4b9fbdb5a7 ; Update TODO item about ligature support
03d44acfdd * doc/lispref/control.texi (Processing of Errors): Improve...
b48ab743a8 Minor fixups for mutability doc
6ac2326e5b Don’t use “constant” for values you shouldn’t change
* doc/lispref/display.texi (Fontsets): Improve the accuracy of a
cross-reference to "Character Properties".
* doc/emacs/mule.texi (Fontsets, Modifying Fontsets): Improve the
documentation of fontsets and how to modify them.
* lisp/xml.el (xml-escape-string): Search for invalid characters.
(xml-invalid-character): New error symbol.
* test/lisp/xml-tests.el (xml-print-invalid-cdata): New unit test.
* etc/NEWS: Document new behavior.
* lisp/emacs-lisp/comp.el (comp-effective-async-max-jobs): Use
NUMBER_OF_PROCESSORS environment variable if system is Windows NT,
"nproc" if it is in PATH or a default of 1.
* lisp/emacs-lisp/comp.el (comp-run-async-workers): Pass the
compilation commands through a temporary file that is loaded by the
child process. This is also done all other operating systems, even
those that support long command lines. It should not be a problem
since libgccjit uses temporary files too.
* lisp/emacs-lisp/package.el (package-name-column-width)
(package-version-column-width, package-status-column-width)
(package-archive-column-width): New defcustoms.
(package-menu-mode):
Use the values of defcustoms instead of hardcoded
values. (Bug#41086)
* configure.ac: don't add linker flags if compiling on
Windows. Compile dynlib.c if modules or native compilation are
enabled. Always compile comp.c
* lisp/term/w32-win.el: Map 'gccjit to "libgccjit.dll" in
`dynamic-library-alist`.
* src/Makefile.in: Update comments. Update to handle changes in
configure.ac.
* src/comp.c: Add declarations of used libgccjit functions using
DEF_DLL_FN. Add calls to load_gccjit_if_necessary() where necessary.
Add `native-comp-available-p`
* src/comp.h: Remove Fnative_elisp_load. Add syms_of_comp().
* src/emacs.c (main): Always call syms_of_comp()
* src/w32.c (globals_of_w32): Clear Vlibrary_cache when starting
because the libraries loaded when dumping will not be loaded when
starting.
* src/w32fns.c: Add Qgccjit symbol.
This was missed in a previous commit to remove obsolete libraries.
Its deletion was already announced in NEWS.
* lisp/obsolete/levents.el: Delete file. This library has been
obsolete since 23.2.
Practice shows that the vast majority of projects only use one main
root. The users of this API very often make this assumption as well.
The rest of the "roots" should be possible to express through
project-external-roots.
* lisp/progmodes/project.el: Update the commentary.
Only 4 non-obsolete generics now.
(project-root): Replacement for `project-roots'.
All callers updated. Implementations too.
(project-roots): Declare obsolete.
(project-external-roots): Simplify the docstring.
(project-ignores): Update the docstring.
(project-find-regexp): Omit the second arg to project-files.
(project--dir-ignores): Simplify.
(project-compile): Simplify, remove outdated comment.
* lisp/cedet/ede.el: Add a FIXME.
* lisp/net/eww.el (eww): Use function-put in place of put, as
recommended in "(elisp) Symbol Plists".
(eww-follow-link):
* lisp/net/shr.el (shr-browse-url): Rather than call browse-url-mail
directly, call browse-url which respects the user options
browse-url-handlers and browse-url-mailto-function. (Bug#41133)
(shr--current-link-region): Return nil if there is no link at point.
(shr--blink-link): Adapt accordingly.
(shr-fill-line, shr-indent, shr-table-body): Refactor to avoid some
unnecessary allocations.
* etc/NEWS: Announce that eww-follow-link and shr-browse-url support
custom URL handlers.
* etc/NEWS: Announce that json-read-number is now stricter.
* json.el: Bump package version.
(json-encoding-lisp-style-closings, json-pre-element-read-function)
(json-post-element-read-function, json-advance, json-peek)
(json--path): Clarify and improve style of doc strings.
(json-join): Define as an obsolete alias of string-join.
(json-alist-p, json-plist-p): Refactor for speed and declare as
pure, side-effect-free, and error-free.
(json--plist-reverse): Rename function...
(json--plist-nreverse): ...to this, making it destructive for speed.
All callers changed.
(json--plist-to-alist): Remove, replacing single use with map-pairs.
(json--with-indentation): Accept multiple forms as arguments, fix
their indentation, and allow them to be instrumented for debugging.
Add docstring.
(json-pop, json-read-keyword, json-add-to-object)
(json-encode-array): Simplify for speed.
(json-skip-whitespace): Put newline before carriage return for
likely frequency of occurrence, and so that the characters appear in
increasing order.
(json--check-position): Use 1+.
(json-path-to-position): Open code apply-partially.
(json-keywords): Turn into a defconst and mark as obsolete now that
it is no longer used.
(json--post-value, json--number, json--escape): New rx definitions.
(json-encode-keyword): Declare as side-effect-free.
(json-read-number): Reject leading zeros and plus signs, and make
integer part mandatory in accordance with JSON standards and for
consistency with native JSON parsing functions. Eagerly signal
json-number-format when garbage follows a valid number, e.g., when
reading "1.1.1", instead of leaving that up to the caller. Remove
optional internal argument from advertised calling convention now
that the function is no longer recursive.
(json-encode-number): Define as an alias of number-to-string.
(json-special-chars): Turn into a defconst.
(json-read-escaped-char, json-new-object, json-read-file)
(json-pretty-print): Simplify.
(json-read-string): For consistency with other json.el error
reporting, remove check for leading '"', and use the integer value
rather than the printed representation of characters in error data.
At EOB signal json-end-of-file instead of json-string-format.
(json--long-string-threshold, json--string-buffer): New variables.
(json-encode-string): Reimplement in terms of buffer manipulation
for speed (bug#20154).
(json-read-object): Escape ?\} properly.
(json--encode-alist): New function extracted from json-encode-alist.
(json-encode-hash-table, json-encode-alist, json-encode-plist): Use
it to avoid destructively modifying the argument when
json-encoding-object-sort-predicate is non-nil without incurring
unnecessary copying (bug#40693). Encode empty object as "{}" even
when pretty-printing. Simplify for speed.
(json-read-array): Avoid recomputing list length on each iteration
when json-pre-element-read-function is non-nil. Make first element
of json-array-format error data a string for consistency with
json-object-format and to make the displayed error message clearer.
(json-readtable-dispatch): Accept any kind of argument, not just
symbols. Generate the table in a simpler manner so the dispatch
order is clearer. Remove dispatch on ?+ and ?. now that
json-read-number is stricter and for consistency with native JSON
parsing functions. Signal json-end-of-file if argument is nil.
(json-read): Simplify accordingly.
(json-encode): Avoid allocating a list on each invocation.
* lisp/jsonrpc.el (jsonrpc--json-read, jsonrpc--json-encode): Check
whether native JSON functions are fboundp only once, at load time.
* lisp/progmodes/python.el (python--parse-json-array): New function.
(python-shell-prompt-detect): Use it to parse JSON directly as a
list rather than converting from a vector.
* test/lisp/json-tests.el (json-tests--with-temp-buffer): Allow
instrumenting for debugging.
(test-json-join, test-json-plist-to-alist): Remove tests.
(test-json-alist-p, test-json-plist-p, test-json-advance)
(test-json-peek, test-json-pop, test-json-skip-whitespace)
(test-json-read-keyword, test-json-encode-keyword)
(test-json-encode-number, test-json-read-escaped-char)
(test-json-read-string, test-json-encode-string)
(test-json-encode-key, test-json-new-object)
(test-json-encode-hash-table, test-json-encode-plist)
(test-json-encode-list, test-json-read-array)
(test-json-encode-array, test-json-read)
(test-json-read-from-string, test-json-encode): Extend tests.
(test-json-plist-reverse): Rename test...
(test-json-plist-nreverse): ...to this and avoid modifying literal
lists.
(test-json-read-number): Rename test...
(test-json-read-integer): ...to this, focusing on integers.
(test-json-add-to-object): Rename test...
(test-json-add-to-alist): ...to this, focusing on alists.
(json-encode-simple-alist): Rename test...
(test-json-encode-alist): ...to this, extending it.
(test-json-encode-alist-with-sort-predicate): Rename test...
(test-json-encode-alist-sort): ...to this, extending it.
(test-json-encode-plist-with-sort-predicate): Rename test...
(test-json-encode-plist-sort): ...to this, extending it.
(test-json-read-keyword-invalid, test-json-read-fraction)
(test-json-read-exponent, test-json-read-fraction-exponent)
(test-json-read-number-invalid)
(test-json-read-escaped-char-invalid, test-json-add-to-plist)
(test-json-add-to-hash-table, test-json-read-object-empty)
(test-json-read-object-invalid, test-json-read-object-function)
(test-json-encode-hash-table-pretty)
(test-json-encode-hash-table-lisp-style)
(test-json-encode-hash-table-sort, test-json-encode-alist-pretty)
(test-json-encode-alist-lisp-style, test-json-encode-plist-pretty)
(test-json-encode-plist-lisp-style, test-json-read-array-function)
(test-json-encode-array-pretty, test-json-encode-array-lisp-style)
(test-json-read-invalid): New tests.
(test-json-path-to-position-no-match): Use should-not.
(test-json-read-object): Move error check to new test
test-json-read-object-invalid.
(test-json-pretty-print-object): Adapt test now that empty objects
are pretty-printed as "{}".
* lisp/version.el: Don’t put an RCS Id style string into the
executable via purecopy, as this does not work with the pdumper.
* src/emacs.c (RCS_Id): New constant, for 'ident'.
(cherry picked from commit 3d1bcfba5e)
* doc/lispref/control.texi (Processing of Errors): Improve indexing by
adding the word form "handle" in addition to "handling". With thanks
to Eli Zaretskii.
This commit makes ido completion respect the user's setting for
`completion-auto-help' by default. It does this by defining a wrapper
function `ido-completion-auto-help', which calls `ido-completion-help'
only when `completion-auto-help' is non-nil.
* lisp/ido.el (ido-completion-auto-help): New function.
(ido-cannot-complete-command):
Use it as the new default (bug#41340).
* etc/NEWS: Document feature.
* lisp/vc/vc-dir.el (vc-dir-mode): Set local bookmark-make-record-function.
(bookmark-make-record-default, bookmark-prop-get, bookmark-default-handler)
(bookmark-get-bookmark-record): Declarations.
(vc-dir-bookmark-make-record): Make record used to bookmark a `vc-dir' buffer.
(vc-dir-bookmark-jump): Provides bookmark-jump behavior for a `vc-dir' buffer.
* src/comp.c: Introduce the Lisp_X, Lisp_Word, and Lisp_Word_tag
types. These types are used instead of long or long long. Use
emacs_int_type and emacs_uint_types where appropriate.
(emit_coerce): Add special logic that handles the case when
Lisp_Object is a struct. This is necessary for handling the
--enable-check-lisp-object-type configure option.
* src/lisp.h: Since libgccjit does not support opaque unions, change
Lisp_X to be struct. This is done to ensure that the same types are
used in the same binary. It is probably unnecessary since only a
pointer to it is used.
* src/comp.c: Taking the address of an array is the same as casting it
to a pointer. Therefore, the C expression `(EMACS_INT **) &pure` is in
fact adding a layer of indirection that is not necessary. The fix is
to cast the `pure` array to a pointer and store that in a void pointer
that is part of the compiled shared library.
* src/comp.c: Add `define_setjmp_deps()` and `emit_setjmp()` which
abstract over this difference in behavior between operating systems.
WARNING: Not all cases are handled by this patch. The Mingw-64
setjmp.h header deals with many other combinations. I don't think it
is a good idea to replicate the logic of that header inside
emacs. (Maybe a few lines in the configure script could be added to
handle this problem?)
Thanks to Bob Newell for finding this.
* lisp/gnus/gnus-registry.el (gnus-registry-handle-action): If a
message entry ends up with no groups in its 'group key, that means the
entry should be deleted.
, rather than find-file-hook. This keeps which-function-mode active should
the major mode be reinitialized. Also accept a null result from
add-log-current-defun as definitive, should that function have run. This
fixes bug #40714.
* lisp/progmodes/which-func.el (which-func-ff-hook): Put on
after-change-major-mode-hook.
(which-function): Enhance the logic to accept a null result from
add-log-current-defun.
* src/comp.c (comp--compile-ctxt-to-file): Add a preprocessor check to
avoid blocking SIGIO in platforms that don't have it.
Signed-off-by: Andrea Corallo <akrl@sdf.org>
* lisp/net/tramp-archive.el (tramp-archive-file-name-handler):
Increase `max-specpdl-size' temporarily.
* lisp/net/tramp-rclone.el (tramp-rclone-flush-directory-cache):
Fix a problem with older Emacsen.
Check Lisp_Compiled objects better as they’re created,
so that the byte-code interpreter needn’t do the checks
each time it executes them. This improved performance
of ‘make compile-always’ by 1.5% on my platform. Also,
improve the quality of the (still-incomplete) checks, as
this is more practical now that they’re done less often.
* src/alloc.c (make_byte_code): Remove. All uses removed.
(Fmake_byte_code): Put a better (though still incomplete)
check here instead. Simplify by using Fvector instead
of make_uninit_vector followed by memcpy, and by using
XSETPVECTYPE instead of make_byte_code followed by XSETCOMPILED.
* src/bytecode.c (Fbyte_code): Do sanity check and conditional
translation to unibyte here instead of each time the function is
executed.
(exec_byte_code): Omit no-longer-necessary sanity and
unibyte checking. Use SCHARS instead of SBYTES where
either will do, as SCHARS is faster.
* src/eval.c (fetch_and_exec_byte_code): New function.
(funcall_lambda): Use it.
(funcall_lambda, lambda_arity, Ffetch_bytecode):
Omit no-longer-necessary sanity checks.
(Ffetch_bytecode): Add sanity check if actually fetching.
* src/lisp.h (XSETCOMPILED): Remove. All uses removed.
* src/lread.c (read1): Check byte-code objects more thoroughly,
albeit still incompletely, and do translation to unibyte here
instead of each time the function is executed.
(read1): Use XSETPVECYPE instead of make_byte_code.
(read_vector): Omit no-longer-necessary sanity check.
