19204 commits

This branch

This branch

All branches

Author	SHA1	Message	Date
Jesse Miller	073957f014	Cache HarfBuzz text shaping results ... Wire the shaper-cache to actually be used for shaped runs. On a 10s loop of the same short-word screen (~16ms/frame), main-thread CPU dropped ~28% vs master and hb_shape samples went to 0. Parse speed remains unchanged (and fast), verified with kitten __benchmark__.	2026-09-08 13:56:39 -06:00
Jesse Miller	c655a592f9	Add shaped run cache ... Per-font verstable map of exact run keys to glyph infos, positions and cell groups. Skip runs longer than 32 cells. Full-drop at 2048 entries or 1 MiB of live storage. The intention here is to avoid shaping the same runs over and over again on an active screen. This used to not have much of an impact, but now "modern" full screen TUIs are actively drawing the screens in such a way that harfbuzz shaping is showing up as a more significant portion in profiling runs. The cache is sized to not be a long-lived history but able to cache most repeated runs on a few active screens.	2026-09-08 13:56:39 -06:00
Kovid Goyal	0b12ed8c41	...	2026-09-08 18:55:51 +05:30
Kovid Goyal	df999b05e5	Fix drag gesture state leaking past the button event early returns ... Follow-ups to the drag_or_normal_select selected text drag support: - Reset the pending drag at the top of handle_button_event again, before the scrollbar, missing screen and mouse position early returns. Moving it into a helper placed it after those returns, so a scrollbar track click (with scrollbar_jump_on_click disabled) no longer cleared an armed URL drag, letting it fire on a later mouse move with no button held. Split the helper in two, as arming the drag needs an up to date mouse position but clearing it does not. - Clear the click queue for all drag kinds, not just text drags. A native drag consumes mouse-up, so the press lingered and turned the next press into a double press, selecting a word instead of starting a new gesture. Reproducible with link drags before this change. - Note that handle_potential_drag can call into Python, which can re-alloc the windows array, so w must not be used after it returns. Also share the drag thumbnail construction between drag_url and drag_selection (and skip the thumbnail rather than dividing by zero on a zero width render), use None instead of -1 for the mock event pixel position so negative positions remain testable, and tighten the off-by-one bound in clear_click_queue.	2026-09-08 18:51:02 +05:30
Kovid Goyal	d90869a8c6	Merge branch 'feature/drag-selected-text' of https://github.com/xiaobai050/kitty	2026-09-08 18:44:56 +05:30
Kovid Goyal	0d0de4d5d8	Merge branch 'update-wayland-dependencies' of https://github.com/nitin88/kitty	2026-09-08 10:46:38 +05:30
nitin88	5e00f55ac8	Bump Wayland and wayland-protocols versions ... Use Wayland 1.26.0 and wayland-protocols 1.49 with verified release checksums. The updated protocol definition fixes the zero blur capability that prevents background-effect support from being detected.	2026-09-08 10:10:12 +05:30
Kovid Goyal	024d34ec8a	Make ty 79 happy	2026-09-08 09:35:54 +05:30
Kovid Goyal	1433653e7c	Remove changelog entry as this is not a significant enough change to merit one	2026-09-08 09:02:33 +05:30
Kovid Goyal	5abdd00373	Merge branch 'increase-max-font-zoom' of https://github.com/rudolf-blue/kitty	2026-09-08 09:02:26 +05:30
rudolf-blue	18bddffcca	Raise the maximum font size and fix the legacy zoom limit ... The ceiling in change_font_size is ten times the configured font_size, so the largest size you can reach depends on the size you configured. With font_size 6 zooming stops at 60pt, while font_size 20 reaches 200pt. That is backwards: a small configured font is if anything a reason to want a larger temporary zoom, not a smaller one. Use the larger of that multiple and an absolute 256pt. No existing setup loses range, since the multiple still wins whenever it is the bigger of the two. The bounds and the clamp move into options/utils.py so the numbers are in one place and can be tested directly. increase_font_size also had its own cap of five times font_size. That was the old limit, missed when 25d8d9fe6 raised change_font_size to ten, and because the wrapper clamps before delegating, the stale bound is the one that wins. Both legacy wrappers can simply drop their bounds: they go through change_font_size, which already clamps.	2026-09-07 19:41:08 -04:00
Kovid Goyal	df00dd1cec	Fix #10448	2026-09-07 20:51:54 +05:30
Kovid Goyal	4498c00282	Merge branch 'dependabot/github_actions/actions-39eae9928b' of https://github.com/kovidgoyal/kitty	2026-09-07 09:20:54 +05:30
Kovid Goyal	3a93141fec	Merge branch 'dependabot/go_modules/all-go-deps-bfb0acd25d' of https://github.com/kovidgoyal/kitty	2026-09-07 09:20:39 +05:30
dependabot[bot]	1aac2ec474	Bump github/codeql-action from 4.37.8 to 4.37.9 in the actions group ... Bumps the actions group with 1 update: [github/codeql-action](https://github.com/github/codeql-action). Updates `github/codeql-action` from 4.37.8 to 4.37.9 - [Release notes](https://github.com/github/codeql-action/releases) - [Changelog](https://github.com/github/codeql-action/blob/main/CHANGELOG.md) - [Commits](https://github.com/github/codeql-action/compare/v4.37.8...v4.37.9) --- updated-dependencies: - dependency-name: github/codeql-action dependency-version: 4.37.9 dependency-type: direct:production update-type: version-update:semver-patch dependency-group: actions ... Signed-off-by: dependabot[bot] <support@github.com>	2026-09-07 03:45:33 +00:00
dependabot[bot]	9cbf912acc	Bump github.com/shirou/gopsutil/v4 in the all-go-deps group ... Bumps the all-go-deps group with 1 update: [github.com/shirou/gopsutil/v4](https://github.com/shirou/gopsutil). Updates `github.com/shirou/gopsutil/v4` from 4.26.7 to 4.26.8 - [Release notes](https://github.com/shirou/gopsutil/releases) - [Commits](https://github.com/shirou/gopsutil/compare/v4.26.7...v4.26.8) --- updated-dependencies: - dependency-name: github.com/shirou/gopsutil/v4 dependency-version: 4.26.8 dependency-type: direct:production update-type: version-update:semver-patch dependency-group: all-go-deps ... Signed-off-by: dependabot[bot] <support@github.com>	2026-09-07 03:42:41 +00:00
白鸽	2cd6733c75	Allow dragging selected text with drag_or_normal_select	2026-09-06 21:16:02 +08:00
Kovid Goyal	d124d4f097	Proper fix for custom shader scaling on OS window grow + shrink	2026-09-06 14:58:18 +05:30
Kovid Goyal	e55739bd10	was_maximized should use pre full screen state as well	2026-09-06 14:50:21 +05:30
Kovid Goyal	df2986feb3	Run autoformat after gen/config	2026-09-06 07:09:40 +05:30
Kovid Goyal	f90c26f1d0	Fix regression in master that broke text_fg_override_threshold	2026-09-06 06:51:31 +05:30
Kovid Goyal	cb1b153526	...	2026-09-06 06:47:26 +05:30
Kovid Goyal	f3fdc21850	Fix #10436	2026-09-04 21:04:24 +05:30
Kovid Goyal	b14ae3bf21	...	2026-09-04 06:35:44 +05:30
Kovid Goyal	0dfa8dc4fd	Fix deprecation warning when building docs	2026-09-04 06:22:51 +05:30
Kovid Goyal	49b7bf6646	Fix #10434	2026-09-04 03:58:20 +05:30
Kovid Goyal	0dbbedfb5e	Add a note about text/uri-list lifetime	2026-09-02 10:14:32 +05:30
Kovid Goyal	e2093178e5	Fix empty kitty config dir causing watching for changes to kitty.conf failing	2026-09-02 09:36:03 +05:30
Kovid Goyal	b869fe4e7f	macOS: Fix pasting paths from dropped file promises ... Previously they were always invalid (deleted in a few milliseconds). Now, kitty preserves them for ten minutes or till kitty exit, whichever is sooner. This is still not ideal, but is likely the best that can be done with File Promises since we have no idea when the client will consume them. It is at any rate better than the old behavior and is superior to the alternative of letting the drag source keep the promised data alive for some indeterminate time period. Now we can guarantee ten minutes of validity. Note I still need to test this on an actual mac as I currently dont have access to one. Fixes #10430	2026-09-02 09:35:55 +05:30
Kovid Goyal	ede47d2f93	file transfer protocol: harden against malicious clients planting symlinks ... This is actually not a security boundary for this protocol. Since it is explicitly *not* designed to be proof against malicious clients. However, to avoid more AI generated security reports against it...	2026-09-01 22:23:24 +05:30
Kovid Goyal	03fbdad855	Color control protocol: Respond to unknown field names safely using base64 encoding	2026-09-01 21:20:04 +05:30
Kovid Goyal	57bb144d96	More efficient is digit check ... Also fixes the previous check allowing trailing newlines	2026-09-01 20:59:31 +05:30
Kovid Goyal	35f88faeea	Make UBSan happy	2026-09-01 20:31:30 +05:30
Kovid Goyal	b54bfbb4f8	Drag and drop: Fix a use-after-free	2026-09-01 20:30:45 +05:30
Kovid Goyal	398fa49084	Skip test inapplicable on macOS	2026-09-01 20:28:26 +05:30
Kovid Goyal	3281919056	ssh kitten: Verify owner and permissions of SHM memory used for askpass	2026-09-01 20:14:03 +05:30
Kovid Goyal	41095672e8	Fix #10427	2026-08-31 10:36:38 +05:30
Kovid Goyal	4c784bc8e3	Merge branch 'dependabot/github_actions/actions-ebe228fd2f' of https://github.com/kovidgoyal/kitty	2026-08-31 09:17:44 +05:30
Kovid Goyal	4d92f6ad37	Merge branch 'dependabot/go_modules/all-go-deps-20a44d9da3' of https://github.com/kovidgoyal/kitty	2026-08-31 09:17:31 +05:30
dependabot[bot]	1889d266f0	Bump github/codeql-action from 4.37.7 to 4.37.8 in the actions group ... Bumps the actions group with 1 update: [github/codeql-action](https://github.com/github/codeql-action). Updates `github/codeql-action` from 4.37.7 to 4.37.8 - [Release notes](https://github.com/github/codeql-action/releases) - [Changelog](https://github.com/github/codeql-action/blob/main/CHANGELOG.md) - [Commits](https://github.com/github/codeql-action/compare/v4.37.7...v4.37.8) --- updated-dependencies: - dependency-name: github/codeql-action dependency-version: 4.37.8 dependency-type: direct:production update-type: version-update:semver-patch dependency-group: actions ... Signed-off-by: dependabot[bot] <support@github.com>	2026-08-31 03:45:52 +00:00
dependabot[bot]	b145cfadd1	Bump the all-go-deps group with 2 updates ... Bumps the all-go-deps group with 2 updates: [github.com/ALTree/bigfloat](https://github.com/ALTree/bigfloat) and [github.com/nwaples/rardecode/v2](https://github.com/nwaples/rardecode). Updates `github.com/ALTree/bigfloat` from 0.2.0 to 0.3.0 - [Commits](https://github.com/ALTree/bigfloat/compare/v0.2.0...v0.3.0) Updates `github.com/nwaples/rardecode/v2` from 2.3.0 to 2.4.1 - [Commits](https://github.com/nwaples/rardecode/compare/v2.3.0...v2.4.1) --- updated-dependencies: - dependency-name: github.com/ALTree/bigfloat dependency-version: 0.3.0 dependency-type: direct:production update-type: version-update:semver-minor dependency-group: all-go-deps - dependency-name: github.com/nwaples/rardecode/v2 dependency-version: 2.4.1 dependency-type: direct:production update-type: version-update:semver-minor dependency-group: all-go-deps ... Signed-off-by: dependabot[bot] <support@github.com>	2026-08-31 03:42:58 +00:00
Kovid Goyal	d5e912d999	Merge branch 'empty-envvar' of https://github.com/MithicSpirit/kitty	2026-08-31 06:14:17 +05:30
MithicSpirit	04930efa44	Treat empty display server envvars as unset ... This makes the DISPLAY, WAYLAND_DISPLAY, WAYLAND_SOCKET, and KITTY_DISABLE_WAYLAND envvars be treated as unset if they are the empty string. Therefore, users can simply do "WAYLAND_DISPLAY= kitty" as opposed to the full incantation "env -u WAYLAND_DISPLAY kitty" (in this case setting KITTY_DISABLE_WAYLAND would also work, but that wouldn't propagate to any clients launched through kitty). This matches the behavior of many other programs.	2026-08-30 15:54:55 -04:00
Kovid Goyal	1de47c6cf4	...	2026-08-30 21:12:41 +05:30
Kovid Goyal	8dd6436917	Fix #10423	2026-08-30 21:12:13 +05:30
Kovid Goyal	33a2685550	autoformat	2026-08-30 08:28:03 +05:30
Kovid Goyal	fa2e694c38	explicitly reject POSIX SHM names that dont start with / to match spec requirement	2026-08-30 08:25:57 +05:30
Kovid Goyal	049574d489	Update changelog	2026-08-30 08:19:14 +05:30
Kovid Goyal	8e58d32dae	Fix rounded window borders leaving corner regions unpainted ... When not rendering in layers (opaque windows) kitty never clears the framebuffer and relies on full pixel coverage. The rounded border pass left the corner regions between the rounded outline and its bounding rect unwritten, so they accumulated stale framebuffer content, and in the layered path (e.g. during live resize) they blitted as black on opaque windows. Add an outer_color uniform to the rounded rect shader that fills the area outside the outline, and paint it with the default background at the same opacity draw_borders() uses for non-border rects, so corners match the surrounding gaps in all rendering paths. Also skip the rounded border pass entirely when no rounded rects exist, instead of binding GL state and iterating all border rects every frame with the default window_border_radius of 0.	2026-08-30 08:16:42 +05:30
Kovid Goyal	3d600a5ef2	Merge pull request #10421 from NightWatcher314/master ... Add configurable rounded window borders	2026-08-30 08:02:52 +05:30