* lisp/emacs-lisp/ring.el (ring-elements): Don't use the RESULT
argument of `dotimes' when the iteration variable isn't referred by
it.
(ring-member): Don't pass nil as the RESULT argument of `dotimes'
since it's the default.
Having one test for all character classes it is not always trivial to
determine which class is failing. This happens when failure is caused
by ‘(should (equal (point) (point-max)))’ not being met.
With per-character class tests, it is immidiatelly obvious which test
causes issues plus tests for all classes are run even if some of them
fail.
* test/src/regex-tests.el (regex-character-classes): Delete and split
into…
(regex-tests-alnum-character-class, regex-tests-alpha-character-class,
regex-tests-ascii-character-class, regex-tests-blank-character-class,
regex-tests-cntrl-character-class, regex-tests-digit-character-class,
regex-tests-graph-character-class, regex-tests-lower-character-class,
regex-tests-multibyte-character-class,
regex-tests-nonascii-character-class,
regex-tests-print-character-class, regex-tests-punct-character-class,
regex-tests-space-character-class,
regex-tests-unibyte-character-class,
regex-tests-upper-character-class, regex-tests-word-character-class,
regex-tests-xdigit-character-class): …new tests.
* lisp/emacs-lisp/regexp-opt.el (regexp-opt-charset): Do not use
'case-table as charmap char-table’s property. The function has nothing
to do with casing and in addition using 'case-table causes unnecessary
extra slots to be allocated which ‘regexp-opt-charset’ does not use.
RE_CHAR_TO_MULTIBYTE(c) yields c for ASCII characters and a byte8
character for c ≥ 0x80. Furthermore, CHAR_BYTE8_P(c) is true only
for byte8 characters. This means that
c = RE_CHAR_TO_MULTIBYTE (ch);
if (! CHAR_BYTE8_P (c) && re_iswctype (c, cc))
is equivalent to:
c = c;
if (! false && re_iswctype (c, cc))
for 0 ⪬ c < 0x80, and
c = BYTE8_TO_CHAR (c);
if (! true && re_iswctype (c, cc))
for 0x80 ⪬ c < 0x100. In other words, the loop never executes for
c ≥ 0x80 and RE_CHAR_TO_MULTIBYTE call is unnecessary for c < 0x80.
* src/regex.c (regex_compile): Simplyfy a for loop by eliminating
dead iterations and unnecessary macro calls.
decimalnump was used in regex.c only in ISALNUM macro which ored it with
alphabeticp. Because both of those functions require Unicode general
category lookup, this resulted in unnecessary lookups (if alphabeticp
return false decimalp had to perform another lookup). Drop decimalnump
in favour of alphanumericp which combines decimelnump with alphabeticp.
* src/character.c (decimalnump): Remove in favour of…
(alphanumericp): …new function.
* src/regex.c (ISALNUM): Use alphanumericp.
* src/character.h (STRING_CHAR): Update doc.
* src/buffer.h (FETCH_MULTIBYTE_CHAR): Update doc. While at it, change
the function to use BYTE_POS_ADDR instead of open-coding it.
These changes are needed to conform to the C standard's rule for
allocating structs containing flexible array members. C11 says
that malloc (offsetof (struct s, m) + n) does not suffice to
allocate a struct with an n-byte tail; instead, malloc’s arg
should be rounded up to the nearest multiple of alignof (struct s).
Although this is arguably a defect in C11, gcc -O2 + valgrind
sometimes complains when this rule is violated, and when debugging
it’s better to keep valgrind happy.
For details please see the thread containing the message at:
https://gcc.gnu.org/ml/gcc-patches/2016-09/msg00416.html
* lib-src/ebrowse.c, src/alloc.c, src/image.c, src/process.c:
Include flexmember.h.
* lib-src/ebrowse.c (add_sym, add_member, make_namespace)
(register_namespace_alias):
* src/alloc.c (SDATA_SIZE, allocate_string_data):
* src/image.c (xpm_cache_color, imagemagick_create_cache):
* src/process.c (Fmake_network_process):
Use FLEXSIZEOF instead of offsetof and addition.
* src/alloc.c (SDATA_SIZE, vector_alignment):
Use FLEXALIGNOF instead of sizeof (ptrdiff_t).
* src/lisp.h (ALIGNOF_STRUCT_LISP_VECTOR):
Remove, as alloc.c can now calculate this on its own.
This incorporates:
2016-09-07 flexmember: new macro FLEXALIGNOF
2016-09-07 flexmember: port better to GCC + valgrind
2016-08-18 Port modules to use getprogname explicitly
2016-09-02 manywarnings: add -fno-common
* admin/merge-gnulib (GNULIB_TOOL_FLAGS): Don’t avoid flexmember,
since time_rz now uses part of it. Instead, remove m4/flexmember.m4.
* configure.ac (AC_C_FLEXIBLE_ARRAY_MEMBER): Define away,
since Emacs assumes C99 and therefore removes m4/flexmember.m4.
* lib/euidaccess.c, lib/group-member.c, lib/time_rz.c:
* m4/manywarnings.m4: Copy from gnulib.
* lib/flexmember.h: New file, from gnulib.
* lib/gnulib.mk, m4/gnulib-comp.m4: Regenerate.
* lisp/startup.el (command-line-1): Only pass expanded FILENAME argument
of --load when it refers to a normal file, since `load' doesn't handle
directories (Bug #16406).
* lisp/calendar/icalendar.el (icalendar--read-element): Avoid a regex
stack overflow by not using regex to extract values from calendar
events. (Bug#24315)
* lisp/files.el (convert-standard-filename): Doc fix. (Bug#24387)
* etc/NEWS: Suggest a way for mirroring slashes where previously
'convert-standard-filename' was used.
* src/conf_post.h (DEV_TTY): Move from here ...
* src/keyboard.c, src/keyboard.h: ... to here, as it doesn’t need
to be visible everywhere. Make it a constant.
* src/keyboard.c (handle_interrupt, Fset_quit_char):
* src/process.c (create_process): Prefer DEV_TTY to "/dev/tty".
* src/intervals.c (set_point_from_marker): If MARKER comes from
another buffer, recalculate its byte position before using it to
set point.
* src/marker.c (set_marker_internal): If POSITION is a marker from
another buffer, recalculate its byte position before using it.
(Bug#24368)
* lisp/progmodes/cc-engine.el (c-syntactic-re-search-forward): `noerror' can
be given the values `before-literal' and `after-literal', so that when a
search fails, and the `bound' is inside a literal, point is left respectively
before or after that literal.
* src/window.c (window_scroll_pixel_based):
* src/xdisp.c (pos_visible_p): Don't allow simulated redisplay to
start outside the accessible portion of the buffer. This avoids
assertion violations when some Lisp narrows the buffer to less
than the current window, and then attempts to scroll the buffer.
* src/w32proc.c (sys_signal): Don't reject SIGINT, as it is
supported by MS runtime.
* src/term.c (DEV_TTY): Move from here ...
* src/conf_post.h (DEV_TTY): ... to here. Separate definitions
for WINDOWSNT and for the rest.
* src/keyboard.c (handle_interrupt_signal): Use DEV_TTY instead of
a literal "/dev/tty".
* etc/NEWS: Mention the behavior change.
See discussion on:
https://lists.gnu.org/archive/html/emacs-devel/2016-08/msg00552.html
* lisp/image-dired.el (image-dired-cmd-rotate-original-program)
(image-dired-cmd-create-thumbnail-program)
(image-dired-cmd-create-temp-image-program)
(image-dired-cmd-rotate-thumbnail-program)
(image-dired-cmd-write-exif-data-program)
(image-dired-cmd-read-exif-data-program):
Use executable-find to set the defaut value of this option.
(image-dired-cmd-rotate-original-program): Idem.
Search for program 'convert' if 'jpegtran' is not available.
(image-dired-cmd-rotate-original-options):
Set the default value consistent with the executable in
image-dired-cmd-rotate-original-program.
(image-dired-create-thumb, image-dired-display-image)
(image-dired-rotate-thumbnail, image-dired-rotate-original)
(image-dired-set-exif-data, image-dired-get-exif-data):
Throw and error when the executable used in the function is missing.
(image-dired-next-line, image-dired-previous-line):
Use 'forward-line'.
Fix Bug#24317
* lisp/image.el (image-type-from-file-name): Bind case-fold-search
to a non-nil value to force a case insensitive match.
* lisp/image-dired.el (image-dired-rotate-original):
Use image-type (Bug#24317).
(image-dired-get-exif-file-name): Idem.
Set 'no-exif-data-found' and 'data' in same setq call.
Use file-attribute-modification-time.
Passing the prefix argument as the 3rd argument to 'call-interactively'
causes the prefix argument to be interpreted as events, which is not
only wrong, but also causes a type error, as 'current-prefix-arg' can
never be a vector as 'call-interactively' expects. 'call-interactively'
automatically passes its prefix argument to the called function, so just
do that, eliminating faulty behavior.
* lisp/emacs-lisp/checkdoc.el (checkdoc-ispell):
(checkdoc-ispell-current-buffer):
(checkdoc-ispell-interactive):
(checkdoc-ispell-message-text):
(checkdoc-ispell-start):
(checkdoc-ispell-continue):
(checkdoc-ispell-comments):
(checkdoc-ispell-defun):
Do not pass 'current-prefix-arg' to 'call-interactively' as an event
vector; merely allow it to propagate forward to the interactive call.
* lisp/userlock.el (userlock--check-content-unchanged)
(userlock--ask-user-about-supersession-threat): New functions.
* src/filelock.c (lock_file): Use them to avoid spurious prompting.
* doc/lispref/buffers.texi (Modification Time): Update doc of
ask-user-about-supersession-threat.
This avoids a potential crash if too many subprocesses (Bug#24325).
* src/process.c [HAVE_SETRLIMIT]: Include <sys/resource.h>.
(init_process_emacs): If ulimit -n is greater than FD_SETSIZE,
set it to FD_SETSIZE.
This function is the guts of c-indent-defun and c-mark-function.
In particular, when c-defun-tactic is nil, return a correct value rather than
always nil, and when it's 'go-outward, go through an intricate algorithm to
determine the requisite narrowing before the "top-level" defuns go to work.
* lisp/progmodes/cc-cmds.el (c-narrow-to-most-enclosing-decl-block): Enhance
to take additional optional parameter LEVEL, saying how many enclosing levels
of decl-block to narrow to.
(c-declaration-limits): Introduce algorithm to determine narrowing. Use
c-where-wrt-to-brace-block to determine whether to go back to BOD to determine
lower bound.
* lib-src/etags.c (stack_entry): New struct.
(push_node, pop_node, put_entry): New functions.
(free_tree, add_node, invalidate_nodes, put_entries): Re-implement
in a non-recursive way, to avoid stack overflow. (Bug#5847)
* doc/misc/htmlfontify.texi (Interactive):
* lisp/htmlfontify.el (htmlfontify-buffer):
Spell out character names, for clarity. The old doc string
generated the *Help* text ‘^L ([FF]) or ¤ (244)’, where ‘[FF]’
stands for a form feed character; this was confusing.
* lisp/electric.el (electric-quote-mode):
* src/doc.c (syms_of_doc):
* src/editfns.c (Fformat_message):
Remove no-longer-necessary ‘\=’s in doc strings.
This incorporates:
2016-08-30 intprops: tune INT_NEGATE_OVERFLOW for GCC 5 and 6
2016-08-29 xalloc-oversized.h: port change to GCC 6.2.0
2016-08-29 intprops.h: port recent changes to GCC 6.2.0
2016-08-29 intprops.h: use __typeof__ with GCC 7
2016-08-29 intprops.h, xalloc-oversized.h: work with gcc 7
2016-08-24 intprops: fix paren typo on old platforms
2016-08-24 intprops: port to OpenVMS
* lib/intprops.h, lib/xalloc-oversized.h: Copy from gnulib.
* lisp/leim/quail/latin-ltx.el: Special-case phi/varphi because those
are reversed in ucs-names. Also remove FIXME stating sigma/varsigma
were reversed which is not true (anymore?).
* lisp/textmodes/tex-mode.el (tex--prettify-symbols-alist): Fix
prettification of \phi, \varphi, and \varsigma.
Concretely, \phi's symbol was actually varphi, \varphi was missing, and
\varsigma was commented out with a FIXME that it was reversed with
\sigma which it was not.
* lisp/textmodes/tex-mode.el (tex-insert-quote): When in a comment,
default to inserting a normal double quote (") instead of TeX-style
open (``) or close ('') quote marks.
* etc/NEWS (TeX mode): Note the change.
* test/lisp/electric-tests.el (autowrapping-7): Adjust accordingly.
See this thread for discussion:
https://lists.gnu.org/archive/html/emacs-devel/2016-08/msg00611.html
From: Karl Fogel <kfogel {_AT_} red-bean.com>
To: Emacs Development <emacs-devel {_AT_} gnu.org>
Subject: [PATCH] Have LaTeX mode use normal double quotes in comments.
Date: Mon, 29 Aug 2016 14:44:12 -0500
Message-ID: <87twe3icmr.fsf@red-bean.com>
* lisp/progmodes/cc-engine.el (c-back-over-list-of-member-inits): Remove a
superfluous call to c-backward-syntactic-ws.
(c-looking-at-or-maybe-in-bracelist): Change calling convention, so that it
reports encountering the Java "new" keyword. Add, as an alternative, a check
that we can move back over member initializations, and handle this as finding
a brace block.
(c-looking-at-special-brace-list, c-guess-continued-construct): Adapt to the
new calling convention of c-looking-at-or-maybe-in-bracelist.
(c-guess-basic-syntax, CASE 5A.3): Replace lots of inline ad-hoc code with
calls to c-backward-over-enum-header and c-looking-at-or-maybe-in-bracelist,
using the extra info from the value of that function to generate a
topmost-into-cont element where needed (in Java).
* lisp/progmodes/cc-fonts.el (c-font-lock-declarations): Adapt to the new
calling convention of c-looking-at-or-maybe-in-bracelist.
* lisp/progmodes/cc-mode.el (c-fl-decl-start): Adapt to the new calling
convention of c-looking-at-or-maybe-in-bracelist.
