procyberian/emacs
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Don't set internal_last_event_frame to nil

Browse source 
* src/frame.c (do_switch_frame): Do not set internal_last_event_frame to
nil on ttys.
This commit is contained in:
Gerd Möllmann 2024-10-21 11:21:17 +02:00
parent 6aa203bb04
commit 0012e555d9
1 changed files with 23 additions and 2 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

25
src/frame.c
View file

@ -1792,8 +1792,29 @@ do_switch_frame (Lisp_Object frame, int track, int for_deletion, Lisp_Object nor
(select-window (frame-root-window (make-frame))) doesn't end up
with your typing being interpreted in the new frame instead of
the one you're actually typing in. */
if (!frame_ancestor_p (f, sf))
internal_last_event_frame = Qnil;
/* FIXME/tty: I don't understand this. Setting the last event
frame to nil leads to switch-frame events being generated even
if they normally wouldn't because the frame in question equals
selected-frame. This leads to problems at least on ttys.
Imagine that we have functions on post-command-hook that use
with-selected-window (which is the case with Vertico-Posframe),
Secondly, let these functions select/restore windows on different
frames, so there will be select-frame calls with different frames
during the execution of post-command-hook. Setting
internal_last_event_frame to nil makes these select-frame calls
potentially generate switch-frame events (but only in one direction
(frame_ancestor_p), which I also don't understand).
These switch-frame events form an endless loop in command_loop_1 by
running post-command-hook, which generates switch-frame events,
which command_loop_1 finds (bound to '#ignore) and executes, which
again runs post-command-hook etc. ad infinitum. */
if (!is_tty_frame (f))
if (!frame_ancestor_p (f, sf))
internal_last_event_frame = Qnil;
return frame;
}