* lisp/calc/calc.el (calc-unread-command): Prevent recording
key twice when defining a keyboard macro. Patch by Christoph
Arenz <tiga.arenz@web.de>.
Copyright-paperwork-exempt: yes
`auth-source-search' can be called with an integer port number from
`network-stream-certificate`, while the backend implementation doesn't
allow non-string attributes.
* lisp/auth-source.el (auth-source-plstore-search): Ensure attributes
are string.
* lisp/auth-source.el (auth-source--obfuscate): Convert the string
to bytes before obfuscating.
(auth-source--deobfuscate): Convert back into a string (bug#37758).
The problem originating this report was:
-------------
In [15]: "
File "<ipython-input-15-3b7a06bb1102>", line 1
"
^
SyntaxError: EOL while scanning string literal
In [16]: string face still here"
-------------
This happens because
python-shell-font-lock-comint-output-filter-function is called twice,
first for the error output and then for the "In [16]: " part. The
first time python-shell-comint-end-of-output-p returns nil since
we're *not* at the end of an input prompt. The second time it returns
0 since we're at the end of *just* an input prompt. So we don't call
python-shell-font-lock-cleanup-buffer either time.
The current code is relying in a very weak rule: it considers "just an
input prompt" to be a continuation prompt. Another unreliable aspect
of the current rule is that sometimes
(python-shell-comint-end-of-output-p (ansi-color-filter-apply output))
returns 1 and not 0 for continuation prompts. In short, the rule does
a very poor job identifying continuations.
* lisp/progmodes/python.el (python-shell-font-lock-cleanup-buffer):
Don't check for empty OUTPUT since python-shell-comint-end-of-output-p
returns nil for that anyway. Don't check for
python-shell-comint-end-of-output-p returning a specific number
because it's unreliable, just check for any non-nil. Identify
continuation prompts by looking for "...".
* lisp/tab-bar.el (tab-bar-mode): Set display property with image files
on tab-bar-new-button and tab-bar-close-button when tab-bar-mode is enabled
the first time.
(tab-bar-new-button, tab-bar-close-button): Use default values
without display image properties. (Bug#37685)
* lisp/files.el (find-file-visit-truename): Remove the
safe-local-variable put, because the variable is checked too late
to actually make a difference (bug#36830).
* leim/leim-ext.el ("quail/PY-b5"): Use U+3007 instead of U+25CB for
ideographic zero. U+25CB (used previously) is only one of the several
lookalikes (bug#3179).
New face attribute :extend to control the face extension after
EOL.
necessary, # especially if it merges an updated upstream into a topic
branch. # # Lines starting with '#' will be ignored, and an empty
message aborts # the commit.
src/xdisp.c (face_at_pos): Removed code to that modified
it->start_of_box_run_p and it->face_box_p. This function should not
modify it.
(handle_face_prop): Added code to update it->start_of_box_run_p and
it->face_box_p.
(underlying_face_id): Input IT made CONST.
* src/xdisp.c (append_space_for_newline): Modified to add the space
with the last face also in terminal interface.
(fill_column_indicator_column): Modified to group more conditions.
(extend_face_to_end_of_line): Simplified code in
fill_column_indicator to use the new function.
* src/xdisp.c (extend_face_to_end_of_line): Added a call to
handle_face_prop_general to extend face after EOL according to
parameter :extend in the faces.
* src/xdisp.c (handle_face_prop_general): New function to specialize
handle_face_prop with ATTR_FILTER.
* src/dispextern.h (face_at_buffer_position): Added
LFACE_ATTRIBUTE_INDEX.
* src/xfaces.c (merge_face_ref): Added LFACE_ATTRIBUTE_INDEX to merge
conditionally.
* src/dispextern.h (enum face_underline_type): Add FACE_NO_UNDERLINE = 0.
(struct face): Merge `underline_type` and `underline_p` into `underline`.
Update all the occurrences in src/xdisp.c, src/xfaces.c, and src/xterm.c.
