* src/nsterm.m: remove NSWindowTabbingModeDisallowed to respect
system-wide preferences
* etc/NEWS: add mention of native tab support in macOS and where
to specify system-wide setting
See discussion starting at
https://lists.gnu.org/archive/html/emacs-devel/2021-06/msg00034.html
for the details.
* src/keyboard.c (read_decoded_event_from_main_queue): For
raw-text encoed input, if Meta bit is encoded, apply the Meta
modifier to single-byte characters that have the 0x80 bit set.
For input encoded otherwise, if the Meta bit is encoded, remove
the 0x80 bit after decoding the characters.
(tty_read_avail_input): Reset the 0x80 bit only if Meta bit is not
encoded.
(Fset_input_meta_mode): Support 'encoded' as the value of META.
(Fset_input_mode): Document 'encoded' for META.
(Fcurrent_input_mode): Support and document 'encoded' as the value
of META.
(syms_of_keyboard): DEFSYM 'encoded'.
* src/composite.c (find_automatic_composition): Accept one
additional argument BACKLIM; don't look back in buffer or string
farther than that. Add an assertion for BACKLIM.
(composition_adjust_point, Ffind_composition_internal): Callers
adjusted.
* src/composite.h (find_automatic_composition): Adjust prototype.
* src/character.c (lisp_string_width): Call
'find_automatic_composition' with the value of BACKLIM equal to POS,
to avoid costly and unnecessary search back in the string, since
those previous characters were already checked for automatic
compositions. (Bug#48734) (Bug#48839)
* src/xdisp.c (extend_face_to_end_of_line): Fix calculation of
fill-column-indicator on TTY frames. Suggested by Jimmy Aguilar
Mena <spacibba@aol.com>.
* src/nsterm.h: Update EmacsSurface definition.
* src/nsterm.m ([EmacsView focusOnDrawingBuffer]): Don't change the
CGContext's settings directly.
([EmacsView unfocusDrawingBuffer]): Don't release the context here.
(CACHE_MAX_SIZE): Add maximum cache size.
([EmacsView updateLayer]): Send a request for getContext, which will
copy the buffer and create the context if it doesn't already exist, to
the NS run loop.
([EmacsSurface initWithSize:ColorSpace:Scale:]): Add the scale factor
and if there's already a CGContext available, reuse it.
([EmacsSurface dealloc]): No longer need to release lastSurface
separately.
([EmacsSurface getContext]): Don't create more surfaces than we have
spaces for in the cache.
([EmacsSurface releaseContext]): If there's no context don't try to
release it and put currentSurface back on the cache instead of
lastSurface.
([EmacsSurface copyContentsTo:]): Don't try to copy if the source and
destination are actually the same surface.
When a frame in this state is selected again by Fselect_frame (but not by
Fselect_window), the mini-window rather than the frame's currently selected
window, is chosen for selection, should there still be an active minibuffer in
it.
This fixes bug #48674.
* src/frame.h (struct frame): Add new boolean field select_mini_window_flag.
* src/frame.c (make_frame): Initialize select_mini_window_flag to false.
(do_switch_frame): Set the new flag appropriately for the old frame, and
process the new frame's setting of this flag, before setting it to false.
* src/window.c (select_window): Set f->select_mini_window_flag to false.
(Fset_frame_selected_window, Fdelete_other_windows_internal)
(Fdelete_window_internal): Add comments clarifying that there is no clearing
of f->select_mini_window_flag in these functions.
Do this by using frame relative positions rather than window relative ones,
which gave rise to spurious drag events when the origin of the window changed
between the mouse down and up events. This fixes bug #48409.
* keyboard.c (frame_relative_event_pos): New static variable.
(make_lispy_event): Record frame relative position of down event. When the up
event is in the same position, "move" this event into the window of the down
event when this window (typically a mini-window) is no longer under the mouse.
Call make_lispy_position a second time to generate this changed event.
(syms_of_keyboard): Declare Qwindow_edges. static_pro
frame_relative_event_pos.
'lisp_string_width' is called from 'format' and 'format-message',
which can be called both very early into Emacs initialization and in
other contexts where using the font backend is impossible or
undesirable. So this commit changes 'lisp_string_width' to try
accounting for automatic compositions only when explicitly requested,
and only 'string-width' does that; 'format' and 'format-message'
don't.
* src/character.c (lisp_string_width): Accept an additional
argument AUTO_COMP; attempt accounting for auto-compositions only
if that argument is non-zero. (Bug#48732)
* src/editfns.c (styled_format):
* src/character.c (Fstring_width): Callers of 'lisp_string_width'
adjusted.
* lisp/international/quail.el (quail-add-unread-command-events):
New function.
(quail-start-translation, quail-start-conversion)
(quail-update-translation, quail-next-translation)
(quail-prev-translation, quail-next-translation-block)
(quail-prev-translation-block, quail-minibuffer-message): Use
'quail-add-unread-command-events' (and partly revert commit
03e3440dbb). (Bug#48042)
* lisp/subr.el (inhibit--record-char): Now obsolete.
* lisp/term/xterm.el (xterm--init): New function, with most of the
code of former 'terminal-init-xterm'.
(terminal-init-xterm): Clear the lossage after terminal
initialization (see Bug#44908).
(xterm--read-event-for-query): Do not use 'inhibit--record-char'
anymore (revert commit 3e6525d69f).
* src/keyboard.c (syms_of_keyboard): Remove 'inhibit--record-char'
(partly revert 03e3440dbb).
(record_char, syms_of_keyboard_for_pdumper): Do not use
'inhibit_record_char anymore'.
* src/character.c (lisp_string_width): Compute C pointer to data
of STRING immediately before using it, since STRING could be
relocated by GC triggered by processing compositions. (Bug#48711)
* src/comp.c (Fcomp_el_to_eln_rel_filename): Don't use
'file-truename', as it is only available once files.el is loaded,
which doesn't work during dumping, until loadup loads files.el.
Instead, use 'realpath'. (Bug#48578)
* src/w32.c (realpath): New function.
* src/w32.h (realpath): Add prototype.
* nt/mingw-cfg.site (ac_cv_func_realpath)
(gl_cv_func_realpath_works): Define to "yes", as this function is
now implemented in w32.c.
Numbers with a trailing dot and an exponent were incorrectly read as
integers (with the exponent ignored) instead of the floats they should
be. For example, 1.e6 was read as the integer 1, not 1000000.0 as
every sane person would agree was meant. (Bug#48678)
Numbers with a trailing dot but no exponent are still read as
integers.
* src/lread.c (string_to_number): Fix float lexing.
* test/src/lread-tests.el (lread-float): Add test.
* doc/lispref/numbers.texi (Float Basics): Clarify syntax.
* src/character.c (lisp_string_width): Compute the width when
automatic compositions can happen more accurately, by using the
pixel widths of the grapheme clusters, divided by the default
face's font width. Disregard the current state of
'auto-composition-mode', for consistency with 'current-column' .
* src/composite.c (find_automatic_composition): Now extern.
(char_composable_p): Don't assume 'unicode-category-table' is
always available.
* src/composite.h (find_automatic_composition): Add prototype.
* src/character.c (lisp_string_width): Support automatic
compositions; call 'find_automatic_composition' when
'auto-composition-mode' is ON.
* src/sysdep.c (system_process_attributes): Fix misprint in 'tty' attr
- should be 'ttname' instead. Change 'utime', 'stime', 'time',
'majflt' attrs to obtain them from proc_pid_rusage, as sysctl call
used before doesn't give correct values; remove 'minflt' because it's
not available. Obtain 'vsize' / 'rss' / 'thcount' from proc_pidinfo.
Use sysctl with KERN_PROCARGS2 to obtain args: value contains both
argc and argv, so argv can be reliably cut out.
* src/process.c (wait_reading_process_output): When looping through
fds, continue from where we left off.
(syms_of_process): Vprocess_prioritize_lower_fds: New variable
(bug#48118).
* src/fns.c (Fline_number_at_pos): Pass character position to
args_out_of_range. Suggested by Andreas Schwab
<schwab@linux-m68k.org>. Call args_out_of_range_3 to show both
ends of the accessible portion.
* src/gmalloc.c (hybrid_free_1): New function, with the body of
the previous 'hybrid_free'.
(hybrid_free): Call 'hybrid_free_1' while preserving the value of
'errno'. Suggested by Paul Eggert <eggert@cs.ucla.edu>.
* src/conf_post.h [HYBRID_MALLOC || DARWIN_OS && HAVE_UNEXEC]:
Include <stdlib.h> here, before redirecting 'malloc' and friends
to their hybrid_* and unexec_* equivalents. #undef malloc and
friends before redefining. Provide prototypes for the
replacements. Suggested by Paul Eggert <eggert@cs.ucla.edu>.
* src/gmalloc.c [HYBRID_MALLOC]: Remove declarations of 'malloc'
and friends, as they are now redundant: we include <stdlib.h> in
conf_post.h before redefining 'malloc' etc., and that provides
prototypes from system headers.
* configure.ac (HYBRID_MALLOC): Remove kludge to avoid replacement
of 'free' by Gnulib. (Bug#36649)
* src/search.c (Freplace_match): Don't test for change in search_regs
start and end, this is unreliable if change hooks modify text earlier
in the buffer (bug#35264).
* src/emacs.c (main): Call 'set_initial_minibuffer_mode' before
entering recursive-exit.
* src/minibuf.c (init_minibuf_once_for_pdumper): Don't call
'set_minibuffer_mode' here...
(set_initial_minibuffer_mode): ... set it in this new function.
(Bug#48446)
* src/lisp.h: Add prototype for 'set_initial_minibuffer_mode'.
* src/minibuf.c (read_minibuf): Call get_minibuffer before incrementing
minibuf_level, in case a hook function calls Factive_minibuffer_window.
(init_minibuf_once_for_pdumper): Create *Minibuf-0* here (moved from
init_minibuf_once), and set its mode, so that clicking in the mini-window
immediately after start up works (thanks, Eli Z.).
Also fix some unsafe coding.
* lisp/window.el (push-window-buffer-onto-prev): New function, extracted from
(record-window-buffer): Refactor by extracting the above, and removing the now
redundant parameter DO-MINIBUF.
* src/minibuf.c (zip_minibuffer_stacks, read_minibuf): Replace calls to
get_minibuffer (0) by nth_minibuffer (0). Replace calls to
Qrecord_window_buffer by calls to Qpush_window_buffer_onto_prev.
(Factive_minibuffer_window, read_minibuf_unwind): Call abort_emacs should an
"impossible" null value be returned by nth_minibuffer.
(read_minibuf): Move the get_minibuffer_call to just after the incrementation
of minibuf_level as a precaution against a missing buffer in
Vminibuffer_list.
(nth_minibuffer): Replace XCAR by Fcar, to allow (car nil) to work.
(init_minibuf_once): Create the inactive buffer *Minibuf-0*.
(syms_of_minibuf): New DEFSYM, Qpush_window_buffer_onto_prev.
* src/window.c (restore_window_configuration): Replace some XCARs and XCDRs
by Fcar_safe and Fcdr_safe.
* src/process.c (add_non_keyboard_read_fd): Make this a public function.
(add_process_read_fd): Fold old, static add_non_keyboard_read_fd guts
into here.
* src/xsmfns.c (ice_conn_watch_CB): Call add_non_keyboard_read_fd
(bug#43834).
