* lisp/subr.el (event--posn-at-point): Leave POSN alone if it
doesn't have at least 6 members. This follows more faithfully
what 'event-start' and 'event-end' did before they started using
this function, see commit c1cead89f5. Call posn-at-point with
the minibuffer-window when in the minibuffer. (Bug#60252)
* src/fileio.c (write_region) [WINDOWSNT]: Ignore EBADF errors
from fsync -- this means fsync is not supported for this file.
Happens, for example, with the null device. (Bug#59545)
* src/xdisp.c (string_buffer_position): Make sure the TO argument
of 'string_buffer_position_lim' is always inside [BEGV..ZV].
Otherwise 'string_buffer_position_lim' might call
'get-char-property' and friends with invalid position, which will
just signal an error and do nothing useful. (Bug#60144)
* src/treesit.c (init_treesit_functions, ts_tree_cursor_copy)
(ts_tree_cursor_delete): Add boilerplate for using new tree-sitter
functions.
(ts_node_parent): Delete boilerplate of unused function.
This is the last part of the change that fixes bug#60054. The
previous change fixes it for searching functions, this fixes for
treesit-node-parent.
* src/treesit.c (Ftreesit_node_parent): Use the new cursor API.
ts_node_parent has bugs (bug#60054), using cursor API avoids that.
Tree-sitter's author might remove ts_node_parent in the future, so
might as well switch to use cursors now. We are basically
reimplementing some of the logic of ts_node_prev_sibling and
ts_node_parent in the sibling helper and cursor helper functions.
See also https://github.com/tree-sitter/tree-sitter/issues/1992
* src/treesit.c (treesit_traverse_sibling_helper)
(treesit_traverse_child_helper)
(treesit_traverse_match_predicate): Reimplemented to use the cursor API.
(treesit_search_dfs)
(treesit_search_forward): Use the new cursor helper functions.
(Ftreesit_search_subtree)
(Ftreesit_search_forward)
(Ftreesit_induce_sparse_tree): Use cursors.
* test/src/treesit-tests.el (treesit-search-subtree): New test.
(treesit--ert-search-setup): New macro.
(treesit-search-forward)
(treesit-search-forward-named-only)
(treesit-search-backward)
(treesit-search-backward-named-only)
(treesit-cursor-helper-with-missing-node): New tests.
This is part 1 of the change to change node API to cursor API. See
the second part for more detail. (I splitted the change to make the
diff more sane.)
* src/treesit.c (treesit_assume_true)
(treesit_cursor_helper): New functions.
Formerly, the code supported both make-directory and
make-directory-internal handlers. This led to confusion and meant than
in a few cases (nnmaildir, ido) remote directories could not be used in
some cases. Fix this by using only make-directory handlers.
Perhaps there used to be a reason for why there were both
make-directory and make-directory-internal handlers, but whatever that
reason was, it seems to have vanished even before now.
There is no longer any need for make-directory-internal handlers, as
the few remaining callers that use make-directory-internal do so only
when there are no handlers. However, this change keeps the existing
make-directory-internal handlers for now, in case this code is ever
used in older Emacs versions that still call those handlers.
* lisp/gnus/nnmaildir.el (nnmaildir--mkdir):
* lisp/ido.el (ido-file-internal):
* lisp/net/tramp-smb.el (tramp-smb-handle-make-directory):
Use make-directory, not make-directory-internal.
* lisp/net/tramp-smb.el (tramp-smb-handle-make-directory-internal):
Now obsolete.
* src/fileio.c (Fmake_directory_internal): Do not look for or
use a make-directory-internal handler.
* test/lisp/files-tests.el:
(files-tests-file-name-non-special-make-directory-internal):
Remove, as this test incorrectly assumes that make-directory-internal
must support handlers.
Fix SVG scaling with librsvg>2.52 and SVG file with only one known
dimension.
* src/image.c (svg_load_image): Compute a percentage dimension with
the other known dimension.
* lisp/startup.el (initial-scratch-message):
* lisp/simple.el (get-scratch-buffer-create): Revert last changes.
* src/window.c (Fset_window_configuration): Force recalculation of
Vwindow_list after restoring the windows.
* src/buffer.c (other_buffer_safely): Make sure we always return a
valid buffer, even if 'get-scratch-buffer-create' signals an
error.
* src/buffer.c (Foverlay_recenter, Foverlay_lists): Update the doc
strings.
* lisp/subr.el (copy-overlay): Update comment.
* doc/lispref/internals.texi (Buffer Internals): Remove buffer
fields relevant to the old implementation; add the new interval
tree field.
* doc/lispref/display.texi (Overlays, Managing Overlays): Update
text to be consistent with the new implementation of overlays.
(Managing Overlays): Remove documentation of 'overlay-recenter'.
* etc/NEWS: Mention incompatible aspects of overlay
reimplementation.
* src/xfaces.c (syms_of_xfaces)<face-font-lax-matched-attributes>:
Change the default value to t. Update doc string.
(realize_gui_face): When 'Vface_font_lax_matched_attributes' is t,
reset the 3 default font attributes efficiently. Call
'font_maybe_unset_attribute' only if the value is neither nil nor
t. (Bug#59347)
* src/xfaces.c (font_unset_attribute): New function.
(realize_gui_face): Improve commentary and use list instead of
bitmask.
(syms_of_xfaces): Get rid of bitmask. Replace it by a list,
there is no reason any user should have to think about bitmasks
in Emacs lisp.
Between commits bf0d3f76dc (2014) and 6b1ed2f2c9 (2022),
realize_gui_face called font_load_for_lface with an empty or
partly emptied font spec, i.e. it ignored a part of its attrs
argument. The rationale given in bug#17973, which led to
bf0d3f76dc, is not clear. However, 6b1ed2f2c9, which passes
the full font spec to font_load_for_lface and
font_find_for_lface, leads to suboptimal font choices, for
example when the font chosen for the default face has a
weight, slant or width that is not supported by other
available fonts on the system, such as 'medium' or 'heavy'.
If these attributes are not unset here, the call to
font_list_entities in font_find_for_lface arbitrarily limits
the candidate font list to those that are perfect matches for
these attributes, which means that the scoring mechanism is
bypassed. Note that the size attribute in spec is also unset,
in font_find_for_lface.
Also allow unsetting the other attributes, for debugging purposes.
* src/xfaces.c (realize_gui_face): Unset the weight, slant and
width of the font spec. Fixes bug#57555 and bug#59347.
(syms_of_xfaces): New variable
'realize-gui-face-ignored-spec-attributes'.
Fix the problem described in bug#59693.
* src/treesit.c (treesit_record_change): Always use the base buffer.
(Ftreesit_parser_create): Always use the base buffer. Also change the
for loop into FOR_EACH_TAIL (stylistic change).
(Ftreesit_parser_list): Always use the base buffer.
* doc/lispref/parsing.texi (Using Parser): Update manual.
* test/src/treesit-tests.el (treesit-indirect-buffer): New test.
This reverts commit c774e83e36. It
causes crashes on GNU/Linux systems using libgmp, as there a shared
library initializer is used to set the malloc functions on startup, so
they are not dumped.
* src/treesit.c (treesit_check_position): Extract out new function.
(Ftreesit_node_first_child_for_pos)
(Ftreesit_node_descendant_for_range): Replace code with the new
function.
(Ftreesit_query_capture): Add missing check for node and parser. Add
check for range for BEG and END. Move treesit_initialize to the
beginning of the function.
* test/src/treesit-tests.el (treesit-node-api)
(treesit-query-api): Add tests for out-of-range error.
* src/lread.c (read_bool_vector, skip_lazy_string):
Replace `|` with `||` to explicitly introduce sequence points since
the variables, `length` and `nskip`, are mutated more than once.
The `|` was just a weak attempt at micro-optimisation in any case;
sorry about that.
Before this change Ftreesit_query_capture doesn't convert character
position to byte position for BEG and END parameters. I observed
fontification issue in css files but couldn't figure out why, now I
know :-)
I decide to keep treesit--font-lock-query-expand-range, since it might
provide a escape hatch for problems we discover in the future, and it
should be very cheap so no downside of keeping it.
* lisp/textmodes/css-mode.el (css-ts-mode): Stop setting
treesit--font-lock-query-expand-range.
* lisp/treesit.el (treesit--font-lock-query-expand-range): Update
docstring.
* src/treesit.c (Ftreesit_query_capture): Convert BEG and END to byte
position. Also added parentheses wround "beg_byte - visible_beg" in
the call to ts_query_cursor_set_byte_range (i.e., style change).
* nsterm.m ([EmacsView resetCursorRects:]): Be defensive when
accessing FRAME_OUTPUT_DATA. [resetCursorRects:] can be called
from the event loop after the frame is deleted. When this
happens, emacsframe is NULL. This means there is an underlying
leak of the EmacsView object! (Bug#59794)
Do not merge to master.
Copyright-paperwork-exempt: yes
* .dir-locals.el (c-mode): Add more noise macro names.
* src/frame.c (delete_frame): Do not delete terminal if force is
Qnoelisp.
* src/xterm.c (x_wm_supports): Fix coding style.
(x_delete_display): Delete the supported atoms list.
(x_delete_terminal): Delete all terminals. Free the scratch
cursor GC.
* src/treesit.c (treesit_sync_visible_region): Set nee_reparse flag to
true if buffer range changes. Add some assertion.
* src/treesit.c (treesit_ensure_parsed): Move
treesit_sync_visible_region in front of the check for need_reparse.
* src/xdisp.c (get_locked_narrowing_begv)
(get_locked_narrowing_zv): Safer handling of negative values.
(handle_fontified_prop): Do not use locked narrowing if the region
size is <= 0.
* src/keyboard.c (safe_run_hooks_maybe_narrowed): Do not use
locked narrowing if the region size is <= 0.
