Commit log

c8ada5b agent_ui: Reduce label repetitiveness on new thread menu (#42001)

Click to expand commit body 
Mostly just removing "thread" from all external agent menu items; I
think we can do without it and it already becomes much better/cleaner.

Release Notes:

- N/A

Danilo Leal created

27a1884 open_ai: Make the deltas optional (#39142)

Click to expand commit body 
I am using an Azure OpenAI instance since that is what is provided at
work and with how they have it setup not all responses contain a delta,
which lead to errors and truncated responses. This is related to how
they are filtering potentially offensive requests and responses. I don't
believe this filter was made in-house, instead I believe it is provided
by Microsoft/Azure, so I suspect this fix may help other users.

Release Notes:

- N/A

Co-authored-by: Bennet Bo Fenner <bennetbo@gmx.de>

Techy and Bennet Bo Fenner created

2bc1d60 remote: Fix open terminal fails when $SHELL is not set (#41990)

Click to expand commit body 
Closes #41644

Release Notes:

- Fixed issue where it failed to spawn terminal on systems such as
Alpine.

Smit Barmase created

17933f1 Update documentation on Java support (#41758)

Click to expand commit body 
This brings the documentation on Java in line with the much changed
reality of the Java extension.
Note that the correctness of this is contingent on
https://github.com/zed-industries/extensions/pull/3745 being merged.

Release Notes:

- N/A

Karl-Erik Enkelmann created

cd87307 language: Fix language detection for injected syntax layers (#41111)

Click to expand commit body 
Closes #40632

**TL;DR:** The `wrap selections in tag` action was unavailable in ERB
files, even when the cursor was positioned in HTML content (outside of
Ruby code blocks). This happened because `syntax_layer_at()` incorrectly
returned the Ruby language for positions that were actually in HTML.
**NOTE:** I am not familiar with that part of Zed so it could be that
the fix here is completely incorrect.

Previously, `syntax_layer_at` incorrectly reported injected languages
(e.g., Ruby in ERB files) even when the cursor was in the base language
content (HTML). This broke actions like `wrap selections in tag` that
depend on language-specific configuration.

The issue had two parts:
1. Missing start boundary check: The filter only checked if a layer's
end was after the cursor (`end_byte() > offset`), not if it started
before, causing layers outside the cursor position to be included. See
the `BEFORE` video: when I click on the HTML part it reports `Ruby`
language instead of `HTML`.
2. Wrong boundary reference for injections: For injected layers with
`included_sub_ranges` (like Ruby code blocks in ERB), checking the root
node boundaries returned the entire file range instead of the actual
injection ranges.

This fix:
- Adds the containment check using half-open range semantics [start,
end) for root node boundaries. That ensures proper reporting of the
detected language when a cursor (`|`) is located right after the
injection:

   ```
   <body>
    <%= yield %>|
  </body>
   ```

- Checks `included_sub_ranges` for injected layers to determine if the
cursor is actually within an injection
- Falls back to root node boundaries for base layers without sub-ranges.
This is the original behavior.

Fixes ERB language support where actions should be available based on
the cursor's actual language context. I think that also applies to some
other template languages like HEEX (Phoenix) and `*.pug`. On short
videos below you can see how I navigate through the ERB template and the
terminal on the right outputs the detected language if you apply the
following patch:

```diff
diff --git i/crates/editor/src/editor.rs w/crates/editor/src/editor.rs
index 15af61f5d2..54a8e0ae37 100644
--- i/crates/editor/src/editor.rs
+++ w/crates/editor/src/editor.rs
@@ -10671,6 +10671,7 @@ impl Editor {
         for selection in self.selections.disjoint_anchors_arc().iter() {
             if snapshot
                 .language_at(selection.start)
+                .inspect(|language| println!("Detected language: {:?}", language))
                 .and_then(|lang| lang.config().wrap_characters.as_ref())
                 .is_some()
             {
```

**Before:**


https://github.com/user-attachments/assets/3f8358f4-d343-462e-b6b1-3f1f2e8c533d


**After:**



https://github.com/user-attachments/assets/c1b9f065-1b44-45a2-8a24-76b7d812130d



Here is the ERB template:

```
<!DOCTYPE html>
<html>
  <head>
    <meta http-equiv="Content-Type" content="text/html; charset=utf-8">
    <style>
      /* Email styles need to be inline */
    </style>
  </head>
  <body>
    <%= yield %>
  </body>
</html>
```

Release Notes:

- N/A

Vitaly Slobodin created

11b29d6 ruby: Add note about enabling Ruby LSP for ERB files (#41851)

Click to expand commit body 
Hi, this is a follow-up change for
https://github.com/zed-industries/zed/pull/41754 I think it important to
keep existing things working. So add notes to the Ruby extension doc
about enabling Ruby LSP for ERB files as well. Thanks!

Release Notes:

- N/A

Vitaly Slobodin created

c061698 project: Fetch latest lsp data in `deduplicate_range_based_lsp_requests` (#41971)

Click to expand commit body 
Fixes ZED-2MK

Release Notes:

- Fixed a panic in inlay hints

Lukas Wirth created

b4f7af0 Work around codegen bug with GetKeyboardState (#41970)

Click to expand commit body 
Works around an issue, which can be reproduced in the following program:
```rs
use windows::Win32::UI::Input::KeyboardAndMouse::{GetKeyboardState, VK_CONTROL};

fn main() {
    let mut keyboard_state = [0u8; 256];
    unsafe {
        GetKeyboardState(&mut keyboard_state).unwrap();
    }

    let ctrl_down = (keyboard_state[VK_CONTROL.0 as usize] & 0x80) != 0;
    println!("Is Ctrl down: {ctrl_down}");
}
```

In debug mode, this program prints the correct answer. In release mode,
it always prints false. The optimizer appears to think that
`keyboard_state` isn't mutated and remains zeroed, and folds the
`modifier_down` comparisons to `false`.

Release Notes:

- N/A

John Tur created

c83621f editor: Fix setting multi_cursor_modifier opens implementation in new pane instead of new tab (#41963)

Click to expand commit body 
Closes #41014

Release Notes:

- Fixed an issue where `multi_cursor_modifier` set to `cmd_or_ctrl`
opens implementation in new pane instead of new tab.

Smit Barmase created

0da52d6 Add ACP terminal-login via _meta field (#41954)

Click to expand commit body 
As discussed with @benbrandt and @mikayla-maki:

* We now tell ACP clients we support the nonstandard `terminal-auth`
`_meta` field for terminal-based authentication
* In the future, we anticipate ACP itself supporting *some* form of
terminal-based authentication, but that hasn't been designed yet or gone
through the RFD process
* For now, this unblocks terminal-based auth

Release Notes:

- Added experimental terminal-based authentication to ACP support

Richard Feldman created

60ee0dd Use our `node` runtime for ACP extensions (#41955)

Click to expand commit body 
Release Notes:

- Now ACP extensions use Zed's managed Node.js runtime

---------

Co-authored-by: Mikayla Maki <mikayla.c.maki@gmail.com>

Richard Feldman and Mikayla Maki created

9fc4abd Re-enable preview auto-release (#41952)

Click to expand commit body 
Patch releases to preview will now automatically release

Release Notes:

- N/A

Conrad Irwin created

2ead8c4 Improve Windows text input for international keyboard layouts and IMEs (#41259)

Click to expand commit body 
- Custom handling of dead keys has been removed. UX for dead keys is now
the same as other applications on Windows.
- We could bring back some kind of custom UI, but only if UX is fully
compatible with expected Windows behavior (e.g. ability to move the
cursor after typing a dead key).
  - Fixes https://github.com/zed-industries/zed/issues/38838
- Character input via AltGr shift state now always has priority over
keybindings. This applies regardless of whether the keystroke used the
AltGr key or Ctrl+Alt to enter the shift state.
- In particular, we use the following heuristic to determine whether a
keystroke should trigger character input first or trigger keybindings
first:
- If the keystroke does not have any of Ctrl/Alt/Win down, trigger
keybindings first.
- Otherwise, determine the character that would be entered by the
keystroke. If it is a control character, or no character at all, trigger
keybindings first.
- Otherwise, the keystroke has _any_ of Ctrl/Alt/Win down and generates
a printable character. Compare this character against the character that
would be generated if the keystroke had _none_ of Ctrl/Alt/Win down:
- If the character is the same, the modifiers are not significant;
trigger keybindings first.
- If there is no active input handler, or the active input handler
indicates that it isn't accepting text input (e.g. when an operator is
pending in Vim mode), character entry is not useful; trigger keybindings
first.
- Otherwise, assume the modifiers enable access to an otherwise
difficult-to-enter key; trigger character entry first.
  - Fixes https://github.com/zed-industries/zed/issues/35862
- Fixes
https://github.com/zed-industries/zed/issues/40054#issuecomment-3447833349
  - Fixes https://github.com/zed-industries/zed/issues/41486
- TranslateMessage calls are no longer skipped for unhandled keystrokes.
This fixes language input keys on Japanese and Korean keyboards (and
surely other cases as well).
- To avoid any other missing-TranslateMessage headaches in the future,
the message loop has been rewritten in a "traditional" Win32 style,
where accelerators are handled in the message loop and TranslateMessage
is called in the intended manner.
  - Fixes https://github.com/zed-industries/zed/issues/39971
  - Fixes https://github.com/zed-industries/zed/issues/40300
  - Fixes https://github.com/zed-industries/zed/issues/40321
  - Fixes https://github.com/zed-industries/zed/issues/40335
  - Fixes https://github.com/zed-industries/zed/issues/40592
  - Fixes https://github.com/zed-industries/zed/issues/40638
- As a bonus, Alt+Space now opens the system menu, since it is triggered
by the WM_SYSCHAR generated by TranslateMessage.
- VK_PROCESSKEYs are now ignored rather than being unwrapped and matched
against keybindings. This ensures that IMEs will reliably receieve
keystrokes that they express interest in. This matches the behavior of
native Windows applications.
  - Fixes https://github.com/zed-industries/zed/issues/36736
  - Fixes https://github.com/zed-industries/zed/issues/39608
  - Fixes https://github.com/zed-industries/zed/issues/39991
  - Fixes https://github.com/zed-industries/zed/issues/41223
  - Fixes https://github.com/zed-industries/zed/issues/41656
  - Fixes https://github.com/zed-industries/zed/issues/34180
  - Fixes https://github.com/zed-industries/zed/issues/41766


Release Notes:

- windows: Improved keyboard input handling for international keyboard
layouts and IMEs

John Tur created

0a4b1ac inline assistant: Mention ability to add context with @ in the placeholder (#41950)

Click to expand commit body 
This has been possible in the inline assistant for ages now and maybe
you didn't know because we didn't say anything about it! This PR fixes
that by including that you can @-mention context on it the same you can
in the agent panel.

Release Notes:

- N/A

Danilo Leal created

f9fb855 Fetch (just) enough refs in script/cherry-pick (#41949)

Click to expand commit body 
Before this change we'd download all the tagged commits, but none of
their ancestors,
this was slow and made cherry-picking fail.

Release Notes:

- N/A

Conrad Irwin created

b587a62 No-op commit to test cherry-picking (#41948)

Click to expand commit body 
Closes #ISSUE

Release Notes:

- N/A

Conrad Irwin created

1b2e38b More tweaks to CI pipeline (#41941)

Click to expand commit body 
Closes #ISSUE

Release Notes:

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

Conrad Irwin created

4339c77 Tidy up `Edit in json` footer entries (#41890)

Click to expand commit body 
Before:
<img width="350" height="109" alt="before"
src="https://github.com/user-attachments/assets/d5d3e6bd-3a65-4d7d-8585-1e4d8f72997f"
/>

After:
<img width="310" height="103" alt="after"
src="https://github.com/user-attachments/assets/40137084-7323-4a79-b95b-a020c418646b"
/>

* All items got a keybinding label
* All items were made of a non-small size, to match the text size in the
right button
* Keybindings are rendered as disabled for disabled buttons

Release Notes:

- N/A

Co-authored-by: Lukas Wirth <lukas@zed.dev>

Kirill Bulatov and Lukas Wirth created

ba7ea71 node_runtime: Improve proxy mapping (#41807)

Click to expand commit body 
Closes #ISSUE

More accurately map localhost to `127.0.0.1`. Previously we would
lowercase and simply replace all instances of localhost inside of the
URL string, meaning query parameters, username, password etc. could not
contain the string `localhost` or contain uppercase letters without
getting modified. Added a test ensuring the mapping logic works. The
previous implementation would fail this new test.

Release Notes:

- Improved the behavior of mapping `localhost` to `127.0.0.1` when
passing configured proxy urls to `node`

tidely created

cd04450 Refactor completions (#41939)

Click to expand commit body 
This is progress toward multi-word snippets (including snippets with
prefixes containing symbols)

Release Notes:

- Removed `trigger` argument in `ShowCompletions` command

---------

Co-authored-by: Conrad Irwin <conrad.irwin@gmail.com>

Andrew Farkas and Conrad Irwin created

769a8a6 windows: Automatically find Windows SDK when building gpui (#38711)

Click to expand commit body 
### Summary

This PR changes `gpui/build.rs` to look up the Windows SDK directory in
the registry instead of falling back to a hard-coded path.

---

### Problem

Currently, building `gpui` on Windows requires `fxc.exe` to be in `PATH`
or at a predefined location (unless `GPUI_FXC_PATH` is set). This
requires to maintain a certain build environment with proper paths/vars
or to install the specific SDK version.

It is possible to find the SDK automatically using the registry keys it
creates upon installation. Specifically in
`SOFTWARE\\WOW6432Node\\Microsoft\\Microsoft SDKs\\Windows\\v10.0`
branch there are:
* `InstallationFolder` telling the SDK installation location;
* `ProductVersion` telling the SDK version in use.

These keys provide enough information to locate the SDK binaries, with
added robustness:
* handles non-standard SDK installation path;
* deterministically selects the latest SDK when multiple versions are
present.

---

### Changes Made

* **Updated `crates/gpui/build.rs`**:

  * added dependency on `winreg`
  * introduced `find_latest_windows_sdk_binary()` helper
  * updated fallback logic to use registry lookup

This PR only changes the fallback location, and does not touch the
established environment-based workflow.

Release Notes:

- N/A

---

### Impact

Reduces manual configuration needed to build GPUI on Windows.

---------

Co-authored-by: John Tur <john-tur@outlook.com>

Sergey Cherepanoff and John Tur created

94ff4aa AI: Fix Github Copilot edit predictions failing to start (#41934)

Click to expand commit body 
Closes #41457 #41806 #41801

Copilot started using `node:sqlite` module which is an experimental
feature between node v22-v23 (stable in v24). The fix was passing in the
experimental flag when Zed starts the copilot LSP.

I tested this with v20.19.5 and v24.11.0. The fix got v20.19 working and
didn't affect v24.11 which was already working.

Release Notes:

- AI: Fix Github Copilot edit predictions failing to start

Anthony Eid created

1e14804 Fix integer underflow in autosave mode after delay in the settings (#41898)

Click to expand commit body 
Closes #41774 

Release Notes:

- settings_ui: Fixed an integer underflow panic when attempting to hit
the `-` sign on settings item that take delays in milliseconds

Ignasius created

cdd7d4b windows: Skip `AppendCategory` call when there are no shell links (#37926)

Click to expand commit body 
`ICustomDestinationList::AppendCategory` rejects an empty `IObjectArray`
and returns an `E_INVALIDARG` error. Error propagated and caused an
early-return from `update_jump_list()`.
<img width="1628" height="540" alt="image"
src="https://github.com/user-attachments/assets/f8143297-c71e-42a1-a505-66cd77dfa599"
/>

Release Notes:

- N/A

scuzqy created

4fd2b3f editor: Jumping to diagnostics unfolds target locations (#41932)

Click to expand commit body 
Release Notes:

- Jumping to diagnostics no longer skips over folded regions. The folded
region that contains a target diagnostic is now unfolded.

Piotr Osiewicz created

43a7f96 Improve compare_perf.yml, cherry_pick.yml (#41606)

Click to expand commit body 
Release Notes:

- N/A

---------

Co-authored-by: Nia Espera <nia@zed.dev>

Conrad Irwin and Nia Espera created

2a2e04b Add docs for settings profiles (#41931)

Click to expand commit body 
Release Notes:

- N/A

Joseph T. Lyons created

9bf212b lsp: Fix dynamic registration of document diagnostics (#41929)

Click to expand commit body 
- lsp: Fix dynamic registration of diagnostic capabilities not taking
effect when an initial capability is not specified
Gist of the issue lies within use of .get_mut instead of .entry. If we
had not created any dynamic capability beforehand, we'd miss a
registration, essentially

- **Determine whether to update remote caps in a smarter manner**

Release Notes:

- Fixed document diagnostics with Ty language server.

Piotr Osiewicz created

38cd16a Fix save_last_workspace (#41907)

Click to expand commit body 
Closes #37348

Release Notes:

- Fixed last workspace window restoration on linux/windows

localcc created

d8655f0 settings_ui: Fix dropdowns after #41036 (#41920)

Click to expand commit body 
Closes #41533

Both of the issues in the release notes that are fixed in this PR, were
caused by incorrect usage of the `window.use_state` API.
The first issue was caused by calling `window.use_state` in a render
helper, resulting in the element ID used to share state being the same
across different pages, resulting in the state being re-used when it
should have been re-created. The fix for this was to move the
`window.state` (and rendering logic) into a `impl RenderOnce` component,
so that the IDs are resolved during the render, avoiding the state
conflicts.

The second issue is caused by using a `move` closure in the
`window.use_state` call, resulting in stale closure values when the
window state is re-used.

Release Notes:

- settings_ui: Fixed an issue where some dropdown menus would show
options from a different dropdown when clicked
- settings_ui: Fixed an issue where attempting to change a setting in a
dropdown back to it's original value after changing it would do nothing

Ben Kunkle created

91d631c Evaluate zeta2 context retrieval and edit predictions (#41921)

Click to expand commit body 
This PR implements the `zeta-cli eval` command. It will:

- Run the edit prediction model if there are no cached results
- Compute precision/recall/F1 for context retrieval at the line level:
every retrieved line of context is counted as a true positive (correct
retrieval), false positive (retrieved something that was not expected),
or false negative (didn't retrieve an expected line)
- Compute similar metrics for edit predictions
- Pretty-print results, highlighting the difference between actual and
expected when printing to tty

Other changes:
- `zeta-cli predict` accepts a `--format` argument with options `md`,
`json`, `diff`
- Code restructure

Release Notes:

- N/A

---------

Co-authored-by: Piotr Osiewicz <24362066+osiewicz@users.noreply.github.com>
Co-authored-by: Agus Zubiaga <agus@zed.dev>

Oleksiy Syvokon , Piotr Osiewicz , and Agus Zubiaga created

054d2e1 Add Ben K to gpui PR reviewers (#41919)

Click to expand commit body 
Release Notes:

- N/A

Joseph T. Lyons created

982f241 git: Add support for git pull with rebase (#41117)

Click to expand commit body 
- Adds a new action `git::PullRebase` which adds `--rebase` in the final
command invoked by existing Git-Pull implementation.
- Includes the new action in "Fetch/Push" button in the Git Panel
(screenshot below)
- Adds key-binding for `git::PullRebase` in all three platforms,
following the existing key-binding patterns (`ctrl-g shift-down`)
- Update git docs to include the new action.

Sidenote: This is my first ever OSS contribution

Screenshot:

<img width="234" height="215" alt="image"
src="https://github.com/user-attachments/assets/713d068f-5ea5-444f-8d66-444ca65affc8"
/>

---

Release Notes:

- Git: Added `git: pull rebase` for running `git pull --rebase`.

Sathiyaraman M created

9f58046 Add Anthony and Cameron to gpui PR reviewers (#41914)

Click to expand commit body 
Release Notes:

- N/A

Joseph T. Lyons created

fc3e503 remote: Fix incorrect default repository selection when using remote (#41698)

Click to expand commit body 
If I understand this correctly: The `active_repo_id` uses
`get_or_insert_with`, which makes it dependent on the `RepositoryAdded`
event sequence. To ensure correct initialization of the `active_repo_id`
on the remote side, the first local `RepositoryAdded` event must
synchronously send an `UpdateRepository` to `updates_tx`.

Closes #30694

Release Notes:

- Fixed incorrect default repository selection when using remote

ᴀᴍᴛᴏᴀᴇʀ created

52c49b8 Sort PR reviewer categories (#41912)

Click to expand commit body 
Release Notes:

- N/A

Joseph T. Lyons created

07b7071 Sort PR reviewers per category (#41909)

Click to expand commit body 
Release Notes:

- N/A

Joseph T. Lyons created

75cef88 agent_ui: Render error context when failing to spawn agent thread (#41908)

Click to expand commit body 
Release Notes:

- Fixed not telling the user what went wrong when spawning ACP agents

Lukas Wirth created

46b39f0 editor: Clean up edit prediction preview when edit prediction is disabled (#40159)

Click to expand commit body 
Update the `editor::Editor.handle_modifiers_changed` method to ensure
that the `editor::Editor.update_edit_prediction_preview` method is
called even if edit prediction preview is disabled, if there's an active
edit prediction preview.

Without this change it was possible for users to get into a state where
holding the modifiers to show the prediction were part of the modifiers
used to disable edit prediction. When that keybinding was used, edit
prediction would be disabled, but the edit prediction preview would
remain as active, so the context menu for the editor would never be
shown again, as the editor would assume it was still showing the edit
prediction preview.

Closes #40056 

Release Notes:

- Fixed a bug that could cause the completions menu to stop being show
when edit predictions were disabled

---------

Co-authored-by: dino <dinojoaocosta@gmail.com>

Pranav Joglekar and dino created

fb410ab Support relative line number on wrapped lines (rework) (#41805)

Click to expand commit body 
## Add relative line numbers on wrapped lines, take 2

This is a rework of https://github.com/zed-industries/zed/pull/39268
that excludes
https://github.com/zed-industries/zed/pull/39268/commits/e7096d27a6463f6eb7c2a821637c5773b2460c10.
This commit introduced some line number rendering issues as described in
https://github.com/zed-industries/zed/issues/41422.

While @ConradIrwin suggested we try to pass in the buffer rows from the
calling method instead of the snapshot, that
appears to have had unintended consequences and I don't think the two
calculations were intended to do the same thing. Hence, this PR has
removed those changes.

This PR also includes the migration fix originally done by @MrSubidubi
in https://github.com/zed-industries/zed/pull/41351.

## Original PR description and release notes.

**Problem:** Current relative line numbering creates a mismatch with
vim-style navigation when soft wrap is enabled. Users must mentally
calculate whether target lines are wrapped segments or logical lines,
making `<n>j/k` navigation unreliable and cognitively demanding.

**How things work today:**
- Real line navigation (`j/k` moves by logical lines): Requires
determining if visible lines are wrapped segments before jumping. Can't
jump to wrapped lines directly.
- Display line navigation (`j/k` moves by display rows): Line numbers
don't correspond to actual row distances for multi-line jumps.

**Proposed solution:** Count and number each display line (including
wrapped segments) for relative numbering. This creates direct
visual-to-navigational correspondence, where the relative number shown
always matches the `<n>j/k` distance needed.

**Benefits:**
- Eliminates mental overhead of distinguishing wrapped vs. logical lines
- Makes relative line numbers consistently actionable regardless of wrap
state
- Preserves intuitive "what you see is what you navigate" principle
- Maintains vim workflow efficiency in narrow window scenarios

Also explained and discussed in
https://github.com/zed-industries/zed/discussions/25733.

Release Notes:

- Added support for counting wrapped lines as relative lines and for
displaying line numbers for wrapped segments. Changes
`relative_line_numbers` from a boolean to an enum: `enabled`,
`disabled`, or `wrapped`.

Thomas Heartman created

1b93242 project_panel: Add hidden files glob patterns and action toggle hidden files visibility (#41532)

Click to expand commit body 
This PR adds the ability to configure which files are considered
"hidden" in the project panel and toggle their visibility with a
keyboard shortcut. Previously, the editor hardcoded dotfiles as hidden -
now users can customize the pattern and quickly show/hide them.

### Release Notes

- Added `project_panel::ToggleHideHidden` action with keyboard shortcuts
to toggle visibility of hidden files
- Added configurable `hidden_files` setting to customize which files are
marked as hidden (defaults to `**/.*` for dotfiles)

### Motivation

This change allows users to:
1. Quickly toggle hidden file visibility with a keyboard shortcut
2. Customize which files are considered "hidden" beyond just dotfiles
3. Better organize their project panel by hiding build artifacts, logs,
or other generated files

### Usage

**Toggle hidden files:**
- **macOS:** `cmd-alt-.`
- **Linux:** `ctrl-alt-.`
- **Windows:** `ctrl-alt-.`

**Customize patterns in settings:**
```json
{
  "hidden_files": ["**/.*", "**/*.tmp", "**/build/**"]
}
```

### Changes

**Core Implementation:**
- Added `hidden_files` setting (defaults to `**/.*` to match current
dotfile behavior)
- Replaced hardcoded `name.starts_with('.')` logic with configurable
pattern matching using `PathMatcher`
- Hidden status propagates through directory hierarchies (if a directory
is hidden, all children inherit that status)

**User-Facing:**
- Added `ToggleHideHidden` action in the project panel
- Added keyboard shortcuts for all platforms
- Added settings UI entry for configuring `hidden_files` patterns

**Testing:**
- Added comprehensive test coverage validating default behavior, custom
patterns, propagation, and settings changes

### Implementation Notes

- Uses `PathMatcher` for efficient glob matching
- Settings changes automatically trigger worktree re-indexing
- No breaking changes - defaults maintain current behavior (hiding
dotfiles)

---

**Disclaimer:** This was implemented with a fair amount of copy/paste
(particularly the gitignore handling), trial and error, and a healthy
dose of Claude.

### Screenshots

**Project Panel with hidden files visible:**
<img width="1368" height="935" alt="Screenshot 2025-10-30 at 3 15 53 AM"
src="https://github.com/user-attachments/assets/1cbe90ce-504c-4f9b-bca8-bef02ab961be"
/>

**Project Panel with hidden files hidden:**
<img width="1363" height="917" alt="Screenshot 2025-10-30 at 3 16 07 AM"
src="https://github.com/user-attachments/assets/9297f43e-98c7-4b19-be8f-3934589d6451"
/>

**Toggle action in command palette:**
<img width="565" height="161" alt="Screenshot 2025-10-30 at 3 17 26 AM"
src="https://github.com/user-attachments/assets/4dc9e7b6-9c29-4972-b886-88d8018905da"
/>

Release Notes:

- Added the ability to configure glob patterns for files treated as
hidden in the project panel using the `hidden_files` setting.
- Added an action `project panel: toggle hidden files` to quickly show
or hide hidden files in the project panel.

---------

Co-authored-by: Smit Barmase <heysmitbarmase@gmail.com>

B. Collier Jones and Smit Barmase created

fb6e41d remote_server: Fix panic due to invalid settings access (#41904)

Click to expand commit body 
Closes https://github.com/zed-industries/zed/issues/41860

Release Notes:

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

Lukas Wirth created

0ec31db util: Add missing quotes in shell env capturing on windows (#41902)

Click to expand commit body 
Release Notes:

- Fixed shell env capturing failing if zed is installed on a path with
whitespace in it

Lukas Wirth created

a262ca1 util: Log JSON with envs if failed to deserialize (#41894)

Click to expand commit body 
Release Notes:

- N/A

Jakub Konka created

6a38d69 Fix autoupdate nuking the app on Windows (#41571)

Click to expand commit body 
Closes #41477

Release Notes:

- N/A

localcc created

95feefc remote: Fix terminal crash on Elvish shell (#41893)

Smit Barmase created

a827f25 file_finder: Respect .gitignore and file_scan_inclusions with ** in glob (#40654)

Click to expand commit body 
Closes #39037 

Previously, the code split the `**/.env` glob in `file_scan_inclusions`
into two sources for the `PathMatcher`: `["**", "**/.env"]`. This
approach works for directories, but including `**` will match all
directories and their files. To address this, I now select the
appropriate `PathMatcher` using only `**/.env` when specifically
targeting a file to determine whether to include it in the file finder.

Release Notes:

- Fixed: respect `.gitignore` and `file_scan_inclusions` settings with
`**` in glob for file finder

---------

Signed-off-by: Benjamin <5719034+bnjjj@users.noreply.github.com>
Co-authored-by: Julia Ryan <juliaryan3.14@gmail.com>

Coenen Benjamin and Julia Ryan created

cc6208b Show inlay label parts' tooltips if those are present and hovered (#41889)

Click to expand commit body 
Part of https://github.com/zed-industries/zed/issues/33715


https://github.com/user-attachments/assets/d2d6f47d-3974-4c8c-aab9-9046891186bf

Unlike VSCode that does more advanced hovering, this one only works for
inlays with tooltip LSP data in them.

Release Notes:

- Started to show inlay label parts' tooltips when they are hovered

Co-authored-by: Lukas Wirth <lukas@zed.dev>

Kirill Bulatov and Lukas Wirth created

8d15ec7 vim: Fix mini delimiters in multibuffer (#41834)

Click to expand commit body 
- Update `vim::object::find_mini_delimiters` in order to filter out the
  ranges before calling `vim::object::cover_or_next`, ensuring that the
  provided ranges are converted from multibuffer space into buffer
  space.
- Remove the `range_filter` from `vim::object::cover_or_next` was the
  `find_mini_delimiters` function is the only caller and no longer uses
  it

Closes #41346 

Release Notes:

- Fixed a crash that could occur when using `vim::MiniQuotes` and
`vim::MiniBrackets` in a multibuffer

Dino created

ca5a4dc terminal: Resolve env based on the project dir on the target (#41867)

Click to expand commit body 
Prior to this change we would always resolve envs when spawning a new
terminal window based on the inherited CLI environment. This works fine
as long as we open a new Zed instance in the terminal when using it
locally only. When using Zed connected to a remote server, it would not
be meaningful however. WIth this change, we correctly ping the remote
for the project-local envs and use that instead. This change should also
fix a pesky issue when updating Zed - after Zed restarts, opening a new
terminal window will not run `direnv` for example.

Release Notes:

- N/A

Jakub Konka created