless
Emacs is dumping.
* src/alloc.c (alloc_unexec_pre, alloc_unexec_post, check_pure_size):
* src/image.c (reset_image_types):
* src/lastfile.c (my_endbss, _my_endbss, my_endbss_static):
Do not define if CANNOT_DUMP.
* src/emacs.c (might_dump) [CANNOT_DUMP]: Now always false and local.
(daemon_pipe) [!WINDOWSNT]: Now static.
* test/Makefile.in (mostlyclean): Remove *.tmp files.
(make-test-deps.mk): Elide CANNOT_DUMP chatter.
Eli Zaretskii [Sun, 20 Nov 2016 17:28:37 +0000 (19:28 +0200)]
Fix redrawing non-selected frame after resize on MS-Windows
* src/xdisp.c (redisplay_internal): If all the frames were
successfully updated, reset the "garbaged" flag of each frame, to
make sure it doesn't stay set.
* src/w32term.c (w32_read_socket): Don't clear the frame if it's
"garbaged", since expose_frame won't redraw the foreground then.
(Bug#24642)
Paul Eggert [Sun, 20 Nov 2016 00:00:57 +0000 (16:00 -0800)]
Merge from gnulib
This incorporates:
2016-11-15 sys_time: add gnulib::timeval for C++
2016-11-14 snippet/c++defs: fix real-floating arg functions in C++ mode
2016-11-13 strftime: don't use __THROW
2016-11-12 strftime: tune %q
2016-11-12 Merge strftime.c changes from glibc
2016-11-09 manywarnings: fix -Wno-missing-field-initializers detection
2016-11-05 strftime,strptime: support %q to represent the quarter
The glibc changes in turn incorporate the following strftime.c changes:
2015-10-20 Convert misc function definitions to prototype style
2015-09-26 [BZ #18985] out of range data to strftime() causes segfault
2010-01-09 Add support for XPG7 testing
2009-10-30 Implement Burmese language locale for Myanmar
2008-06-13 [BZ #6612] pass reference to tzset_called around
2007-10-16 [BZ #5184] Add tzset_called argument
* build-aux/snippet/c++defs.h, lib/strftime.c, lib/sys_time.in.h:
* m4/manywarnings.m4: Copy from gnulib.
Paul Eggert [Sat, 19 Nov 2016 22:31:05 +0000 (14:31 -0800)]
Merge from origin/emacs-25
4af5981 Add a comment in generated refcards about the source ef880a5 ; * etc/refcards/calccard.tex: Remove obsolete comment. 4887e7c js-mode: Fix indent problem after a regexp e992ac0 Fix sluggish display of symbols in UTF-8 language environment 1fc101b Don't confuse how Texinfo outputs @var with the input 91aa5d1 * doc/lispref/display.texi (Scroll Bars): * doc/lispref/frame... f758fcd * doc/emacs/cmdargs.texi (Initial Options): Copyedit for --da... 5b0cddd More fixes in copyright notices in etc/refcards/ f994c20 Update copyright text in refcards 9ad2ae7 Fix Outline command names 26c3554 Send text received by bracketed paste to process db0b58d Correct the statement about programming modes always running ... 78aece4 Improve documentation of 'occur' eb364fd Do call debugger on failed cl-assert 3ef4ee8 Avoid infloop in python 8da810f Don't refer to obsolete FEATURE-unload-hook 4f478ca Improve documentation of dabbrevs 7272e5d * lisp/chistory.el (list-command-history): Doc fix. (Bug#24890) 89b7482 * lisp/simple.el (set-mark-command): Doc fix. (Bug#24890) 3b199f7 Improve documentation of some Help commands 93d3a0e Fix documentation of yes-or-no prompts af04919 Fix documentation of partial completion style ed80184 Fix documentation of the mode line on emacsclient frames e6be855 Fix description of 'C-z' in User manual 16f7007 Improve and clarify documentation of Outline Mode 31d93aa Add Emacs version number to nt/README.W32 0b6b815 Fix python-mode hideshow regexp dc152c5 Modernize usage of 'macOS' in doc and comments 84c5343 Prefer comments /* like this */ in C code bb61e50 * doc/lispref/loading.texi (Autoload): Better link (Bug#24845). 3ef86fd Clarify documentation of face attribute functions de51d59 ; * nt/README.W32: Minor copyedits. db436e9 Don't call debug on failed cl-assert
Mark Oteiza [Sat, 19 Nov 2016 18:34:24 +0000 (13:34 -0500)]
Port RefTeX to cl-lib
* lisp/textmodes/reftex-auc.el:
* lisp/textmodes/reftex-cite.el: Use cl-lib.
(reftex-do-citation, reftex-create-bibtex-file): Substitute cl-lib
macros.
* lisp/textmodes/reftex-dcr.el: Use cl-lib.
(reftex-view-regexp-match): Substitute cl-lib macro.
* lisp/textmodes/reftex-global.el: Use cl-lib.
(reftex-find-duplicate-labels, reftex-renumber-simple-labels):
(reftex-translate): Substitute cl-lib macros.
* lisp/textmodes/reftex-index.el: Use cl-lib.
(reftex-index, reftex-index-select-tag, reftex-index-mode-map):
(reftex-index-next-phrase, reftex-index-phrases-info):
(reftex-query-index-phrase): Substitute cl-lib macros.
* lisp/textmodes/reftex-parse.el: Use cl-lib.
(reftex-parse-from-file, reftex-where-am-I, reftex-what-macro):
(reftex-nth-arg, reftex-init-section-numbers, reftex-section-number):
Substitute cl-lib macros.
* lisp/textmodes/reftex-ref.el: Use cl-lib.
(reftex-uniquify-label, reftex-offer-label-menu): Substitute cl-lib
macros.
* lisp/textmodes/reftex-sel.el: Use cl-lib.
(reftex-select-shared-map): Set keymap parent to special-mode-map.
Flatten loop and remove digits and hyphen definitions from the map.
(reftex-select-label-mode-map):
(reftex-select-bib-mode-map): Use cl-lib macro, and flatten other loop.
(reftex-insert-docstruct, reftex-select-unmark): Use cl-lib macros.
* lisp/textmodes/reftex-vars.el (reftex-vref-is-default): Use
cl-pushnew.
* lisp/textmodes/reftex.el: Use cl-lib.
(reftex-docstruct-symbol): Use cl-incf.
(reftex-ref-style-toggle): Replace add-to-list with append.
(reftex-compile-variables): Use cl-lib macros, and functions with
compiler macros. cl-first is just an alias.
(reftex-parse-args, reftex-scanning-info-available-p):
(reftex-select-external-document, reftex-get-file-buffer-force): Use
cl-lib macros.
(reftex-isearch-minor-mode): Replace add-to-list with append.
Eli Zaretskii [Sat, 19 Nov 2016 10:17:23 +0000 (12:17 +0200)]
Implement getrlimit and setrlimit for MS-Windows
* src/w32heap.c (getrlimit, setrlimit): New functions.
Include w32.h.
* src/emacs.c (main): Use 'rlim_t', not 'long', for values that
should be compatible with 'struct rlimit' members.
* nt/inc/sys/resource.h: New header file.
* nt/mingw-cfg.site (ac_cv_func_getrlimit, ac_cv_func_setrlimit):
Set to "yes".
Mark Oteiza [Sat, 19 Nov 2016 00:36:18 +0000 (19:36 -0500)]
Clean up reftex-toc-mode-map
* lisp/textmodes/reftex-toc.el: Remove cl.
(reftex-toc-mode-map): Flatten loop. Remove mapping digit keys and
hyphen, as reftex-toc-mode is derived from special-mode nowadays.
Eli Zaretskii [Fri, 18 Nov 2016 11:02:34 +0000 (13:02 +0200)]
Improve documentation of functions that accept time values
* doc/lispref/os.texi (Time Calculations): Mention the meaning of
'nil' or a scalar number as the time-value argument. Add a
cross-reference to 'float-time' for computing a time difference as
a scalar number of seconds.
* src/editfns.c (Fformat_time_string, Ftime_less_p)
(Ftime_subtract, Ftime_add, Fdecode_time, Fcurrent_time_string)
(Fcurrent_time_zone): Mention in the doc strings the meaning of
nil argument and the fact that a time value can be a scalar number
of seconds since the epoch.
(Ftime_subtract): Mention 'float-time'.
Alex [Fri, 18 Nov 2016 09:02:55 +0000 (11:02 +0200)]
Extend 'indent-relative' when its arg is non-nil
* lisp/indent.el (indent-relative-maybe): New obsolete alias.
(indent-relative-first-indent-point): Renamed from
'indent-relative-maybe'.
(indent-relative): Now accepts an additional optional argument.
The first argument was renamed to FIRST-ONLY. Doc fix. Support
the additional arg. (Bug#24766)
Mark Oteiza [Thu, 17 Nov 2016 23:27:38 +0000 (18:27 -0500)]
Port registry.el the rest of the way to cl-lib
This file already is using cl-lib functions at runtime; eieio ultimately
loads cl-lib, which explains why doing so wasn't an issue.
* lisp/registry.el: Require cl-lib.
(registry-db, registry--match, registry-search, registry-delete):
(registry-insert, registry-reindex): Replace cl macros with cl-lib ones.
(registry-collect-prune-candidates): Replace cl function with cl-lib one.
Eli Zaretskii [Wed, 16 Nov 2016 16:11:33 +0000 (18:11 +0200)]
Fix sluggish display of symbols in UTF-8 language environment
* lisp/international/fontset.el (setup-default-fontset): Make sure
Symbola and FreeMono are set up in the default fontset as belonging
to the "iso10646-1" registry. In the UTF-8 locale, this avoids a
long and futile search for a suitable font, whose side effect is a
lot of consing, which then hits the font-cache compacting issue,
and causes very sluggish redisplay of characters displayed by
these fonts. All this happens because the default for the
registry is "iso8859-1". (Bug#24953)
Glenn Morris [Wed, 16 Nov 2016 07:28:47 +0000 (23:28 -0800)]
Add --new-daemon, which runs in the foreground and does not fork
This is intended for modern init systems such as systemd,
which manage many of the traditional aspects of daemon behavior
themselves. (Bug#2677)
* src/emacs.c (daemon_type): New integer.
(usage, standard_args): Add --old-daemon and --new-daemon.
(main): Handle --old-daemon and --new-daemon arguments.
Restrict all the forking and complicated daemon stuff to old-daemon.
(Fdaemon_initialized): Handle new-style daemon.
* src/lisp.h (IS_DAEMON, DAEMON_RUNNING) [!WINDOWNT]:
Replace daemon_pipe with daemon_type.
* doc/emacs/cmdargs.texi (Initial Options):
* doc/emacs/glossary.texi (Glossary):
* doc/emacs/misc.texi (Emacs Server):
* doc/lispref/display.texi (Window Systems):
* doc/lispref/os.texi (Startup Summary): Related doc updates.
* etc/NEWS: Mention this.
* etc/emacs.service: Use Type=simple and --new-daemon.
Glenn Morris [Wed, 16 Nov 2016 00:18:28 +0000 (19:18 -0500)]
Don't confuse how Texinfo outputs @var with the input
* doc/emacs/rmail.texi (Movemail):
* doc/lispref/control.texi (Pattern matching case statement):
* doc/lispref/frames.texi (Size and Position):
* doc/lispref/processes.texi (Asynchronous Processes):
* doc/lispref/text.texi (Document Object Model):
* doc/lispref/windows.texi (Coordinates and Windows):
Do not upper-case the argument of @var.
Ken Brown [Mon, 14 Nov 2016 22:26:12 +0000 (17:26 -0500)]
Simplify case-insensitivity checks on Mac OS X
* src/fileio.c (file_name_case_insensitive_p): Try skipping the
Darwin code and instead using pathconf with _PC_CASE_SENSITIVE.
Leave in two alternatives conditionally compiled based on
DARWIN_OS_CASE_SENSITIVE_FIXME in case pathconf doesn't work.
* etc/PROBLEMS: Mention the possible problem with pathconf on
Mac OS X.
Simen Heggestøyl [Tue, 15 Nov 2016 18:08:22 +0000 (19:08 +0100)]
Complete the name of PostgreSQL databases
* lisp/progmodes/sql.el (sql-postgres-login-params): Complete database
name.
(sql-postgres-list-databases): New function returning a list of
available PostgreSQL databases.
* test/lisp/progmodes/sql-tests.el: New file with tests for sql.el.
Wilson Snyder [Mon, 14 Nov 2016 18:47:31 +0000 (13:47 -0500)]
Update verilog-mode.el
* verilog-mode.el (verilog-read-decls, verilog-calc-1): Fix
"default clocking" indentation and preventing AUTOs from working,
bug1084. Reported by Alan Morgan.
(verilog-diff-report): Fix `verilog-diff-report'
not returning bad status on differences, bug1087. Reported by
Eric Jackowski.
(verilog-auto-inst-param-value)
(verilog-auto-inst-param-value-type, verilog-read-sub-decls)
(verilog-read-sub-decls-expr, verilog-read-sub-decls-gate)
(verilog-read-sub-decls-line, verilog-read-sub-decls-sig)
(verilog-read-sub-decls-type): When
`verilog-auto-inst-param-value-type' is set, which is now the
default, AUTOINPUT etc will now substitute parameter types from
submodules, bug1061. Reported by Brad Dobbie.
(verilog-auto-reset, verilog-backward-case-item)
(verilog-extended-case-re, verilog-read-always-signals-recurse):
Fix indentation of randcase, bug1072. Reported by David Rogoff.
(verilog-read-sub-decls-expr)
(verilog-sig-multidim-string): Fix AUTOINST ordering of dimensions
in generated comments, bug1057. Reported by Kaushal Modi.
(verilog-auto-wire-comment, verilog-insert-definition):
Add `verilog-auto-wire-comment' to suppress wire comments. Reported by
Eric Jackowski.
(verilog-extended-complete-re): Fix indentation
of class static functions, bug1053. Reported by Gregory
Czajkowski.
(verilog-module-filenames): Support tramp for
finding verilog modules. Reported by Nevada Sanchez.
Paul Eggert [Mon, 14 Nov 2016 17:08:06 +0000 (09:08 -0800)]
Improve case-insensitive checks (Bug#24441)
* doc/lispref/files.texi (Truenames): Simplify documentation,
to avoid giving too much platform-specific information that
may not be accurate anyway.
* src/fileio.c (file_name_case_insensitive_p): Use pathconf with
_PC_CASE_SENSITIVE if _PC_CASE_INSENSITIVE is not available.
Otherwise if one approach fails (e.g., with errno == EINVAL), fall
back on an alternative rather than returning false. Try skipping
the Darwin code, as it (1) no longer seems to be needed and (2)
does not seem to match the Apple documentation. Leave in two
alternatives conditionally compiled based on
DARWIN_OS_CASE_SENSITIVE_FIXME in case (1) or (2) is incorrect.
projects / emacs.git / log