Paul Eggert [Sun, 12 Jun 2016 17:31:25 +0000 (10:31 -0700)]
New macro GNUC_PREREQ for GCC version
* src/conf_post.h (GNUC_PREREQ): New macro.
Change uses of __GNUC_MINOR__ and __GNUC_PATCHLEVEL__ to use this
macro instead, for clarity and consistency.
(PRINTF_ARCHETYPE): New macro. Define it to __gnu_printf__ only
if glibc, since non-GNU platforms don’t necessarily support GNU
printf formats.
(ATTRIBUTE_FORMAT_PRINTF): Use it.
Eli Zaretskii [Sun, 12 Jun 2016 08:30:14 +0000 (11:30 +0300)]
Fix printf-related compilation warnings on MinGW
* src/conf_post.h (ATTRIBUTE_FORMAT_PRINTF) [__MINGW32__]: Use
'__ms_printf__', not '__gnu_printf__', as the latter is not what
MS 'printf' supports, and causes bogus compilation warnings.
* src/lisp.h (pI) [__MINGW32__]: Define to "I64", as MS 'printf'
doesn't support the "ll" modifier.
Paul Eggert [Sun, 12 Jun 2016 06:48:13 +0000 (23:48 -0700)]
emacs_strerror cleanups
* src/buffer.c, src/emacs.c, src/lread.c: Don’t include coding.h;
no longer needed, now that emacs_strerror is declared by lisp.h.
* src/coding.c (emacs_strerror): Remove; moved to emacs.c.
* src/coding.h (emacs_strerror) [emacs]: Remove decl; moved
to lisp.h.
* src/emacs.c (emacs_strerror): Move here from coding.c. Do not
convert result string; this is now the caller’s responsibility,
as some need conversion and others don’t.
* src/fileio.c (report_file_errno, report_file_notify_error):
Use emacs_strerror rather than rolling it ourselves.
* src/lisp.h (emacs_strerror): Move decl here from coding.h.
* src/lread.c (dir_warning): Just call emacs_strerror rather than
both strerror and emacs_strerror. Convert its result from
locale-coding-system, since it no longer does that conversion.
* src/sound.c (sound_perror):
* src/sysdep.c (emacs_perror, str_collate):
Use emacs_strerror, not strerror.
Ken Brown [Sat, 11 Jun 2016 12:31:27 +0000 (08:31 -0400)]
Fix dbus crash on 32-bit Cygwin
* src/dbusbind.c (Fdbus__init_bus): Use make_save_pointer to store
connection address. (Bug#23741)
(xd_lisp_dbus_to_dbus): Use XSAVE_POINTER to retrieve connection
address.
Paul Eggert [Sat, 11 Jun 2016 00:18:24 +0000 (17:18 -0700)]
Catch malloc_get_state, malloc_set_state failure
This should help insulate Emacs better from configuration screwups.
Future versions of the GNU C library are planned to deprecate
these functions, but will continue to support them in
already-built-and-dumped Emacs executables.
* src/alloc.c (malloc_initialize_hook) [DOUG_LEA_MALLOC]:
Abort if malloc_set_state fails.
(alloc_unexec_pre) [DOUG_LEA_MALLOC]:
Report malloc_get_state failure, and exit.
Jules Tamagnan [Fri, 10 Jun 2016 09:08:29 +0000 (12:08 +0300)]
Fix eldoc-related freezes in python mode
* lisp/progmodes/python.el (python-eldoc-get-doc): New defvar.
(python-eldoc-function-timeout)
(python-eldoc-function-timeout-permanent): New defcustoms.
(python-eldoc-function): If python-eldoc--get-doc-at-point times
out, effectively turn off ElDoc in current buffer. (Bug#23609)
Eli Zaretskii [Fri, 10 Jun 2016 08:57:17 +0000 (11:57 +0300)]
Show returned value after gdb-mi "finish" command
* lisp/progmodes/gdb-mi.el (gdb-stopped): Display the expected
"Value returned" message in response to "finish", when not
produced by GDB/MI. (Bug#23720)
May address an issue where Emacs consumed large amounts of CPU
because of neverending toolbar updating (which was caused
by, but also called this function).
Martin Rudalics [Fri, 10 Jun 2016 06:21:33 +0000 (08:21 +0200)]
Revert "Bind `widget-button-click' to mouse-1/-2 instead of down-mouse-1/-2"
* lisp/wid-edit.el (widget-keymap): Bind `widget-button-click'
to mouse-1/-2 instead of down-mouse-1/-2. Suggested by Stefan
Monnier. (Bug#19185, Bug#20398)
Paul Eggert [Fri, 10 Jun 2016 06:11:40 +0000 (23:11 -0700)]
Prefer getsockopt to getpeername on non-MS-Windows
* admin/CPP-DEFINES: Remove HAVE_GETPEERNAME.
* configure.ac: Don’t check for getpeername.
* src/process.c (wait_reading_process_output) [!WINDOWSNT]:
Use getsockopt even if not GNU/Linux, as the platforms where
getsockopt used to hang are no doubt long dead.
Paul Eggert [Fri, 10 Jun 2016 04:58:16 +0000 (21:58 -0700)]
Fix XFASTINT of non-fixnum in process status
* src/process.c (decode_status): 3rd arg is now Lisp_Object *,
not int *, and is not decoded. All uses changed.
(status_message): Do not assume ‘failed’ code is an integer.
* src/process.h: Document codes better.
Glenn Morris [Fri, 10 Jun 2016 04:55:02 +0000 (00:55 -0400)]
; Refer to version 25.2 rather than 26.1.
; I hope the next release from master is indeed 26.1, but for reasons
; that passeth my understanding we currently call it 25.2 everywhere else,
; so we should at least be consistent.
* lisp/emulation/viper-init.el (viper-heading-end):
* lisp/gnus/mm-url.el (mm-url-html-entities):
* lisp/gnus/nnmaildir.el (nnmaildir-article-file-name):
* lisp/mh-e/mh-e.el (mh-invisible-header-fields-compiled):
* lisp/net/tramp-adb.el (tramp-adb-method): Remove leading * from doc.
* lisp/gnus/messcompat.el: Mark (pointless) file not for compilation.
(message-from-style, message-interactive, message-indentation-spaces)
(message-signature, message-signature-file):
Remove settings that match the defaults.
(message-setup-hook, message-mode-hook, message-default-headers)
(message-send-hook, message-send-mail-function):
Just use setq rather than redefining.
Eli Zaretskii [Thu, 9 Jun 2016 13:52:08 +0000 (16:52 +0300)]
Fix copying text properties by 'format'
* src/editfns.c (styled_format): Fix copying text properties from
the format specification to the produced string representation.
(Bug#23730)
(Fformat) Doc fix.
* doc/lispref/strings.texi (Formatting Strings): Document that
text properties from the format specifiers are also copied to the
produced string.
Alan Mackenzie [Thu, 9 Jun 2016 12:24:27 +0000 (12:24 +0000)]
Handle C++ raw strings.
* lisp/progmodes/cc-engine.el (c-raw-string-pos, c-depropertize-raw-string)
(c-depropertize-raw-strings-in-region,
c-before-change-check-raw-strings)
(c-propertize-raw-string-opener, c-after-change-re-mark-raw-strings): New
functions.
* lisp/progmodes/cc-fonts.el (c-basic-matchers-before): Insert a clause for
c-font-lock-raw-strings.
(c-font-lock-raw-strings): New function.
* lisp/progmodes/cc-langs.el (c-get-state-before-change-functions): Insert
c-before-change-check-raw-strings into the C++ value, and c-depropertize-CPP
into the values for C, C++, and Objective C.
(c-before-font-lock-functions): Insert c-after-change-re-mark-raw-strings into
the C++ value.
* lisp/progmodes/cc-mode.el (c-old-BEG, c-old-END): New variables.
(c-depropertize-CPP): New function, extracted from
c-neutralize-syntax-in-and-mark-CPP.
(c-neutralize-syntax-in-and-mark-CPP): Remove the call to
c-clear-char-property-with-value for 'syntax-table value '(1) at the beginning
of the function.
(c-after-change): Set c-old-BEG and c-old-END to the current values of
c-new-BEG and c-new-END.
Martin Rudalics [Thu, 9 Jun 2016 08:12:48 +0000 (10:12 +0200)]
Handle mouse leaving initial window in `mouse-set-region' (Bug#23707)
* lisp/mouse.el (mouse-set-region): If the mouse ends up in
another window or on the menu bar, use `window-point' of
selected window instead of `posn-point' of the event end
(Bug#23707).
Marco Wahl [Wed, 8 Jun 2016 23:03:34 +0000 (16:03 -0700)]
org.el: Fix bindings of < and > for calendar scrolling
[This patch taken from upstream Org repo with 8b63dc9 dated
2014-10-20 (Bug#23725).]
* lisp/org/org.el (org-read-date-minibuffer-local-map):
Switch to the current calendar API for scrolling the calendar.
Paul Eggert [Wed, 8 Jun 2016 17:33:34 +0000 (10:33 -0700)]
Replace IF_LINT by NONVOLATILE and UNINIT
Inspired by a suggestion from RMS in: http://bugs.gnu.org/23640#58
* .dir-locals.el (c-mode): Adjust to macro changes.
* src/conf_post.h (NONVOLATILE, UNINIT): New macros (Bug#23640).
(IF_LINT): Remove. All uses replaced by the new macros.
Michal Nazarewicz [Tue, 7 Jun 2016 20:32:59 +0000 (22:32 +0200)]
Remove ‘ert-with-function-mocked’ macro in favour of ‘cl-letf’ macro
* lisp/emacs-lisp/ert-x.el (ert-with-function-mocked): Remove macro
in favour of ‘cl-letf’ macro which is more generic. All existing
uses are migrated accordingly. The macro has not been included in
an official release yet so it should be fine to delete it.
Glenn Morris [Wed, 8 Jun 2016 03:50:35 +0000 (20:50 -0700)]
Try to avoid hangs and stray procs in network-stream-tests. (Bug#23560)
* test/lisp/net/network-stream-tests.el (connect-to-tls-ipv4-wait)
(connect-to-tls-ipv4-nowait, connect-to-tls-ipv6-nowait):
Ensure gnutls-serv process gets killed.
(echo-server-nowait, connect-to-tls-ipv4-nowait):
Limit the amount of time we might wait.
Glenn Morris [Wed, 8 Jun 2016 01:34:51 +0000 (21:34 -0400)]
Reduce allout.el's pollution of the namespace.
* lisp/allout.el (allout-set-regexp): Rename from set-allout-regexp.
Keep old name as obsolete alias.
(allout-produce-mode-menubar-entries, allout-nullify-prefix-data)
(allout-solicit-char-in-string)
(allout-count-trailing-whitespace-region, allout-regexp-sans-escapes):
Rename to add an "allout-" prefix.
Glenn Morris [Wed, 8 Jun 2016 01:29:30 +0000 (21:29 -0400)]
Misc small webjump updates.
* lisp/net/webjump.el (webjump): Add custom group.
(webjump-sample-sites): Make it a constant.
Remove explicit, old list of GNU ftp mirrors.
(webjump-state-to-postal-alist): Make it a constant.
(webjump-sites): Make it a defcustom.
(webjump-to-iwin): Update for changed remote service.
Glenn Morris [Wed, 8 Jun 2016 01:25:20 +0000 (21:25 -0400)]
Do not hard-code port for package test server. (Bug#23708)
* test/lisp/emacs-lisp/package-resources/package-test-server.py:
Do not hard-code port.
* test/lisp/emacs-lisp/package-tests.el (package-test-update-archives-async):
Update for the above change.
Paul Eggert [Tue, 7 Jun 2016 23:18:02 +0000 (16:18 -0700)]
Port --enable-gcc-warnings to clang 3.7.0
* configure.ac: Add -Wno-tautological-compare to avoid bogus
warnings about 0 <= rlim.rlim_max. Remove flags that no longer
seem to be needed, at least in Fedora 23 x86-64.
Paul Eggert [Tue, 7 Jun 2016 22:41:51 +0000 (15:41 -0700)]
Use __builtin_assume_aligned on untagged Lisp vals
* src/conf_post.h (__has_builtin, __builtin_assume_aligned):
New macros, for compilers not already defining them.
(__has_builtin___builtin_assume_aligned): New macro.
* src/lisp.h (lisp_h_XUNTAG): Use __builtin_assume_aligned.
This shrinks text space by 0.2% on x86-64 with GCC 6.1.
Paul Eggert [Tue, 7 Jun 2016 16:34:27 +0000 (09:34 -0700)]
Merge from origin/emacs-25
6e3adf8 Fix crash in syntax.c after GC 973ce5a Improve squiggly heredoc support in non-SMIE Ruby mode 9d5cceb Fix doc string quoting 0b33a23 Fix mouse dragging of vertical dividers with scroll bars on l... a5d05f4 * etc/PROBLEMS: Mention the link-time problems on FreeBSD 11.
Paul Eggert [Tue, 7 Jun 2016 16:28:49 +0000 (09:28 -0700)]
Fix crash in syntax.c after GC
Problem reported by Vincent Belaïche (Bug#23704).
* src/syntax.c (skip_chars): Recompute pointers into the
buffer after every call to update_syntax_table_forward,
as it can GC.
Michael Albinus [Mon, 6 Jun 2016 16:25:28 +0000 (18:25 +0200)]
Some fixes in filenotify-tests.el for cygwin
* test/lisp/filenotify-tests.el (file-notify--test-timeout):
Reintroduce value for cygwin, it's needed on slow systems.
(file-notify--wait-for-events): Move up definition.
(file-notify--test-no-descriptors): Use `file-notify--wait-for-events'.
(file-notify--test-with-events-check, file-notify--test-with-events)
(file-notify-test08-watched-file-in-watched-dir):
Use :random rather than `random.
(file-notify-test06-many-events): Do not skip for cygwin.
Martin Rudalics [Sun, 5 Jun 2016 09:50:47 +0000 (11:50 +0200)]
Fix mouse dragging of vertical dividers with scroll bars on left (Bug#23690)
* lisp/mouse.el (mouse-drag-line): With scroll bars on the left
adjust trailing edge of window on the left when dragging the
vertical divider of the mode line. (Bug#23690)
Paul Eggert [Thu, 2 Jun 2016 04:00:58 +0000 (21:00 -0700)]
Port angle-bracket TZ settings to MS-Windows
* doc/lispref/os.texi (Time Zone Rules): Document MS-Windows
lack of support for numeric time zone abbreviations.
* src/w32.c (sys_putenv): Convert angle-bracket TZ syntax
to MS-compatible syntax if possible, and to "ZZZ" otherwise.
Problem reported by Kazuhiro Ito (Bug#23600).
Paul Eggert [Wed, 1 Jun 2016 20:25:09 +0000 (13:25 -0700)]
Avoid delving into Git internals for version
* lisp/loadup.el (exec-path): Set it to nil later, so that
emacs-repository-get-version can invoke git commands in the PATH.
* lisp/version.el (emacs-repository--version-git-1): Remove.
(emacs-repository-get-version): Let Git do it rather than
delving into Git internals.
Paul Eggert [Wed, 1 Jun 2016 17:28:43 +0000 (10:28 -0700)]
Fix emacs-repository-get-version with packed .git
* lisp/version.el (emacs-repository-get-version):
Parse .git/packed-refs if it exists.
Problem reported by Martin Rudalics in:
http://lists.gnu.org/archive/html/emacs-devel/2016-05/msg00554.html
K. Handa [Tue, 31 May 2016 22:53:03 +0000 (07:53 +0900)]
Fix incomplete handling of translation table in a coding system.
* coding.c (get_translation): New arg NCHARS. Even if TRANS
is an alist, return a character or a vector of character.
(produce_chars): Adjust for the above change.
(consume_chars): Likewise.
* lisp/net/tramp-sh.el (tramp-do-copy-or-rename-file-out-of-band):
Use local `default-directory'. Simplify command to send.
Don't check return code, this was already done in
`tramp-process-actions'.
(tramp-get-inline-coding): Don't set `default-directory'.
* lisp/net/tramp.el (tramp-action-out-of-band):
Throw `out-of-band-failed'.
(tramp-process-actions): Handle `out-of-band-failed'.
(tramp-call-process, tramp-call-process-region):
Use local `default-directory'.