Kovid Goyal
aaf0dea8dc
Use a single wrapper binary for all command line tools
...
This is because Go has a multi megabyte overhead for its binaries
2022-11-14 15:41:43 +05:30
Kovid Goyal
abaafc2d68
Parse roles in the help strings
2022-11-14 15:41:43 +05:30
Kovid Goyal
c97556de65
Show help when command is specified
2022-11-14 15:41:43 +05:30
Kovid Goyal
bbf7504303
Start work on implementing kitty @ as a static binary using Go
2022-11-14 15:41:43 +05:30
Kovid Goyal
126468a5dd
Fix #5668
2022-11-14 15:40:24 +05:30
Kovid Goyal
8280899ce1
IBUS: When creating the connection mark as focused if any top level window has focus. Fixes #5655
2022-11-10 07:21:34 +05:30
Kovid Goyal
b15c9f7419
Function should be static
2022-11-10 07:18:17 +05:30
Kovid Goyal
2e8ef66496
Another mypy update another round of spurious errors
2022-11-08 17:17:40 +05:30
Kovid Goyal
72f92b395f
version 0.26.5
2022-11-07 15:36:42 +05:30
Kovid Goyal
c9c8eb6e94
...
2022-11-05 11:37:07 +05:30
Kovid Goyal
59ded41f7a
When no OS Window is focused the active_* should return those belonging to the last focused OS Window
2022-11-05 11:34:10 +05:30
Kovid Goyal
10ad56885e
Properly specify semantics of active and focused
...
active means active in parent. focused means actually has keyboard
focus.
2022-11-04 10:38:18 +05:30
Kovid Goyal
d1eb9340ac
Output the actual tab id in its repr
2022-11-04 09:29:33 +05:30
Kovid Goyal
a856c64104
Have the semantics of is_focused match that of state:focused otherwise things are liable to get confusing
...
The previous value is now available as is_active_window
2022-11-03 22:32:07 +05:30
Kovid Goyal
c0b11c5656
Also change the behavior of state:focused matching for tabs
2022-11-03 21:56:23 +05:30
Kovid Goyal
935c4ded6b
Remote control: When matching window by state:focused match the window belonging to the OS window that was last focused
...
Fixes #5602
2022-11-03 21:52:20 +05:30
Kovid Goyal
f5de08d5fa
Splits layout: Add a new mappable action to move the active window to the screen edge
...
Fixes #5643
2022-11-03 17:40:04 +05:30
Kovid Goyal
9d471782dd
Increase some more timeouts
2022-11-03 13:13:14 +05:30
Kovid Goyal
dea3df66f8
system python appears broken on macos-latest image
...
The pip scripts install directory is not on PATH
2022-11-03 13:05:03 +05:30
Kovid Goyal
ec80a1532c
Try to make mypy happy with the contortions needed for importlib.resources
2022-11-03 12:41:13 +05:30
Kovid Goyal
93563f1280
More pointless code churn thanks to importlib.resources
...
They've deprecated a bunch of API but the suggested replacement is
either not present in versions of python that are not EOLed or doesn't
actually work.
2022-11-03 12:13:47 +05:30
Kovid Goyal
e309e54002
Allow using the re module in tab title templates
...
Fix #5639
2022-11-01 22:25:58 +05:30
Kovid Goyal
0b3228ab16
...
2022-11-01 12:51:36 +05:30
Kovid Goyal
3f28cf45d4
Dont use the deprecated importlib.resources API
2022-11-01 12:38:24 +05:30
Kovid Goyal
88b829fc9b
cleanup comment
2022-11-01 12:22:05 +05:30
Kovid Goyal
458ac6953f
Update changelog for last PR
2022-11-01 12:21:09 +05:30
Kovid Goyal
a6855bde3c
Merge branch 'no_cursor_update_if_not_changed' of https://github.com/wengxt/kitty
2022-11-01 12:20:18 +05:30
Kovid Goyal
f6f2f00c53
Add a note about wayland semantics
2022-11-01 12:13:23 +05:30
Kovid Goyal
00324c96fc
...
2022-10-31 22:09:01 +05:30
Kovid Goyal
2435a8ccfd
Wayland GNOME: Fix incorrect window size when in some circumstances when switching between windows with window decorations disabled
...
Only call wl_surface_commit() after a resize when the correct size buffer is
attached to the surface. This is ensured by setting a flag on the window that prevents
all surface commits till it is cleared. The flag is cleared at next
eglSwapBuffers().
I dont actually understand if this guarantees that the
buffer size is always correct. For example, if the back buffer is
latched when wl_egl_resize_window() is called, the backbuffer will be
correct only after two swaps (I think). Or maybe the old back buffer is
discarded, I cant find any documentation about it.
All I can say is that doing it this way seems to fix the issue.
Thanks to @jadahl for his help with tracking down the root cause.
Fixes #4802
2022-10-31 21:59:01 +05:30
Kovid Goyal
de122ed727
Bind eglQuerySurface
2022-10-31 21:03:39 +05:30
Kovid Goyal
c2ead407ae
...
2022-10-31 18:14:38 +05:30
Kovid Goyal
16b322616a
Fix cursor position at x=0 changing to x=1 on resize
...
Fixes #5635
2022-10-31 08:03:42 +05:30
Kovid Goyal
e0b4c7edc5
Increase the timeout for shell integration tests
...
Needed for ARM running non-natively
2022-10-30 11:09:36 +05:30
Kovid Goyal
a8ab4eaf23
Use a larger timeout when running prewarm test
2022-10-30 10:36:10 +05:30
Kovid Goyal
3847837bd0
Forgot to run gen-config.py
2022-10-19 20:50:53 +05:30
Kovid Goyal
1747bbbbcb
ssh kitten: Allow using absolute paths for the location of transferred data
...
Fixes #5607
2022-10-19 20:43:53 +05:30
Kovid Goyal
90e80477e9
Fix background_tint with background_opacity < 1 causing colors in cell area and padding to not match
2022-10-19 19:24:37 +05:30
Kovid Goyal
4f90110a7c
Fix ghosting when using background_tint under GNOME+Wayland
...
The problem was that on Wayland if the buffer contains pixels with alpha
< 1 they are blended with something, even if the window is opaque. Under
mutter that something was the previous frame, under sway it was the
background/whatever is under the window.
So when blending the tint color, use a blend mode that results in opaque
pixels.
Fixes #5605
2022-10-19 19:02:05 +05:30
Kovid Goyal
28c616d3ff
Move blend mode selection into draw_tint
2022-10-19 17:13:46 +05:30
Kovid Goyal
f5126c664c
Wayland KDE: Fix abort when pasting into Firefox
...
Dunno why kwin uses drag and drop callbacks when pasting into Firefox on
Wayland, but then this is Wayland, so every stupid thing imaginable will
be done at least once.
Fixes #5603
2022-10-19 11:28:33 +05:30
Kovid Goyal
0f340086c0
Fix resize_draw_strategy=static
...
The text is now static again but the background image is still scaled, cant be
bothered about that. Fixes #5601
2022-10-19 08:26:31 +05:30
Kovid Goyal
5c34f16ab4
Fix #5600
2022-10-19 07:45:27 +05:30
Kovid Goyal
03fe928e3f
show progress while compressing release tarballs
2022-10-17 09:55:09 +05:30
Kovid Goyal
55b53bcac6
version 0.26.4
2022-10-17 08:18:51 +05:30
Kovid Goyal
57d044861f
Merge branch 'patch-1' of https://github.com/ms-test/kitty
2022-10-16 13:28:06 +05:30
ms-test
aab149366d
Check correct pair member
...
I haven't looked deeply, but this looked to be a simple typo unless it's checking one both times because of a type invariant. If that is the case, and the check is correct, then a comment and/or code refactoring to call attention to this would likely be a good idea.
2022-10-16 03:53:16 -04:00
Kovid Goyal
896f93822d
kitty shell: Also output active tab id
...
Fixes #5590
2022-10-16 08:52:35 +05:30
Kovid Goyal
6890e265b6
GNOME Wayland: Fix a memory leak in gnome-shell when using client side decorations
...
destroy CSD buffers that were never attached explicitly, as mutter does
not send release events for these even when the pool is destroyed.
2022-10-12 17:52:47 +05:30
Kovid Goyal
48a4edc199
Merge branch 'action_cwd_doc' of https://github.com/cbosdo/kitty
2022-10-10 14:04:29 +05:30
