* doc/lispref/os.texi (Desktop Notifications): Document support
for `:resident'.
* java/org/gnu/emacs/EmacsService.java (cancelNotification):
* src/android.c (android_init_emacs_service):
* src/android.h (struct android_emacs_service): New function.
* src/androidselect.c (android_notifications_notify_1)
(Fandroid_notifications_notify): New parameter QCresident; save
it within notification lists.
(android_notification_deleted, android_notification_action):
Adjust for changes to the format of notification lists and
cancel non-resident notifications when an action is selected.
(syms_of_androidselect): <QCresident>: New symbol.
* doc/lispref/os.texi (Desktop Notifications): Document that
:on-cancel, :on-action and :actions are now supported on
Android.
* java/org/gnu/emacs/EmacsActivity.java (onNewIntent): New
function.
* java/org/gnu/emacs/EmacsDesktopNotification.java
(NOTIFICATION_ACTION, NOTIFICATION_TAG, NOTIFICATION_DISMISSED):
New constants. <actions, titles>: New fields.
(insertActions): New function.
(display1, display): Insert actions on Jelly Bean and up, and
arrange to be notified when the notification is dismissed.
(CancellationReceiver): New class.
* java/org/gnu/emacs/EmacsNative.java (sendNotificationDeleted)
(sendNotificationAction): New functions.
* src/android.c (sendDndDrag, sendDndUri, sendDndText): Correct
return types.
(sendNotificationDeleted, sendNotificationAction)
(android_exception_check_5, android_exception_check_6): New
functions.
* src/android.h:
* src/androidgui.h (struct android_notification_event): New
structure.
(union android_event): New member for notification events.
* src/androidselect.c (android_init_emacs_desktop_notification):
Update JNI signatures.
(android_notifications_notify_1, Fandroid_notifications_notify):
New arguments ACTIONS, ACTION_CB and CANCEL_CB. Convert and
record them as appropriate.
(android_notification_deleted, android_notification_action): New
functions.
(syms_of_androidselect): Prepare a hash table of outstanding
notifications.
<QCactions, QCon_action, QCon_cancel> New defsyms.
* src/androidterm.c (handle_one_android_event)
<ANDROID_NOTIFICATION_DELETED>
<ANDROID_NOTIFICATION_ACTION>: Dispatch event contents to
androidselect.c for processing.
* src/androidterm.h:
* src/androidvfs.c (java_string_class): Export.
* src/keyboard.c (kbd_buffer_get_event) <NOTIFICATION_EVENT>:
Call callback specified by the event.
* src/termhooks.h (enum event_kind) [HAVE_ANDROID]: New
enum NOTIFICATION_EVENT.
Also, attempt to document the intent better.
Problem reported by Alan Mackenzie (Bug#69684).
* src/data.c (Fbare_symbol): Do not signal if the SYM is a symbol
with position and symbols-with-pos-enabled is nil. Instead,
ignore symbols-with-pos-enabled, as that was the intent.
* test/src/data-tests.el (data-tests-bare-symbol):
New test, to help prevent this bug from reoccurring.
* lisp/emacs-lisp/pp.el (pp--quoted-or-unquoted-form-p): New helper
function.
(pp--insert-lisp): Take care of quoted, backquoted and
unquoted expressions; print using an recursive call.
(pp--format-list): Exclude more cases from printing as a function call
by default. Print lists whose second-last element is an (un)quoting
symbol using dotted list syntax; e.g. (a b . ,c) instead of (a b \, c).
Previously it always returned t or nil for NEW-WINDOW-FLAG, but now it
can return the actual prefix arg when appropriate. This lets functions
for 'browse-url-browser-function' consult it and do more things than
just open a new window or not (for example, you could use "C--" as the
prefix arg to do something special in a custom function).
* lisp/net/browse-url.el (browse-url-interactive-arg): Use 'xor' to
adjust the value of 'current-prefix-arg'.
(browse-url): Update docstring.
* admin/syncdoc-type-hierarchy.el: Delay loading `org-table` so as
not to "pollute" the table with Org-specific types.
(syncdoc-all-types): Sort the types topologically from the root.
(syncdoc-hierarchy): Use `cl--class-parents` instead if
`comp--direct-supertypes` so we don't depend on `comp-cstr`.
(syncdoc-make-type-table): Sort the table so supertypes always come before
their subtypes.
(syncdoc-make-type-table): Require `org-table` here.
* doc/lispref/elisp_type_hierarchy.jpg:
* doc/lispref/elisp_type_hierarchy.txt: Refresh.
'read-char' will no longer return -1 as of
ac82baea1c41ec974ad49f2861ae6c06bda2b4ed. This switches to a cleaner
method of detecting whether the end of a keyboard macro has been
reached.
* lisp/calc/calc-prog.el (calc--at-end-of-kmacro-p): New function.
(calc-kbd-skip-to-else-if): Use the function.
Co-authored-by: Stefan Monnier <monnier@iro.umontreal.ca>
* lisp/subr.el (read-char-choice-with-read-key):
* lisp/net/dbus.el (dbus-call-method):
Remove workarounds for the bug fixed in the previous commit
ac82baea1c41ec974ad49f2861ae6c06bda2b4ed, where 'read-event',
'read-char' and 'read-char-exclusively' could return wrongly -1.
In the case of lisp/dbus.el, this reverts commit
7177393826.
This fixes a bug that could make 'read-event', 'read-char', and
'read-char-exclusive' erroneously return -1, an internal magic return
value of 'read_char' leaked from C to lisp. Instead of returning -1, the
aforementioned lisp functions now transparently continue reading
available input (e.g., from the keyboard) when reaching the end of a
keyboard macro.
* src/keyboard.c (read_char, read_key_sequence): Move handling
of the end of a keyboard macro from 'read_char' to its caller
'read_key_sequence', which is the only caller that can
meaningfully deal with this case.
* src/macros.c (Fexecute_kbd_macro): Document how the end of keyboard
macro is processed.
* etc/NEWS: Announce this change.
* src/keyboard.c, src/keyboard.h (requeued_events_pending_p): Add
function 'requeued_events_pending_p' (whose name was made available in
the previous commit). As opposed to the previous function with the same
name, the new function covers both command and other events.
* src/keyboard.c (Finput_pending_p): Use the new function.
* src/keyboard.c, src/keyboard.h (requeued_events_pending_p): Rename to
'requeued_command_events_pending_p' to clarify that the function covers
only command events. Fix wrong comment that claimed that the function
was unused.
* src/process.c (wait_reading_process_output): Update caller to use the
new name.
* lisp/erc/erc-stamp.el (erc-stamp--lr-date-on-pre-modify): Remove
disruptive assertion for now.
(erc-stamp--time-as-day): Attempt to fix date being rewound by a whole
day when daylight saving time is in effect. Do this by forcing the
`dst' slot of the `decoded-time' object to -1 and the `zone' to nil.
* lisp/erc/erc-goodies.el (erc-spoiler-face): Redefine role and redo
definition to inherit from `erc-control-default-face'.
(erc-controls-propertize): Include `cursor-face' in the applied hover
properties for spoiler text, and ensure they aren't clobbered by other
built-in modules, like `button'.
(Bug#69597)
Copyright-paperwork-exempt: yes
* lisp/erc/erc-button.el (erc-button-remove-old-buttons): Restore
original `mouse-face' values in areas marked as important after
clobbering.
* lisp/erc/erc.el (erc--reserve-important-text-props): New function.
(erc--restore-important-text-props): New function.
* test/lisp/erc/erc-tests.el (erc--restore-important-text-props): New
test.
(Bug#69597)
* lisp/erc/erc-goodies.el (erc-inverse-face): Specify face attribute
`:inverse-video' (née :reverse-video) to swap foreground and
background colors over affected intervals, as per
https://modern.ircdocs.horse/formatting#reverse-color.
(erc-control-default-fg erc-control-default-bg): New faces for IRC
color-code number 99. Ignore the ERC convention of prefixing
control-code-derived faces with "fg:" and "bg:" because it doesn't
comport with modern sensibilities, which demand identifiers normally
be namespaced.
(erc-get-bg-color-face, erc-get-fg-color-face): Return new, dedicated
faces instead of `default', and don't nest them in a list.
* test/lisp/erc/erc-goodies-tests.el
(erc-controls-highlight--inverse): Redo completely, asserting behavior
described in the spec linked to above.
(erc-controls-highlight--spoilers): New test based on the body of the
old `erc-controls-highlight--inverse', except without shadowing
`erc-insert-modify-hook' with an unrealistic, idealized value. Adjust
expected buffer state to reflect the new role of
`erc-spoiler-face'. (Bug#69597)
Now that we use extra-parents to group alternative major modes,
some tables can be simplified to mention only the group's leader.
* lisp/align.el (align-c++-modes, align-rules-list):
Don't bother listing TS alternatives.
(align-perl-modes): Don't bother listing CPerl alternative.
* lisp/info-look.el (perl-mode): Simplify.
* lisp/cedet/semantic/symref/grep.el (semantic-symref-filepattern-alist):
Don't bother listing TS alternatives.
* lisp/emulation/viper.el (viper-vi-state-mode-list): Don't bother
listing CPerl alternative.
* lisp/progmodes/gud.el (gud-tooltip-activate-mouse-motions-if-enabled):
Take into account the modes hierarchy.
(gud-tooltip-modes): Don't bother listing TS alternatives.
* .dir-locals.el (c-ts-mode): Simplify.
* src/sysdep.c (handle_sigsegv): Return after restoring the
original signal handler, which should proceed to call debuggerd
to generate a tombstone.
(init_sigsegv): Save the original signal handler on Android, to
be restored after a signal is received.
(init_signals): Call init_sigsegv on Android.
* lisp/progmodes/etags.el (complete-tag): Bind
'completion-ignore-case', so that 'completion-in-region' is
affected by it. This fixes a bug made in 30 Apr 2010, when this
function was refactored to use
'tags-completion-at-point-function'. Reported by Morgan Willcock
<morgan@ice9.digital>.
Not sure why earlier tests did not catch it, but there are more
places where we bump into problems because `eieio--class-precedence-list`
now returns also non-EIEIO classes.
* lisp/obsolete/eieio-compat.el
(eieio--generic-static-object-generalizer):
* lisp/emacs-lisp/eieio-core.el (eieio--generic-generalizer)
(eieio--generic-subclass-specializers): Handle non-EIEIO parents.
* test/lisp/emacs-lisp/eieio-tests/eieio-test-methodinvoke.el
(eieio-test-method-order-list-7): Adjust test.
* lisp/emacs-lisp/eieio-core.el (eieio--class-precedence-c3)
(eieio--class-precedence-dfs, eieio--class-precedence-bfs): Use
`cl--class-parents` since some of the parents aren't EIEIO classes.
Now that built-in types have classes that describe their
relationships exactly like struct/eieio/oclosure classes,
we can the code that navigates that DAG.
* lisp/emacs-lisp/cl-generic.el (cl--generic-struct-tag): Move to
`eieio-core.el`.
(cl--generic-type-specializers): Rename from
`cl--generic-struct-specializers`. Make it work for any class.
(cl--generic-typeof-generalizer, cl--generic-oclosure-generalizer): Use it.
(cl--generic-struct-generalizer): Delete generalizer.
(cl-generic-generalizers :extra "cl-struct"): Delete method.
(prefill 0 cl--generic-generalizer): Move to after the typeof.
(cl-generic-generalizers :extra "typeof"): Rewrite to use
classes rather than `cl--all-builtin-types`.
(cl-generic--oclosure-specializers): Delete function.
* lisp/emacs-lisp/cl-preloaded.el (cl--direct-supertypes-of-type)
(cl--typeof-types, cl--all-builtin-types): Delete constants.
* lisp/emacs-lisp/comp-cstr.el (comp--typeof-builtin-types):
Delete constant.
(comp--cl-class-hierarchy): Simplify.
(comp--compute-typeof-types): Simplify now that
`comp--cl-class-hierarchy` and `comp--all-classes` work for built-in
types as well.
(comp--direct-supertypes): Just use `cl--class-parents`.
(comp-supertypes): Simplify since typeof-types should now be complete.
* lisp/emacs-lisp/eieio-core.el (eieio-defclass-autoload):
Use `superclasses` argument, so we can find parents before it's loaded.
(eieio--class-precedence-c3, eieio--class-precedence-dfs):
Don't add a `eieio-default-superclass` parent any more.
(eieio--class/struct-parents): Delete function.
(eieio--class-precedence-bfs): Use `eieio--class-parents` instead.
Don't stop when reaching `eieio-default-superclass`.
(cl--generic-struct-tag): Move from `cl-generic.el`.
* lisp/emacs-lisp/eieio-core.el (eieio-defclass-internal):
Always put a parent in the `parents` slot of the class.
* lisp/emacs-lisp/eieio.el (eieio-class-parents): Remove the
`eieio-default-superclass` if it's the only parent.
(child-of-class-p): Handle all classes in the parents.
(eieio-default-superclass): Adjust docstring.
* lisp/emacs-lisp/cl-preloaded.el (cl--struct-register-child):
Register child only in struct parents.
(cl-struct-define): Put the "type" as parent of parentless :type structs.
Copy slots only from struct parent classes.
(cl-structure-object): Set (manually) its parent to `record`
and remove assertion that it has no parents.
Add classes describing the built-in types.
* lisp/emacs-lisp/cl-preloaded.el (built-in-class): New type.
(cl--define-built-in-type): New aux macro.
(all built-in types): "Define" them with it.
(cl--builtin-type-p): New aux function.
(cl--struct-name-p): Use it.
(cl--direct-supertypes-of-type, cl--typeof-types, cl--all-builtin-types):
Move the definitions to after the built-in classes are defined,
and rewrite to make use of those classes.
* lisp/emacs-lisp/cl-extra.el (cl-describe-type):
Accept two (unused) optional args, for use with `describe-symbol-backends`.
(describe-symbol-backends): Simplify accordingly and
add ourselves at the end.
(cl--class-children): New function.
(cl--describe-class): Use it. Also don't show a silly empty list of slots
for the built-in types.
