This PR bumps the version of the GLSL extension to v0.2.1.

Release Notes:

- N/A

Co-authored-by: zed-zippy[bot] <234243425+zed-zippy[bot]@users.noreply.github.com>

Closes #51297

On non-QWERTY layouts, all Cmd+key events are routed through the macOS
IME because `key_char` is always `None` when Cmd is held. For certain
characters (dead keys like backtick, and non-dead keys like ç), the IME
calls `insertText:` instead of `doCommandBySelector:`, consuming the
event before it reaches GPUI's keybinding system or macOS system
shortcuts.

This adds `!platform` to the IME-path condition in `handle_key_event` so
Cmd+key events bypass the IME (except when composing). GPUI handles them
if a binding matches, otherwise `performKeyEquivalent:` returns `NO` and
macOS handles them.

**This won't fully fix Cmd+backtick window cycling by itself** because
Zed's key_equivalents system maps default keybindings onto the physical
backtick key on various layouts. For example, `cmd-'`
(ToggleSelectedDiffHunks) maps to the backtick key on Spanish, `cmd-=`
(IncreaseBufferFontSize) on Scandinavian layouts, `cmd-"`
(ExpandAllDiffHunks) on German/Portuguese/Swiss, and `cmd-}`
(ActivateNextItem) on Spanish-ISO. These Zed bindings shadow the
backtick key and consume the event before macOS can cycle windows. I'd
appreciate guidance on the preferred approach to resolve these
keybinding conflicts -- IMO, Zed's default shortcuts should not be
interfering with Cmd+backtick for any layout.

Release Notes:

- Fixed Cmd+key shortcuts being consumed by the IME on non-QWERTY
keyboard layouts, preventing Zed keybindings and macOS system shortcuts
from working with special characters.

Release Notes:

- N/A

Release Notes:

- N/A

Add new page detailing the roles available for users in their
organizations.

Release Notes:

- N/A *or* Added/Fixed/Improved ...

The GLSL language extension was missing the "task" and "mesh" path
suffixes for task and mesh shaders.
"task" and "mesh" are the official suffixes used in glslang.

Release Notes:

- N/A

Co-authored-by: MrSubidubi <finn@zed.dev>

Sigh.. The missing flag caused the wrong output to be used, resulting in
an error in the process.

Release Notes:

- N/A

This will allow us to also use the workflows for this repository, which
will especially come in handy once we revisit provider extensions.

Not perfect, as we will trigger some failed workflows for extensions
that were just added

Release Notes:

- N/A

Previously, if you wanted to have a button that contains icons on both
edges, you'd need to use a `ButtonLike` component, which takes any
children. Meanwhile, the `Button` would only take one icon, where you
could control its position through the `IconPosition` enum. This has
always felt unnecessarily limiting. So, this PR removes this limitation
by adding two new methods to the button: `start_icon` and `end_icon`.

In the meantime, I have also been bothered by the unnecessary
indirection in the `IconButton` due to the existence of the `ButtonIcon`
component. So I figured I could also completely eliminate that by adding
some of its methods directly to the `IconButton` and in the Button, just
using a regular `Icon` component.

---
## Before
```rust
Button::new("id", "Label")
    .icon(IconName::Plus)
    .icon_position(IconPosition::Start)
    .icon_size(IconSize::Small)
    .icon_color(Color::Muted)
```

## After
```rust
Button::new("id", "Label")
    .start_icon(Icon::new(IconName::Check))
    .end_icon(Icon::new(IconName::ChevronDown).size(IconSize::XSmall))
```

This should have no visual impact to the UI.

Release Notes:

- N/A

Follow up to https://github.com/zed-industries/zed/pull/51487

The PR above added the item to the menu, and this one makes the menu
item actually insert a mention crease with the branch diff. That was
missing in the previous one.

Release Notes:

- N/A

Release Notes:

- N/A *or* Added/Fixed/Improved ...

Follow-up of https://github.com/zed-industries/zed/pull/51299

Release Notes:

- N/A

When the connection to X is broken zed will go into an infinite loop and
eat up 100% (of one core) of CPU; this change causes it to exit with an
error instead.

I encountered this behavior while running zed in
[Xephyr](https://freedesktop.org/wiki/Software/Xephyr/) for testing,
though I do sometimes terminate my X server as a way to log out or
attempt to recover from a (very) broken state, and I appreciate a
graceful exit in those situations!

Exiting in case of X server disconnect is common practice in my
observations, likely as the difficulty of recreating state stored
server-side outweighs the potential utility in attempting to recover (if
"reconnecting" to an X server is ever desired in regular usage,
[Xpra](https://xpra.org/index.html) might be able to help!).


Release Notes:

- N/A

This PR introduces the `project_panel.scrollbar.horizontal_scroll`
setting to allow users to toggle the horizontal scroll bar in the
project panel. This was Zed's design before PR #18513, and the default
behavior of VSCode (`workbench.list.horizontalScrolling`).


https://github.com/user-attachments/assets/f633f4e4-a585-4494-8f48-df77c6aca418

## Rationale

Zed's design used to be the same as the default behavior of VSCode.
I.e., no horizontal scrolling, and the view is always snapped to the
left, with long file names clipped of. If you want to see the content
that is out-of-frame, you'll need to drag the handle and expand the
project panel. This could be problematic, especially for large repos
with multiple levels of nested directories, as pointed out by issues
#5550 and #7001.

<img width="1398" height="992" alt="image"
src="https://github.com/user-attachments/assets/d86563f2-0f06-4e9e-818c-155ac45f0f56"
/>\
*VSCode's default setup, for reference.*

Then came PR #18513, which added horizontal scroll and addressed this
pain point, but users didn't have a choice. They're stuck with
horizontal scrolling always turned on. I, for instance, personally
prefer the old, VSCode-default behavior, for most projects I open are
small and don't need horizontal scrolling in the project panel. With
horizontal scrolling always turned on, I find it annoying to have my
project panel view accidentally scrolled to the middle, and I'll have to
grab my mouse and scroll it back. It's also visually redundant.

Thus, why not add an option like VSCode's
`workbench.list.horizontalScrolling` and let users choose? I'd love to
be able to, say, set a per-project override for the projects that need
horizontal scrolling, while having it disabled by default.

## Extra Notes

- I was originally thinking about using `ScrollbarAxes` from
`src/editor_settings.rs` and make the option
`project_panel.scrollbar.axes.horizontal` similar to the global editor
scrollbar settings, but this option is specific to the project panel and
it doesn't quite make sense to allow disabling vertical scrolling on the
project panel, so I added a standalone option for it instead, similar to
VSCode's `workbench.list.horizontalScrolling`.

- I went the conservative route and set horizontal scrolling to enabled
(current behavior) by default. Imo it might make more sense to disable
it by default instead, similar to VSCode, but I'll leave this for the
Zed team to decide.

- I named it `horizontal_scroll` instead of `horizontal_scrolling` to be
consistent with the adjacent setting `sticky_scroll`.

- As for tests, I don't see tests for the scrollbar, so I didn't add
any.

I'd be glad to update the PR if anything is not inline with the
project's requirements or conventions.

---

Release Notes:

- Added `project_panel.scrollbar.horizontal_scroll` setting to allow
toggling horizontal scrolling in the project panel

Signed-off-by: k4yt3x <i@k4yt3x.com>

Release Notes:

- N/A

When on a untrusted project, if one was to try and use the commit
generation functionality, the command would fail because of the `-c
diff.external` configuration provided in `GitBinary::build_command`, as
git would interpret this as `""` and try to run that command.

This `-c diff.external` is a good safeguard to have on untrusted
repositories because it prevents random commands, configured in
`.git/config` from being run. For example, if one uses `git config
diff.external "touch bananas.txt"` and then run `git diff`, a new
`bananas.txt` file would be created.

However, it was still possible to bypass this safeguard using the
following strategy:

1. Specify a custom diff for a specific file format. For example, for
markdown files, with `printf '*.md diff=pwned\n' > .gitattributes`
2. Update the command run by the `pwned` diff, for example, `git config
diff.pwned.command "touch bananas.txt"`
3. Open Zed and attempt to generate a commit message in an untrusted
repository and check that a new `bananas.txt` file was created

This is only prevented by using the `--no-ext-diff` flag on the `diff`
command, so a new `GitBinary::build_diff_command` has been introduced
which simply wraps `GitBinary::build_command` and adds the
`--no-ext-diff` flag, if necessary.

As a side-effect, this also makes it so that generating a commit message
in an untrusted repository works again, which was accidentally broken on
https://github.com/zed-industries/zed/pull/50649 .

Before you mark this PR as ready for review, make sure that you have:
- [X] Added a solid test coverage and/or screenshots from doing manual
testing
- [X] Done a self-review taking into account security and performance
aspects
- [X] Aligned any UI changes with the [UI
checklist](https://github.com/zed-industries/zed/blob/main/CONTRIBUTING.md#uiux-checklist)

Release Notes:

- Fixed commit message generation in untrusted repositories

This PR adds the recently introduced "branch diff" mention option to the
"Add Context" menu in the message editor:

<img width="450" height="582" alt="Screenshot 2026-03-13 at 9  58@2x"
src="https://github.com/user-attachments/assets/1f5102e7-9018-4a5b-992d-43befa55dcbe"
/>

Release Notes:

- N/A

Release Notes:

- Added an error indicator in the edit prediction menu with an error
message when the free tier limit is exceeded

This wil help with releases for extensions living this repository, which
will become more relevant once agent provider extensions are back on the
table.

Release Notes:

- N/A

Add documentation for snippets in extensions.

Feel free to change the wording or add more content. 

Before you mark this PR as ready for review, make sure that you have:
- [ ] Added a solid test coverage and/or screenshots from doing manual
testing
- [ ] Done a self-review taking into account security and performance
aspects
- [ ] Aligned any UI changes with the [UI
checklist](https://github.com/zed-industries/zed/blob/main/CONTRIBUTING.md#uiux-checklist)

Release Notes:

- N/A

---------

Co-authored-by: Finn Evers <finn.evers@outlook.de>

Release Notes:

- agent: Fixed an issue where external agents would return an error
because unsupported ACP method was called

Release Notes:

- N/A

---------

Co-authored-by: cameron <cameron.studdstreet@gmail.com>
Co-authored-by: Bennet Bo Fenner <bennetbo@gmx.de>

Release Notes:

- N/A

Reverts zed-industries/zed#50692

The test here doesn't pass, unsure how this managed to even get merged

Remove the BreadcrumbText struct from workspace and use the existing
HighlightedText struct from the language crate instead. The per-segment
font field is replaced by returning an optional Font alongside the
segments from the breadcrumbs() method, since the font was always
uniform across all segments of a given item.

Release Notes:

- N/A *or* Added/Fixed/Improved ...

Release Notes:

- N/A *or* Added/Fixed/Improved ...

Release Notes:

- N/A *or* Added/Fixed/Improved ...

Closes #51450 

When you restart the lsp, it does not sufficiently clear cached data
regarding the semantic tokens, if using semantic_tokens = "full", this
would mean that you would have no syntax highlighting. also, toggling on
and off semantic tokens in the menu would have no effect. this change
properly clears the cached state and things work again!

Before:


https://github.com/user-attachments/assets/67ac1be1-ae3d-4c84-afbc-056fd81f63f0

After:


https://github.com/user-attachments/assets/644f8297-8003-4d74-b962-81ba9bb8274c

You might notice that the syntax highlighting is quite spare in the
videos, especially compared to the non semantic token based
highlighting, and you would be correct! but thats just how it is with
`semantic_tokens: "full"`, other editors, like neovim, provide basic
syntax highlighting that zed doesn't (because it doesn't need to with
treesitter usually, but here treesitter is disabled), however if we turn
off that syntax highlighting we can see that neovim actually matches zed
here:

<img width="847" height="485" alt="Screenshot 2026-03-12 at 11 33 19 PM"
src="https://github.com/user-attachments/assets/7f90789c-dac3-41bf-9d19-640c6c7b1144"
/>


Before you mark this PR as ready for review, make sure that you have:
- [x] Added a solid test coverage and/or screenshots from doing manual
testing
- [x] Done a self-review taking into account security and performance
aspects
- [x] Aligned any UI changes with the [UI
checklist](https://github.com/zed-industries/zed/blob/main/CONTRIBUTING.md#uiux-checklist)

Release Notes:

- lsp: Fixed restarting the LSP breaking semantic token highlighting.

Closes #49603

Release Notes:
- Added a 7-day grace period to prevent recently used local workspaces
from being deleted when their paths are temporarily unavailable. Session
workspaces are always preserved on restart.

Updates #21054

Authored-By: @ngauder

Release Notes:

- Added _ and newline to subword boundaries

---------

Co-authored-by: Nikolas Gauder <nikolas.gauder@tum.de>

Closes #50436

Before you mark this PR as ready for review, make sure that you have:
- [x] Added a solid test coverage and/or screenshots from doing manual
testing
- [x] Done a self-review taking into account security and performance
aspects
- [ ] Aligned any UI changes with the [UI
checklist](https://github.com/zed-industries/zed/blob/main/CONTRIBUTING.md#uiux-checklist)

Release Notes:

- Fixed XF86 multimedia key names ("New", "Save", "Open") being shown as
keybinding hints in menus instead of the actual keyboard shortcuts.

Closes #ISSUE

Before you mark this PR as ready for review, make sure that you have:
- [ ] Added a solid test coverage and/or screenshots from doing manual
testing
- [ ] Done a self-review taking into account security and performance
aspects
- [ ] Aligned any UI changes with the [UI
checklist](https://github.com/zed-industries/zed/blob/main/CONTRIBUTING.md#uiux-checklist)

Release Notes:

- N/A

Closes #51417

I noticed this bug in the settings menu where when I opened the settings
menu, I could not scroll down through all the available options, eg. on
the initial page I wasn't able to scroll down to privacy. When I saw
that no one else had reported this issue, I figured it may be due to my
setup, and it turns out that using Aerospace, the window manager I use,
was what made this bug visible to me. Because aerospace resizes the
window right after it launches, the originally computed heights for the
list are incorrect, meaning the scroll bar is the wrong size as well.

in the relevant code there was a comment that says "If the width of the
list has changed, invalidate all cached item heights" which wasn't
incorrect per-se, but it just invalidated them without triggering any
re-computation, causing incorrect scroll bars.

My intuition is that window resizes/events that change the width of the
list bounds are fairly rare, so there shouldn't be a large performance
hit from the change.

Also implemented a test that directly showcases the behavior, if you run
the test without the change it fails, as the max_offset_for_scrollbar
will be wrong.

Videos:

Before 


https://github.com/user-attachments/assets/2b680222-7071-4098-863f-519361f0756a

After:


https://github.com/user-attachments/assets/1222a299-23d7-4007-8e88-55d2daccce64


[x] Tests
[x] Video of behavior


Release Notes:

- gpui: fixed list height re-computation when the list width changes.

Just cleaning up our component set a bit. This one wasn't used at all.

Release Notes:

- N/A

Closes #51416

Before you mark this PR as ready for review, make sure that you have:

- [ ] Added a solid test coverage and/or screenshots from doing manual
testing
- [ ] Done a self-review taking into account security and performance
aspects
- [ ] Aligned any UI changes with the [UI
checklist](https://github.com/zed-industries/zed/blob/main/CONTRIBUTING.md#uiux-checklist)

Release Notes:

- N/A

This PR assigns meaningful names to some single-letter bindings we were
using to refer to the organization.

Release Notes:

- N/A

Release Notes:

- N/A *or* Added/Fixed/Improved ...

Closes #49854 

Before you mark this PR as ready for review, make sure that you have:
- [x] Added a solid test coverage and/or screenshots from doing manual
testing
- [x] Done a self-review taking into account security and performance
aspects
- [x] No UI changes

Video for the fix: 
in the video i open a project -> close the project -> quit out of zed ->
reopen zed -> zed opens to an empty workspace which was not the case
before


https://github.com/user-attachments/assets/1afb44a1-932b-4dab-8228-9d9d65750b6e



Release Notes:
 - Fixed closed projects re-opening erroneously

Before you mark this PR as ready for review, make sure that you have:
- [x] Added a solid test coverage and/or screenshots from doing manual
testing
- [x] Done a self-review taking into account security and performance
aspects
- [x] Aligned any UI changes with the [UI
checklist](https://github.com/zed-industries/zed/blob/main/CONTRIBUTING.md#uiux-checklist)

Release Notes:

- N/A

It does not make sense to enable them during the running turn and it can
lead to more confusing states if subagents are used.

Release Notes:

- N/A

---------

Co-authored-by: Danilo Leal <daniloleal09@gmail.com>

- **Fix race condition in test_collaborating_with_completion**
- **WIP: Integrate scheduler crate into GPUI TestDispatcher**
- **WIP: scheduler integration debugging**
- **Fix formatting**
- **Unify RunnableMeta and add execution tracking to TestScheduler**
- **Remove unused execution tracking from TestScheduler and
TestDispatcher**
- **Add is_ready() to GPUI Task for API parity with scheduler**
- **Eliminate RunnableVariant::Compat - all runnables now have source
location metadata**
- **Update integration plans to reflect completed phases**
- **Simplify RunnableVariant to type alias**
- **Delegate TestDispatcher task queues to TestScheduler (Phase 2b)**
- **Remove waiting_hint/waiting_backtrace and debug logging from
TestDispatcher**
- **Remove wrapper methods from TestDispatcher - access scheduler()
directly**
- **Update integration plan with complete state and instructions for
full scheduler migration**
- **Use scheduler's native timer() and simplify TestDispatcher**
- **Fix rng() usage to lock mutex, update plan with SharedRng wrapper**
- **Add SharedRng wrapper for ergonomic random number generation**
- **Update plan: mark Phase 1 (SharedRng) as complete**
- **Update scheduler integration plan with Phase 2 investigation notes**
- **Phase 3: Delegate simulate_random_delay to
scheduler.yield_random()**
- **Phase 4: Remove TaskLabel**
- **Phase 5 (WIP): Simplify block_internal and remove unparkers**
- **Phase 5 Complete: Scheduler integration finished**
- **Update integration plan with code review findings**
- **Phase 6 & 7: Restore realtime priority support and delete dead
code**
- **Add TestApp and TestAppWindow for cleaner GPUI testing**
- **Fix formatting across the branch**
- **Fix Linux build: add explicit type annotation and rename
probability() to weight()**
- **Add TestApp and TestAppWindow for cleaner GPUI testing**
- **Rename TestAppWindow to TestWindow, internal TestWindow to
TestPlatformWindow**
- **Remove unused RunnableVariant imports on Linux**
- **Add STATUS.md for next agent**
- **Run cargo fmt**
- **Use per-app element arena only and scope test draws**
- **Fix collab tests for scheduler timing and ordering**
- **Store element arena on App and route element allocations through
draw scope**
- **Fix TestScheduler lock ordering between rng and state**
- **Fix inlay hints test by explicitly triggering refresh after viewport
setup**
- **Add scheduler integration regression risk analysis doc**
- **Fix tests: avoid caching Entity in global OnceLock for Codestral API
key**
- **Document learned weak point: global cached Entity handles break
across App contexts**
- **Add scheduler regression test for block_with_timeout continuation
and explicit time advancement**
- **Document TestScheduler timeout tick budget behavior and explicit
time advancement guidance**
- **Add test asserting realtime priority spawns panic under
TestDispatcher**
- **Document realtime priority determinism contract in tests**
- **Remove realtime priority until we have a concrete use case (cc
@localcc)**
- **Update STATUS for scheduler integration decisions and realtime
priority removal**
- **Fix prettier docs and clippy in scheduler tests**
- **Remove unused imports from Windows dispatcher**
- **WIP: scheduler integration debugging + agent terminal diagnostics**
- **Update scheduler integration status**
- **Remove temporary planning docs, consolidate into scheduler
integration doc**
- **Remove unrelated changes from scheduler integration**
- **Fix clippy errors**
- **Add STATUS.md with debugging instructions for Linux/Windows hang**
- **WIP: local changes needed by ex**
- **Add pointer capture API for stable drag handling**
- **Add pointer capture API for stable drag handling**
- **chore: update generated cargo manifests**
- **gpui: Expose ShapedLine::width() for pen advancement**
- **Remove git2 usage from util test.rs**
- **Store DiagnosticQuad bounds in logical Pixels**
- **WIP: executor and test_app changes for scheduler integration**
- **Expose font APIs publicly**
- **gpui: add typed diagnostics and record_diagnostic API**
- **WIP: gpui test window diagnostics changes**
- **Add LineCacheKey trait and shape_line_cached API for
content-addressable shaping**
- **Fix RenderGlyphParams field additions for Ex compatibility**
- **Add doc comment for recommended_rendering_mode, fix formatting**
- **Add scheduler_executor() method for Ex compatibility**
- **Fix TestWindow -> TestPlatformWindow in test_context.rs**
- **Add headless metal renderer and window focus improvements**
- **Fix double borrow in TestWindow::simulate_resize**
- **Fix cbindgen panic: remove default type parameter from
Diagnostic<T>**
- **Implement AppContext for HeadlessMetalAppContext**
- **Missing trait impls**
- **Add ShapedLine::split_at and eliminate re-shaping in soft wraps**
- **Add handoff doc for platform-neutral-tests merge**
- **Remove ex-only test infrastructure before merging main**
- **Add cross-platform HeadlessAppContext with pluggable text system**
- **Export platform_text_system() from gpui_windows for cross-platform
tests**
- **Restore TestApp/TestAppWindow with pluggable text system support**
- **Add TestApp::open_window_sized for tests that need specific window
dimensions**
- **Fix some warnings**
- **Fixes**
- **Add a platform-neutral headless renderer interface**
- **Synchronize Managed texture before CPU readback on discrete GPUs**
- **Allow creating TestDispatcher with custom scheduler**

Release Notes:

- N/A

---------

Co-authored-by: Nathan Sobo <nathan@zed.dev>
Co-authored-by: John Tur <john-tur@outlook.com>
Co-authored-by: Agus Zubiaga <agus@zed.dev>
Co-authored-by: Antonio Scandurra <me@as-cii.com>

Small tweaks to our feature doc and a link out to more about how the
Feature Request process works.

Release Notes:

- N/A

Closes #ISSUE

Before you mark this PR as ready for review, make sure that you have:
- [ ] Added a solid test coverage and/or screenshots from doing manual
testing
- [ ] Done a self-review taking into account security and performance
aspects
- [ ] Aligned any UI changes with the [UI
checklist](https://github.com/zed-industries/zed/blob/main/CONTRIBUTING.md#uiux-checklist)

Release Notes:

- N/A *or* Added/Fixed/Improved ...

Adds a loading state to the archive view and a couple of other tiny UI
tweaks to the thread item and such.

Release Notes:

- N/A

Mentioned in #47258 

Release Notes:

- Added hotkey options and actions for toggling light and dark theme.
- Add default keymap as `cmd/ctrl+k cmd/ctrl+shift+t`

Fixes: #48945

Description:

The convert:to-title-case, convert:to-upper-camel-case, and
convert:to-lower-camel-case editor commands were stripping leading
whitespace from each line of a multi-line selection.

Root cause: The conversion functions split on whitespace using
.split_whitespace() and then joined the resulting words, discarding any
leading spaces/tabs before the first word on each line.

Fix: Each line now preserves its leading whitespace by capturing and
re-prepending it before applying the case conversion.

Tests: Added test cases covering multi-line selections with indentation
for all three commands.

Video : 


[bug1fix.webm](https://github.com/user-attachments/assets/f4d25c55-bc6d-44e6-a989-7d9b4bc59ac9)

Release Notes:

- Fixed trailing whitespace handling on text case changes

Release Notes:

- Tweaked the git diff status to render + and - using the font instead
of icons.

In Helix, selecting a line with `x` creates a selection from column 0 of
the current row to column 0 of the next row. The default
`InsertEndOfLine` uses the selection head (which is on the next row) to
find the line end, placing the cursor on the wrong line. 

This commit introduces a new `HelixInsertEndOfLine`, mapped by default
to `shift-a` when Helix mode is enabled, that moves left from the head
first to land on the correct line.

Release Notes:

- Fixed `shift-a` in Helix select mode placing the cursor on the wrong
line after selecting with `x`

---------

Co-authored-by: SkandaBhat <9384046+SkandaBhat@users.noreply.github.com>
Co-authored-by: dino <dinojoaocosta@gmail.com>