Paul Eggert [Mon, 5 Jun 2017 06:52:10 +0000 (23:52 -0700)]
SCHARS and STRING_BYTES are nonnegative
Tell the compiler that SCHARS and STRING_BYTES are nonnegative, in
the hopes that this will optimize a bit better. Also, check this
at runtime if ENABLE_CHECKING.
* src/lisp.h (SCHARS, STRING_BYTES):
eassume that these functions return nonnegative values.
(STRING_SET_CHARS) [ENABLE_CHECKING]:
eassert that newsize is nonnegative.
* src/emacs-module.c (Fmodule_load, funcall_module): Adapt callers.
(finalize_environment): Add parameter for public part of the
environment, like 'initialize_environment'. Add assertions.
Fix a FIXME in emacs-module.c. Put the printing into print.c, like
other types.
* src/print.c (print_vectorlike): Add code to print module functions.
* src/emacs-module.c (funcall_module): Stop calling
'module_format_fun_env'. Now that module functions are first-class
objects, they can be added to signal data directly.
(module_handle_signal): Remove now-unused function
'module_format_fun_env'.
* test/src/emacs-module-tests.el (mod-test-sum-test): Adapt unit test.
* src/eval.c (funcall_lambda): Adapt call to changed signature of
'funcall_module'.
* src/emacs-module.c (MODULE_FUNCTION_BEGIN_NO_CATCH): New helper
macro.
(MODULE_FUNCTION_BEGIN, module_type_of, module_is_not_nil, module_eq):
Use it.
Paul Eggert [Sun, 4 Jun 2017 15:39:37 +0000 (08:39 -0700)]
Tune ‘format’ after recent fix
* doc/lispref/strings.texi (Formatting Strings):
* src/editfns.c (Fformat): Format field numbers no longer need
to be unique, reverting the previous doc change since that has
now been fixed. Also, document that %% should not have modifiers.
* src/editfns.c (styled_format): Improve performance. Remove
the need for the new prepass over the format string, by using
a typically-more-generous bound for the info array size.
Initialize the info array lazily. Move string inspection to
the same area to help caching. Avoid the need for a
converted_to_string bitfield by using EQ. Cache arg in a
local and avoid some potential aliasing issues to help the
compiler. Info array is now 0-origin, not 1-origin.
Previously styled_format overwrite the argument vector. This is no
longer possible because there might be more than one specification per
argument. Use the existing auxiliary info array instead.
* src/editfns.c (styled_format): Record arguments in the info
structure instead of overwriting them.
* test/src/editfns-tests.el (format-with-field): Add unit test.
Glenn Morris [Sat, 3 Jun 2017 00:06:12 +0000 (20:06 -0400)]
Add watch for password back to inferior python comint filter
It was removed along with other items for speed (bug#16875),
but doesn't seem to have been causing an issue, and it's useful to
have it there (bug#27154).
* lisp/progmodes/python.el (inferior-python-mode):
Add comint-watch-for-password-prompt to comint-output-filter-functions.
Ryan [Sat, 3 Jun 2017 00:00:49 +0000 (20:00 -0400)]
Use completing-read-default in tmm-prompt
tmm uses completing-read, but customizes its behavior so much
that any alternative completing-read-function will almost
certainly break it. For example, both ido-ubiquitous and ivy have
special code to deactivate themselves for tmm.
* lisp/tmm.el (tmm-prompt): Use completing-read-default instead of
completing-read. (Bug#27193)
Paul Eggert [Thu, 1 Jun 2017 23:03:12 +0000 (16:03 -0700)]
Limit format fields to more POSIX-like spec
* doc/lispref/strings.texi (Formatting Strings):
Don’t allow mixing numbered with unnumbered format specs.
* src/editfns.c (styled_format): Don’t bother checking for field 0,
since it doesn’t crash and the behavior is not specified.
* test/src/editfns-tests.el (format-with-field): Adjust tests to
match current doc. Add more tests for out-of-range fields.
Paul Eggert [Thu, 1 Jun 2017 23:03:12 +0000 (16:03 -0700)]
Improve performance by avoiding strtoumax
This made (string-to-number "10") 20% faster on my old desktop,
an AMD Phenom II X4 910e running Fedora 25 x86-64.
* admin/merge-gnulib (GNULIB_MODULES): Remove strtoumax.
* lib/gnulib.mk.in, m4/gnulib-comp.m4: Regenerate.
* lib/strtoul.c, lib/strtoull.c, lib/strtoumax.c, m4/strtoull.m4:
* m4/strtoumax.m4: Remove.
* src/editfns.c (str2num): New function.
(styled_format): Use it instead of strtoumax. Use ptrdiff_t
instead of uintmax_t. Check for integer overflow.
* src/lread.c (LEAD_INT, DOT_CHAR, TRAIL_INT, E_EXP):
Move to private scope and make them enums.
(string_to_number): Compute integer value directly during
first pass instead of revisiting it with strtoumax later.
Paul Eggert [Thu, 1 Jun 2017 05:09:39 +0000 (22:09 -0700)]
Minor improvements to format field numbers
* src/editfns.c (styled_format): Allow field numbers in a %% spec.
No need for a special diagnostic for field numbers greater than
PTRDIFF_MAX. Reword diagnostic for field 0.
* test/src/editfns-tests.el (format-with-field): Adjust to match.
Philipp Stephani [Wed, 31 May 2017 22:09:43 +0000 (00:09 +0200)]
Implement field numbers in format strings
A field number explicitly specifies the argument to be formatted.
This is especially important for potential localization work, since
grammars of various languages dictate different word orders.
* src/editfns.c (Fformat): Update documentation.
(styled_format): Implement field numbers.
* doc/lispref/strings.texi (Formatting Strings): Document field numbers.
Limit scope of local overriding-terminal-local-map
The function `binding' may call isearch-done, which globally sets
overriding-terminal-local-map to nil (Bug#23007).
* lisp/isearch.el (isearch-mouse-2): Don't bind
overriding-terminal-local-map around the call to `binding'.
Stephen Berman [Thu, 1 Jun 2017 20:15:50 +0000 (22:15 +0200)]
Correct and isolate the todo-mode test environment
This avoids having to set todo-mode variables globally in the test
file and prevents any exisiting user todo-mode files from influencing
the tests.
* test/lisp/calendar/todo-mode-tests.el:
(with-todo-test): New macro.
(todo-test-todo-quit01, todo-test-todo-quit02)
(todo-test-item-highlighting): Use it.
* src/xdisp.c (get_next_display_element): Dispatch used format string
for unprintables based on new display-raw-bytes-as-hex variable.
(display-raw-bytes-as-hex): New variable. (Bug#27122)
* lisp/cus-start.el: Add defcustom form for display-raw-bytes-as-hex.
* doc/emacs/display.texi: Document the new variable.
* etc/NEWS: Mention display-raw-bytes-as-hex.
* test/manual/redisplay-testsuite.el (test-redisplay-5-toggle)
(test-redisplay-5): New tests.
(test-redisplay): Call test-redisplay-5.
Eli Zaretskii [Thu, 1 Jun 2017 18:12:39 +0000 (21:12 +0300)]
Add customizable to display raw bytes as hex
* src/xdisp.c (get_next_display_element): Dispatch used format string
for unprintables based on new display-raw-bytes-as-hex variable.
(display-raw-bytes-as-hex): New variable. (Bug#27122)
* lisp/cus-start.el: Add defcustom form for display-raw-bytes-as-hex.
* doc/emacs/display.texi: Document the new variable.
* etc/NEWS: Mention display-raw-bytes-as-hex.
* test/manual/redisplay-testsuite.el (test-redisplay-5-toggle)
(test-redisplay-5): New tests.
(test-redisplay): Call test-redisplay-5.
Eli Zaretskii [Thu, 1 Jun 2017 14:55:25 +0000 (17:55 +0300)]
Fix linum under text-scaling when leuven-theme is used
* etc/themes/leuven-theme.el (linum): Make the 'linum' face
inherit from 'default' and 'shadow', so that margins are enlarged
as expected under text-scaling.
Fix memory leak of cwd string in emacsclient (Bug#26628)
* lib-src/emacsclient.c (main): emacsclient retrieves the current
working directory using get_current_dir_name which returns a newly
allocated string. Make sure this string is freed before exiting.
Paul Eggert [Wed, 31 May 2017 23:50:08 +0000 (16:50 -0700)]
Fix bug with "%%" in error format
* src/doprnt.c (doprnt): Format "%%" correctly.
Problem reported by Philipp Stephani in:
http://lists.gnu.org/archive/html/emacs-devel/2017-05/msg00901.html
Eli Zaretskii [Wed, 31 May 2017 16:01:31 +0000 (19:01 +0300)]
Support lower bound on hscrolling when only current line scrolls
* doc/emacs/display.texi (Horizontal Scrolling): Document the new
mode of auto-hscrolling only the current line.
* src/xdisp.c (init_iterator): When hscrolling only the
current line, apply the window's min_hscroll here, so that
non-current lines will be hscrolled by that minimum.
Suggested by Stephen Berman <stephen.berman@gmx.net>.
(hscroll_window_tree): Account for window's min_hscroll when
deciding whether to recompute the hscroll.
(display_line): Subtract window's min_hscroll from x_incr, as that
was already accounted for in init_iterator. (Bug#27008)
Noam Postavsky [Sun, 28 May 2017 21:01:05 +0000 (17:01 -0400)]
cl-print: handle circular objects when `print-circle' is nil (Bug#27117)
* lisp/emacs-lisp/cl-print.el (cl-print--currently-printing): New variable.
(cl-print-object): When `print-circle' is nil, bind it to a list of
objects that are currently printing to avoid printing the same object
endlessly.
* test/lisp/emacs-lisp/cl-print-tests.el (cl-print-circle): New test.
Noam Postavsky [Wed, 31 May 2017 03:31:18 +0000 (23:31 -0400)]
Further simplify test/Makefile, optionally load elc tests
* test/Makefile.in: Use make's error ignoring feature instead of
suppressing test errors with shell. Compile test files in the main
make invocation instead of a recursive 'make' call. Optionally load
.elc test files if TEST_LOAD_EL is set to something other than 'yes'.
Remove obsolete commentary.
Glenn Morris [Wed, 31 May 2017 01:01:56 +0000 (21:01 -0400)]
Avoid subr test failure when source dir has multiple names
* test/lisp/subr-tests.el (subr-tests--this-file):
Use the true name of the file. The following test does a string
comparison of this value with that from method-files, which uses
load-history, which contains true names.
Dmitry Gutov [Tue, 30 May 2017 22:29:34 +0000 (01:29 +0300)]
Extract eldoc--supported-p
* lisp/emacs-lisp/eldoc.el (eldoc--supported-p): New function.
(turn-on-eldoc-mode, eldoc-mode): Use it.
(http://lists.gnu.org/archive/html/emacs-devel/2017-05/msg00865.html)
Paul Eggert [Tue, 30 May 2017 16:27:33 +0000 (09:27 -0700)]
Skip .#* temporaries when finding sources
Without this patch, ‘make check’ can fail with the diagnostic
‘invalid syntax in conditional’ if there is an Emacs temporary
file whose name starts with ‘.#’, because the ‘#’ is treated as
the start of a Make comment.
* lisp/Makefile.in (loaddefs, tagsfiles, check-defun-deps):
* test/Makefile.in (ELFILES):
Skip files starting with ‘.’, so that the .#* files do not cause
trouble. (We cannot easily skip just files starting with ‘.#’,
since ‘#’ starts a Make comment!)
Alan Mackenzie [Tue, 30 May 2017 16:21:31 +0000 (16:21 +0000)]
Fix the mouse help/key map on the "%p" part of the mode line.
* lisp/bindings.el (mode-line-percent-position): give it a
`risky-local-variable' property.
(mode-line-position): correct the quoting on the mode-line-percent-position
part of the variable, allowing the properties to be properly recognized.
Alan Mackenzie [Tue, 30 May 2017 16:21:31 +0000 (16:21 +0000)]
Fix the mouse help/key map on the "%p" part of the mode line.
* lisp/bindings.el (mode-line-percent-position): give it a
`risky-local-variable' property.
(mode-line-position): correct the quoting on the mode-line-percent-position
part of the variable, allowing the properties to be properly recognized.
Paul Eggert [Tue, 30 May 2017 00:15:11 +0000 (17:15 -0700)]
Improve .gdbinit Lisp value pretty-printing
* src/.gdbinit (to_string): Use an unsigned representation for
Lisp values, as requested by Eli Zaretskii (Bug#27098).
Also, use "make_number(N)" for Lisp integers.
Dmitry Gutov [Mon, 29 May 2017 23:55:28 +0000 (02:55 +0300)]
Turn global-eldoc-mode into a globalized minor mode
* lisp/emacs-lisp/eldoc.el (global-eldoc-mode):
Turn into globalized mode (bug#19853).
(turn-on-eldoc-mode): Make it into a wrapper instead of alias.
(eldoc-mode): Only show the message when called interactively.
Stephen Berman [Mon, 29 May 2017 21:49:41 +0000 (23:49 +0200)]
Add initial tests for todo-mode.el
*test/lisp/calendar/todo-mode-tests.el:
*test/lisp/calendar/todo-mode-resources/todo-test-1.toda:
*test/lisp/calendar/todo-mode-resources/todo-test-1.todo: New files.
* .gitattributes: Ignore trailing whitespace in todo-mode test
data files, since it is part of the todo-mode file format.
Stephen Berman [Mon, 29 May 2017 18:26:49 +0000 (20:26 +0200)]
Make `todo-toggle-item-highlighting' work on multiline items (bug#27133)
* lisp/calendar/todo-mode.el (todo-hl-line-range): New named function,
replacing an anonymous function for the sake of `describe-variable'.
(todo-modes-set-2): Use it as buffer-local value of hl-line-range-function
and remove boundp test of this variable, so its value is available on
invoking `todo-toggle-item-highlighting'.
Noam Postavsky [Mon, 22 May 2017 16:36:20 +0000 (12:36 -0400)]
; Fix backslashes in python-tests
* test/lisp/progmodes/python-tests.el (python-indent-after-backslash-1)
(python-indent-after-backslash-2)
(python-indent-after-backslash-3)
(python-indent-after-backslash-4)
(python-indent-after-backslash-5)
(python-nav-beginning-of-statement-1)
(python-nav-end-of-statement-1)
(python-nav-forward-statement-1)
(python-nav-backward-statement-1)
(python-nav-backward-statement-2)
(python-info-statement-starts-block-p-2)
(python-info-statement-ends-block-p-2)
(python-info-beginning-of-statement-p-2)
(python-info-end-of-statement-p-2)
(python-info-beginning-of-block-p-2)
(python-info-end-of-block-p-2)
(python-info-line-ends-backslash-p-1)
(python-info-beginning-of-backslash-1)
(python-info-continuation-line-p-1)
(python-info-block-continuation-line-p-1)
(python-info-assignment-statement-p-1)
(python-info-assignment-continuation-line-p-1): Backslashes in
literals should be doubled only once to produce one backslash in the
buffer. If there backslashes inside a Python string literal in a Lisp
literal, that would need to be doubled twice, but there are no such
cases. Note that `python-tests-looking-at' takes a plain string, not
a regexp.
Jules Tamagnan [Thu, 27 Oct 2016 22:03:31 +0000 (15:03 -0700)]
Comply with pep 8 style guide for backslash in assignment (Bug#24809)
* lisp/progmodes/python.el (python-indent--calculate-indentation):
Increase indent by `python-indent-offset' after
`:after-backslash-assignment-continuation'.
Wilfred Hughes [Mon, 29 May 2017 00:00:05 +0000 (01:00 +0100)]
Ensure button-get works in any buffer
* lisp/button.el (button-get): Previously we assumed that button-get
was called in the buffer containing the button. In other buffers,
button-get always returned nil. Fix this by passing the relevant
buffer from the marker.
Dmitry Gutov [Sun, 28 May 2017 23:55:42 +0000 (02:55 +0300)]
Signal error if find-grep returns a nonzero status
* lisp/progmodes/xref.el (xref-collect-matches): Signal error
if find-grep returns a nonzero status (bug#23451). Remove the
comment: even if some output is present, a non-zero status
means something went wrong and it can't be relied upon.
Stephen Berman [Sun, 28 May 2017 20:52:05 +0000 (22:52 +0200)]
Make sure exiting todo-mode buffer buries it (bug#27121)
This failed due to commit ea3ae33b from 2013-05-16, which prevented
quitting todo-mode buffer after visiting todo-archive buffer from
making the archive buffer current again. Avoid this now by simply
killing the archive buffer, since there's no need to keep it a live
buffer. Consequently, quitting a todo-mode buffer can now use
bury-buffer without an argument, which ensures that is will not
becomes current on quitting the buffer that replaced it in the window.
* lisp/calendar/todo-mode.el (todo-quit): Kill todo-archive-mode
buffer instead of burying it. This now allows exiting the
todo-mode buffer by bury-buffer without an argument, so do that.
Michael Albinus [Sun, 28 May 2017 21:44:10 +0000 (23:44 +0200)]
Some tweaks, almost all for Tramp adb method
* lisp/net/tramp-adb.el (tramp-adb-parse-device-names):
Use `make-tramp-file-name'.
(tramp-adb-get-device): Use `tramp-file-name-port-or-default'.
(tramp-adb-maybe-open-connection): Set "prompt" property.
(tramp-adb-wait-for-output): Use it.
* lisp/net/tramp-cache.el (tramp-cache-print): Use `elt'.
(tramp-dump-connection-properties): Check also that there are
properties to be saved. Don't save "started" property of
"ftp" method.
* lisp/net/tramp-gvfs.el (tramp-gvfs-handle-expand-file-name):
Use `make-tramp-file-name'.
* lisp/net/tramp.el (tramp-remote-file-name-spec-regexp):
Host could be empty.
(tramp-file-name-port-or-default): New defun.
(tramp-dissect-file-name): Simplify `make-tramp-file-name' call.
(tramp-handle-file-name-case-insensitive-p): Use a progress reporter.
(tramp-call-process, tramp-call-process-region):
Use `make-tramp-file-name'.
* test/lisp/net/tramp-tests.el (tramp-test03-file-name-defaults):
Revert change from 2017-05-24.
(tramp-test05-expand-file-name-relative): Let it also pass for
"adb" method.
Jürgen Hötzel [Sun, 28 May 2017 11:06:53 +0000 (13:06 +0200)]
Fix Tramp for Android 7
* tramp-adb.el (tramp-adb-ls-toolbox-regexp):
Username part of prompt is empty on Android 7.
(tramp-adb-ls-toolbox-regexp):
Ignore addition links column on Android 7.
(tramp-adb-get-ls-command):
Dont use --color=none when using toybox (Android 7). It's not
possible to disable coloring explicitly for toybox ls.
Paul Eggert [Sat, 27 May 2017 17:54:25 +0000 (10:54 -0700)]
Depromiscuify inotify with IN_MASK_ADD
Use IN_MASK_ADD instead of using a no-longer-promiscuous-enough
mask. This simplifies the code and restores the ability to
use IN_ACCESS, IN_CLOSE_WRITE, IN_CLOSE_NOWRITE, and IN_OPEN
in some cases (Bug#26973).
* src/inotify.c (INOTIFY_DEFAULT_MASK): Remove.
(Finotify_add_watch): Use IN_MASK_ADD instead.