Instead of referencing obarray directly, that function has to consider
a collection of completions which includes the shorthand versions of
some of the symbols. That collection changes from buffer to buffer,
depending on the choice of elisp-shorthands.
To make this process efficient, and avoid needless recalculation of
the above collection, a new obarray-specific cache was invented. The
Elisp variable obarray-cache is immediately nullified if something
touches the obarray.
* lisp/progmodes/elisp-mode.el : New helper.
(elisp-completion-at-point): Use new helpers.
(elisp--completion-local-symbols)
(elisp--fboundp-considering-shorthands)
(elisp--bboundp-considering-shorthands): New helpers
* src/lread.c (intern_driver): Nullify Qobarray_cache.
(syms_of_lread): Add Qobarray_cache.
* test/lisp/progmodes/elisp-mode-tests.el
(elisp-shorthand-completion-at-point): New test.
* test/lisp/progmodes/elisp-resources/simple-shorthand-test.el
(f-test-complete-me): New fixture.
This simplification in requirements makes for more complex C code but
that code is much less wasteful in Lisp strings than the previous
implementation.
* src/lread.c (read1): Rework.
(Fintern): Rework.
(Fintern_soft): Rework.
(Funintern): Rework.
(oblookup_considering_shorthand): Rewrite.
* test/lisp/progmodes/elisp-mode-tests.el (elisp-shorthand-read-buffer)
(elisp-shorthand-read-from-string): Use new format of
elisp-shorthands.
* test/lisp/progmodes/elisp-resources/simple-shorthand-test.el (f-test)
(f-test2, f-test3): Use new form of elisp-shorthands.
It passes the tests designed for the previous Elisp implementation.
Likely, this isn't the final form of the implementation. For one, the
reader is much slower and allocates a Lisp string for every atom read,
regardless if its already interned or not. This has the potential to
be catastrophic in terms of GC.
Also rename the main variable to elisp-shorthands, from the
repetitive shorthand-shorthands.
For some reason, I had to put 'hack-elisp-shorthands' and
'load-with-shorthands-and-code-conversion', the new source-file
loading functions, in lisp/international/mule.el.
Otherwise, lisp/loadup.el wouldn't see them, for some reason that I
didn't investigate. This should probably be fixed.
* lisp/shorthand.el: Remove.
* test/lisp/shorthand-tests.el: Remove.
* src/lread.c:
(read1, Fintern, Fintern_soft, Funintern): Use
oblookup_considering_shorthand.
(oblookup_considering_shorthand): New helper.
(syms_of_lread): Declare elisp-shorthands.
* lisp/progmodes/elisp-mode.el (elisp-shorthands):
Put a safe-local-variable spec.
* test/lisp/progmodes/elisp-mode-tests.el (elisp-shorthand-read-buffer)
(elisp-shorthand-read-from-string)
(elisp-shorthand-byte-compile-a-file)
(elisp-shorthand-load-a-file): New tests.
* test/lisp/progmodes/elisp-resources/simple-shorthand-test.el: New file
* lisp/loadup.el (load-source-file-function): Set to
load-with-shorthands-and-code-conversion.
* lisp/international/mule.el (hack-elisp-shorthands): Move here.
(load-with-shorthands-and-code-conversion): And here.
* lisp/international/characters.el (update-glyphless-char-display):
(glyphless-char-display-control): Add control knob for U+FE00 through
U+FE0F, defaulting to 'thin-space'.
* doc/lispref/display.texi (Glyphless Chars): Document it.
* lisp/obsolete/erc-compat.el (erc-decode-coding-string)
(erc-encode-coding-string, erc-set-write-file-functions)
(erc-emacs-build-time, erc-replace-match-subexpression-in-string)
(erc-member-if, erc-delete-if, erc-remove-if-not, erc-subseq):
Explicitly declare obsolete.
(erc-define-minor-mode): Make into obsolete function alias for
'define-minor-mode'.
(erc-user-emacs-directory): Make into obsolete variable alias for
'user-emacs-directory'.
* lisp/whitespace.el (whitespace-empty-at-bob-regexp):
* lisp/vc/vc-dir.el (vc-dir-child-files-and-states):
* lisp/vc/vc-annotate.el
(vc-annotate-show-changeset-diff-revision-at-line):
* lisp/vc/ediff.el (ediff-merge-with-ancestor-command):
* lisp/gnus/nnvirtual.el (nnvirtual-update-xref-header)
(nnvirtual-map-article):
* lisp/emulation/viper-init.el (viper-ESC-keyseq-timeout):
* lisp/emulation/viper-cmd.el
(viper-set-parsing-style-toggling-macro):
* lisp/cedet/data-debug.el (data-debug-insert-hash-table-button):
"The first line of a doc string must be a complete sentence."
Also, fix some awkward wording in doc strings while at that.
* lisp/emacs-lisp/ert.el (ert--explain-string-equal): Add fast-path to
avoid doing extra work.
Problem reported by Mattias Engdegård <mattiase@acm.org>.
* doc/lispref/searching.texi (Rx Constructs, Rx Functions):
Add clarifications and improve naming of arguments. Add examples
illustrating the differences between `rx` and `rx-to-string`.
* lisp/emacs-lisp/bytecomp.el
(byte-compile-docstring-length-warn): Warn about overly long
docstring in lambda. (Bug#44858)
(byte-compile--wide-docstring-p): Improve comment.
* test/lisp/emacs-lisp/bytecomp-tests.el
("warn-wide-docstring-defun.el"): Update to test for the above new
warning.
* lisp/custom.el (defcustom): Avoid warning about long docstring
lines. This was caused by the value of the defcustom being treated as
docstring due to it being wrapped in a lambda.
* lisp/emacs-lisp/bytecomp.el (byte-compile--wide-docstring-p):
Ignore more function argument lists.
* test/lisp/emacs-lisp/bytecomp-tests.el
(bytecomp-tests-byte-compile--wide-docstring-p): New test.
* lisp/gnus/mm-view.el (mm-text-html-renderer-alist): Add a new
form for links.
(mm-links-remove-leading-blank): Make obsolete.
(mm-inline-wash-with-file):
(mm-inline-render-with-file): Make obsolete -- they were awkwardly
defined and only used with links.
(mm-inline-render-with-links): New function.
This reverts commit fe5b20410f.
This change was not correct; these are Lisp symbols that should be in
lower-case. Problem reported by Mattias Engdegård <mattiase@acm.org>.
This should really be taken care of by a syntax normalisation step in
the frontend, but there is no such step for non-lexbind code yet.
* lisp/emacs-lisp/byte-opt.el (byte-optimize-letX): Tolerate bindingsa
without initialising expressions.
* test/lisp/emacs-lisp/bytecomp-tests.el (bytecomp-tests--test-cases):
Add test cases.
Previously, variables bound outside `while` loops were not substituted
inside even in the absense of mutation. Add the necessary mutation
checking inside loops to allow propagation of values and aliased
variables.
* lisp/emacs-lisp/byte-opt.el
(byte-optimize--inhibit-outside-loop-constprop): New variable.
(byte-optimize-form-code-walker): First traverse each loop without
substitution to discover mutation, then without restrictions.
* test/lisp/emacs-lisp/bytecomp-tests.el (bytecomp-test-loop): New.
(bytecomp-tests--test-cases): Add test cases.
* lisp/emacs-lisp/ert.el (ert-deftest):
Evaluate the body of `ert-deftest` with the `lexical-binding` value of
the source file (or more precisely the value in force when the
definition is evaluated), which is what everyone expected, instead of
always using dynamic binding which is what they got until now.
* test/lisp/emacs-lisp/ert-tests.el
(ert-test-deftest-lexical-binding-t): New test.
Allowing &rest without a variable name following turned out not to be
very useful, and it never worked properly. Disallow it.
* lisp/emacs-lisp/bytecomp.el (byte-compile-check-lambda-list):
* src/eval.c (funcall_lambda):
Signal error for &rest without variable name.
* doc/lispref/functions.texi (Argument List): Adjust manual.
* etc/NEWS (file): Announce.
* test/src/eval-tests.el (eval-tests--bugs-24912-and-24913):
Extend test, also checking with and without lexical binding.
(eval-tests-accept-empty-optional-rest): Reduce to...
(eval-tests-accept-empty-optional): ...this, again checking
with and without lexical binding.
* lisp/emacs-lisp/derived.el (define-derived-mode): Doc fixes;
correctly mention that the mode name is used in the mode line, clarify
argument types, and how the mode hook is named. (Bug17567)
(derived-mode-hook-name, derived-mode-map-name)
(derived-mode-syntax-table-name, derived-mode-abbrev-table-name):
Clarify that argument is a symbol.
