| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
| |
Bug: https://bugs.gentoo.org/868543
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
|
| |
Bug: https://bugs.gentoo.org/868543
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
|
| |
Bug: https://bugs.gentoo.org/868543
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
|
|
| |
Almost identical to short-lived 0.26.0, minor bugfix release.
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
| |
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
|
|
|
|
|
|
|
| |
Still skipping py3.11 for now. Even if make it ignore
deprecation warning failure, the prewarm tests still fail.
Now using build-launcher for tests, PATH or symlink now confuses tests
(test.py actually has a #!./kitty/launcher shebang rather than python),
and build-launcher is only a single link command after linux-package.
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
|
|
|
|
|
|
|
| |
Thought had tested this, but apparently I didn't.
glfw uses libEGL.so with wayland, so there's no sense in checking
for libGL with -X. Unfortunately X11 (or anything) being optional
at build time is our invention, so needs more sed.
Closes: https://bugs.gentoo.org/857918
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Not great but replace -O3 patch by sed, less likely to break for -9999
and isn't a very important change (and checking with qa-sed on release).
Was under the impression this old IUSE=debug did more but seems it
is only for symbols and was no-op with the patch. Now that don't
cleanup addition of e.g. -g3, the IUSE would get in the way of
users setting their own debug flags.
Live still need more updates for running tests, but will handle after
it settles down and stops changing when add next release.
Closes: https://bugs.gentoo.org/857477
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
| |
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
| |
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
| |
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
|
|
| |
No py3.11 yet, several issues.
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
|
|
|
| |
Early'ish but to get fix for bug #848822 in
Bug: https://bugs.gentoo.org/848822
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
| |
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
| |
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
| |
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
|
|
| |
This was meant to be done together with amd64 last time.
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
|
|
| |
Was brought to my attention it wasn't respected, thanks.
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
| |
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
| |
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
|
|
|
| |
--update-check-interval "shouldn't" be necessary anymore, but
keeping it in the ebuild while the option still exists
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
| |
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
|
|
|
| |
0.24.x been plagued with regressions but seems sorted out with .4
(also, work on 0.25.x has started so probably no .5 coming).
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
| |
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
|
|
|
| |
Thankfully was not affected by the history file regression this
fixes due to disabling the POSIX mode hack in Gentoo.
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
|
|
|
| |
0.24.3 has several regression fixes, so .2 is being dropped
as next stable candidate as well.
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
|
|
|
|
|
| |
Overlooked that when picked up the package despite
it was using xdg.eclass and latest kitty has a new
.desktop with MimeType.
Closes: https://bugs.gentoo.org/834362
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
no-rc option has changed meaning from "don't auto-modify user .bashrc"
to "don't use a hack'ish env-modifying method to load kitty.bash".
Method can make sense on some distros, but we have /etc/bash/bashrc.d
on Gentoo which x11-terms/kitty-shell-integration uses. It also has
less limitations by working for shells that kitty did not start as
well as remote ones if the stand-alone shell-integration package is
installed there. So, still keep no-rc.
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
|
|
| |
0.24.2 will be next stable candidate instead
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
| |
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
|
| |
Closes: https://bugs.gentoo.org/831758
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
| |
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
|
|
| |
for future bumps
Signed-off-by: Georgy Yakovlev <gyakovlev@gentoo.org>
|
|
|
|
| |
Signed-off-by: Georgy Yakovlev <gyakovlev@gentoo.org>
|
|
|
|
| |
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
|
|
|
|
| |
Accidentally reverted the --shell-integration update
when added IUSE=transfer.
Fixes: 5ab5d2999cb504df3136977702e28ef4649b4cb0
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
| |
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
|
|
|
|
|
| |
This is largely our own limitation, so this may need to be
skipped permanently. Reason it sometime worked was due to
the test chroot having 0 users and so kitty also detected
0 matching with 'who' returning nothing.
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Feel only a handful of users will want this extra dependency
used to transfer files over the terminal itself (e.g. serial
link too, albeit kitty must exist on other end), but unfortunately
upstream doesn't offer build-time options to exclude features.
Given these seds are getting heavy (needed 1 more sed + rm to
disable transfer tests), require every USE beside debug to run
tests rather than try to exclude them. Doesn't make a difference
for self-testing albeit tinderboxes may end up skipping.
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
| |
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
| |
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
| |
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
|
|
|
|
|
|
|
|
| |
This only works with imagemagick. It parse outputs of the identify
command which mismatches and also call it using "identify -- files"
while graphicsmagick (currently) doesn't accept '--'.
Also (somehwat) clarify that this is for "kitty icat image.png",
applications can display images through kitty's graphics protocol
using their own image processing dependencies.
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
| |
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
| |
Signed-off-by: Jakov Smolić <jsmolic@gentoo.org>
|
|
|
|
|
| |
Closes: https://github.com/gentoo/gentoo/pull/22724
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
|
|
|
|
|
|
|
|
| |
New dependency on librsync for transfer kitten.
Note about skipped docs:
Since ~3 months ago, a new theme is used and needs some missing
dependencies (not just sphinx) and, as long as it's only needed for
the live ebuild (prebuilt for release), does not feel worth it. Given
the full depgraph, users would likely prefer it optional+disabled too.
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* EAPI-8
* enable py3.10
* make glfw-x11.so optional with IUSE=+X wrt bug #713702
* move most glfw dlopen() deps to DEPEND-only, they are not
essential at runtime for kitty save for libglvnd
* add wayland-scanner BDEPEND (if missing it silently skips
glfw-wayland.so despite USE=wayland)
* export PKGCONFIG_EXE to use right pkg-config
* disable LTO, same reason was already stripping -O3 (if wanted,
should be set by users in CFLAGS -- there is no custom LTO paths)
* use intended --ignore-compiler-warnings to disable -Werror and
simplify -flags.patch to hopefully need less rebasing
* replace -tests.patch used to find bin/kitty by setting PATH
* drop terminfo patch, let kitty install its own /usr/lib64/kitty
terminfo and only skip /usr/share/terminfo from kitty-terminfo
* pin kitty-terminfo to match, the >-r0 check will be removable next
version and is there to avoid collisions (pinning not always
/essential/, but on a same system they originate from the same
package and should match)
* add test dep on dev-python/pillow used by 1 test
* remove unnessary fix shebang (these don't have executable bits and
are used through kitty's linked libpython)
* make libcanberra an optfeature wrt bug #775266
* for imagemagick optfeature, don't use the virtual so it's not noisy
if users have *magick installed but not the virtual
* change HOMEPAGE to actual homepage, github is shown from remote-id
Closes: https://bugs.gentoo.org/713702
Closes: https://bugs.gentoo.org/775266
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
|
|
|
| |
doc is redundant with HOMEPAGE, and bugs-to doesn't
mean much if not different from the github remote-id
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
| |
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|