* lisp/simple.el (delete-trailing-whitespace): Set newline’s character
syntax to non-whitespace so that ‘\s-’ regular expression does not match
it.
This simplifies the loop slightly since a simple ‘\s-+$’ can be used and
as a consequence ‘line-beginning-position’ function does not need to be
called any longer.
Furthermore, when newline has whitespace syntax, ‘\s-$’ regular
expression ends up matching empty lins since ‘\s-’ matches newline
characetr of proceeding line. This leads to needless loop iterations.
Since previous change to ‘delete-trailing-whitespace’ already introduced
‘with-syntax-table’, take advantage of it and also overwrite newline’s
character syntax.
* lisp/simple.el (delete-trailing-whitespace): Treat form fead as
a non-whitespace character (regradless of whether it’s character syntax
is whitespace) and delete any whitespace following it instead of leaving
lines with form feeds completely unchanged. I.e. a line like "\f " will
now became "\f".
* configure.ac [USE_X_TOOLKIT]: Define X_TOOLKIT_EDITRES if
_XEditResCheckMessages is declared in X11/Xmu/Editres.h and may be
linked with -lXmu. This should work with any non-ancient Xmu library.
* xfns.c (toplevel): Remove old cruft.
(x_window) [USE_X_TOOLKIT]: Use X_TOOLKIT_EDITRES.
* xterm.c (toplevel): Remove old cruft.
(handle_one_xevent): Use X_TOOLKIT_EDITRES.
* xterm.h (toplevel): Include X11/Xmu/Editres.h if X_TOOLKIT_EDITRES.
Problem reported by Christer Ekholm (Bug#23864).
* src/process.c (wait_reading_process_output): Further fix for
typo introduced in 2015-07-06T02:19:13Z!eggert@cs.ucla.edu when
wait == INFINITY and got_output_end_time is invalid. See:
http://bugs.gnu.org/23864#20
* lisp/progmodes/cc-fonts.el (c-font-lock-declarations): Remove four arms of
the "main" cond form in the function, which have been redundant since the
extraction of c-font-lock-cut-off-declarators from the function on
2016-06-15/16.
* lisp/progmodes/cc-fonts.el (c-font-lock-cut-off-declarators)
(c-font-lock-enclosing-decls)
* lisp/progmodes/cc-mode.el (c-fl-decl-start): Replace invocations of
c-beginning-of-decl-1 with less accurate invocations of
c-syntactic-skip-backwards to speed up fontification.
* lisp/ibuffer.el (ibuffer-unmark-all-marks): New command (Bug#23680).
(ibuffer-mouse-popup-menu): Use it.
(ibuffer-mode): Update mode doc.
(ibuffer-mode-map): Bind 'ibuffer-unmark-all-marks' to 'U'.
Rebind 'ibuffer-do-replace-regexp' to 'r'.
; * etc/NEWS: Add entry for this change.
Problem reported by Constantin Kulikov (Bug#23684).
* src/process.c (wait_reading_process_output):
Fix typo introduced in 2015-07-06T02:19:13Z!eggert@cs.ucla.edu
when wait == INFINITY and got_output_end_time is invalid.
In this case the code should break, not continue.
* lisp/ibuf-ext.el (ibuffer-mark-by-content-regexp): New command.
(ibuffer-never-search-content-name): New option.
(ibuffer-never-search-content-mode): Idem.
(ibuffer-mark-by-content-regexp): Use them (Bug#23734).
* lisp/ibuffer.el (ibuffer-mode-map): Bind new command to '% c' and '% g'.
(ibuffer-mode): Update mode documentation.
; * etc/NEWS: Add NEWS entry for these changes.
* src/term.c (append_glyph, append_composite_glyph)
(append_glyphless_glyph): Set the avoid_cursor_p and multibyte_p
members of the produced 'struct glyph'. Fixes cursor positioning
on the first character after a stretch glyph produced from
line-prefix, wrap-prefix, etc. on TTY frames.
* src/indent.c (Fvertical_motion): Zero the current_y coordinate
whenever we zero the vpos vertical position of the iterator.
* src/xdisp.c (move_it_in_display_line_to): Handle line-prefix and
wrap-prefix regardless of whether the current Y coordinate is
inside the window dimensions. (Bug#23879)
* src/dispextern.h (FACE_FROM_ID_OR_NULL): Renamed from
FACE_OPT_FROM_ID; all callers changed.
* src/xdisp.c (extend_face_to_end_of_line): Call FACE_FROM_ID, not
FACE_FROM_ID_OR_NULL, as the resulting face is immediately
dereferenced.
(fill_gstring_glyph_string): Call FACE_FROM_ID, not
FACE_FROM_ID_OR_NULL, as the resulting face will be dereferenced
when the glyph string is drawn.
(BUILD_COMPOSITE_GLYPH_STRING): Call FACE_FROM_ID, not
FACE_FROM_ID_OR_NULL, as the resulting face will be dereferenced
in fill_composite_glyph_string.
(calc_line_height_property): Call FACE_FROM_ID_OR_NULL rather that
FACE_FROM_ID, since the function and its caller can cope with that
situation. Conflate 3 tests of missing face or font into just
one.
* src/xfaces.c (Fx_list_fonts, Fface_font, lookup_face): Call
FACE_FROM_ID_OR_NULL rather that FACE_FROM_ID, since these
functions can cope with that situation.
(lookup_derived_face): Don't call FACE_FROM_ID if the result will
not be used.
* src/w32console.c (w32_face_attributes): Remove redundant 'eassert'.
This applies to applying or removing syntax-table text properties in raw
strings which affect the stringiness of a piece of text. This fixes the
bug
reported in
http://lists.gnu.org/archive/html/emacs-devel/2016-06/msg00695.html.
* lisp/progmodes/cc-engine.el (c-truncate-semi-nonlit-pos-cache): new
defsubst.
(c-invalidate-state-cache-1): Call new function in place of inline
manipulation.
(c-depropertize-raw-string, c-propertize-raw-string-opener): truncate the
semi-nonlit cache when pertinent syntax-table text properties get applied to
the text.
Also amend some pertinent documentation. This fixes bug #23818.
* lisp/progmodes/cc-engine.el (c-beginning-of-decl-1): Also check for a
virtual semicolon at a place where we check for other types of statement ends.
* lisp/progmodes/cc-vars.el (c-macro-nacmes-with-semicolon): Remove from the
doc string the bit saying that the variable is a prototype and liable to
change.
* doc/misc/cc-mode.texi (Macros with ;): Enhance, stating that configuring
macros with semicolon can prevent C-M-a missing the beginning of defun.
* lisp/dired.el (dired-re-inode-size): Update 'dired-re-inode-size'
to match when Dired displays the allocated file size column
in human readable units (Bug#22255).
src/nsfns.m: Add colour settings functions to ns_frame_park_handlers.
src/nsterm.m (ns_draw_window_divider): ns_focus has to go before the
attempt to set the colour.
src/nsterm.m (ns_draw_vertical_window_border): This had the same bug as
above, although I didn't see any errors.
* src/window.c (displayed_window_lines): Fix an off-by-one error
when the bottom of the last display line is exactly at window's
last pixel. Remove kludgey fix for TTY frames that is no longer
needed. (Bug#15760)
(Fmove_to_window_line): Doc fix.
* doc/lispref/positions.texi (Screen Lines): Clarify and make more
accurate the documentation of 'move-to-window-line'.
Also amend the code so that low-level searches to the end of literals are done
only when these positions get used.
* lisp/progmodes/cc-engine.el (c-crosses-statement-barrier-p): Use the new
c-literal-start instead of c-literal-limit.
(c-state-semi-nonlit-pos-cache): Change the structure of this cache, such that
it stores details of the literal at a point, rather than merely points outside
of literals.
(c-state-semi-pp-to-literal, c-state-full-pp-to-literal)
(c-cache-to-parse-ps-state, c-parse-ps-state-to-cache, c-ps-state-cache-pos)
(c-parse-ps-state-below, c-literal-start): New functions.
(c-state-semi-safe-place): Removed.
(c-in-literal): Use c-state-semi-pp-to-literal, so as not to scan to its end.
(c-literal-limits, c-determine-limit-get-base): consequential amendments.
(c-find-decl-spots, c-before-change-check-<>-operators, c-raw-string-pos)
(c-guess-basic-syntax (CASE 2)): Avoid needless scans to end of literals.
* lisp/progmodes/cc-fonts.el (c-font-lock-doc-comments): Avoid needless scans
to end of literals.
* lisp/progmodes/cc-mode.el (c-fl-decl-start): Avoid needless scans to end of
literals.
* lisp/progmodes/cc-cmds.el (c-beginning-of-defun, c-end-of-defun)
(c-defun-name, c-declaration-limits): Avoid needless scans to end of literals.
* lisp/progmodes/ruby-mode.el (ruby-toggle-string-quotes):
Handle quoting of the first character in the string.
* test/lisp/progmodes/ruby-mode-tests.el
(ruby-toggle-string-quotes-quotes-correctly): Update.
* lisp/progmodes/ruby-mode.el (ruby-toggle-string-quotes):
Change logic to quote based on the current quote of the string.
* test/lisp/progmodes/ruby-mode-tests.el
(ruby-toggle-string-quotes-quotes-correctly): Add test.
Copyright-paperwork-exempt: yes
Problem reported by Juliusz Chroboczek (Bug#17976)
and by Artur Malabarba (Bug#23620).
Patch from a suggestion by Andreas Schwab in:
http://bugs.gnu.org/17976#39
This patch is for non-MS-Windows platforms.
I don't know the situation on MS-Windows.
* src/process.c (connecting_status):
New function, for (connect . ADDRINFOS).
(connect_network_socket, check_for_dns, wait_for_socket_fds)
(wait_while_connecting, wait_reading_process_output, status_notify):
Use it.
(decode_status, Fmake_network_process):
Support (connect . ADDRINFOS) status.
(connect_network_socket) [!WINDOWSNT]:
If the connection failed and there are other addresses to try, do not
signal an error; instead, loop around to try the next address.
(wait_reading_process_output): Advance to the next address
if there are multiple addresses and the first remaining address
failed.
* src/process.h (struct Lisp_Process.status): Adjust comment
to describe (connect . ADDRINFOS).
* lisp/term.el (ansi-term): Remove unnecessary setting of
'bidi-paragraph-direction'. The underlying problem is solved in
bidi.c by changing the regexps that define the paragraph beginning
and end. (Bug#20611)
