Commit Graph
8 Commits
Author SHA1 Message Date
Ben Kunkle aecaacdb44 git: Move diff hunk operations onto buffer diffs (#63556)
# Objective

Keep Git-specific diff-base knowledge in `GitStore`, where Git
operations are defined.

Related to FR-135.

## Solution

Give each `BufferDiff` its supported hunk operations. Keep editor hooks
limited to rendering custom controls.

## Self-Review Checklist:

- [x] I've reviewed my own diff for quality, security, and reliability
- [x] Unsafe blocks (if any) have justifying comments
- [x] The content adheres to Zed's UI standards
([UX/UI](https://github.com/zed-industries/zed/blob/main/CONTRIBUTING.md#uiux-checklist)
and
[icon](https://github.com/zed-industries/zed/blob/main/crates/icons/README.md)
guidelines)
- [x] Tests cover the new/changed behavior
- [x] Performance impact has been considered and is acceptable

---

Release Notes:

- N/A
2026-09-01 15:52:33 +00:00
Miguel Raz Guzmán Macedo 513ad12cef Avoid redundant string allocations (#63377)
Remove unnecessary intermediate `String` allocations when constructing
`Arc<str>` and `SharedString` values. Use static `SharedString` storage
for fixed worktree picker identifiers.

Validation:

- `cargo fmt --all -- --check`
- `tooling/lints/single-lint owned_string_into_shared -p editor -p
git_ui_core -p toolchain_selector`
- `cargo test -p editor -p git_ui_core -p toolchain_selector`

Release Notes:

- N/A
2026-08-28 17:31:36 +00:00
Smit Barmase 3122f37ff1 git_ui_core: Fix worktree creation for bare repositories (#63351)
Follow-up https://github.com/zed-industries/zed/pull/55153
Fixes
https://github.com/zed-industries/zed/discussions/54553#discussioncomment-17091833

When a repository stores its Git data in `~/work/zed/.bare`, creating a
linked worktree from Zed is broken in two visible ways:

- Zed creates `random-code` at:
```
~/work/worktrees/.bare/random-code/.bare
```
instead of:
```
~/work/worktrees/zed/random-code/zed
```

- Because the checkout ends in `.bare`, Zed displays `.bare` as its
worktree name in the title bar, worktree picker, and agent thread
sidebar. Every worktree Zed creates from that repository gets the same
`.bare` label instead of its generated name.

This problem is specific to the path Zed creates. Worktrees using either
of these layouts are not malformed:
```
~/work/worktrees/zed/random-code/zed
~/work/zed/some-worktree
```

https://github.com/zed-industries/zed/pull/55153 introduced repository
identities for bare repositories. This PR uses that identity in the
remaining worktree naming and creation paths.

There are three repository layouts to account for:

- In a normal repository, the common Git directory is `~/work/zed/.git`,
so Zed derives the main worktree at `~/work/zed`. Existing behavior
remains unchanged.

- With `~/work/zed/.bare`, Git has no main worktree. A checkout named
main is still a linked worktree. Zed now treats `~/work/zed` as the
repository identity, creates new worktrees under `~/work/worktrees/zed`,
and displays their generated names.

- With a bare repository named `~/work/foo.git`, `foo.git` is the
repository identity. Its worktree path remains
`~/work/worktrees/foo.git/random-code/foo.git`. Zed uses foo.git to
derive the worktree name random-code, while displaying the repository
itself as foo.

Existing malformed worktrees such as
`~/work/worktrees/.bare/random-code/.bare` are not moved and may
continue to display `.bare`.

Broken:

<p>
<img width="238" height="58" alt="image"
src="https://github.com/user-attachments/assets/1504e4d0-c34a-471f-8e76-c5ec062b1d77"
/>
</p>
<p>
<img width="274" height="49" alt="image"
src="https://github.com/user-attachments/assets/13266ca9-4e64-426c-b03c-e23c924a3597"
/>
</p>
<p>
<img width="574" height="85" alt="image"
src="https://github.com/user-attachments/assets/0fca58c7-5be7-4d36-9067-066d0b80caf8"
/>
</p>

Release Notes:

- Fixed linked worktree names and creation paths for bare Git
repositories.
2026-08-28 12:15:55 +00:00
Enzo GAZZOLIandKirill Bulatov 1155a5becd Render control characters in tab titles and project panel entries (#62875)
# Objective

Fixes #62664

File names may contain control characters (`\n`, `\r`, `\t`, …) on most
platforms. Zed rendered them verbatim, so a file named `notes\ndraft.md`
broke the layout of its tab and of its project panel entry instead of
showing a readable name. The same held anywhere else a name reached the
screen without going through `Label::single_line`.

## Solution

The substitution lives in one shared place,
`ui::utils::replace_control_characters`, reached through
`Label::single_line`. `\n` keeps rendering as the `⏎` Zed already used,
so nothing that renders correctly today changes; every other C0 control
character maps onto the Unicode "Control Pictures" block by a single
rule (`U+2400 + code point`), giving `␉` for tab and `␍` for carriage
return. `DEL` maps to `␡`. C1 controls are left alone, having no
equivalent there.

Where that is wired in:

- **`Label::single_line`** — replaces its previous `\n`-only
substitution. The project panel already used it everywhere, including
folded directory components, so the panel is fixed by this alone.
- **Editor tabs** and the **default `Item::tab_content`**, neither of
which called `single_line`.
- **Terminal tab titles**, which render through their own `tab_content`.
Those titles come from OSC escape sequences, so the control characters
in them are attacker-controlled.
- **`HighlightedLabel::single_line`**, which never substituted anything
— not even `\n`. See below.
- **The file finder**, for both the file name and the path.
- **Nine `tab_content` overrides in git_ui**, by deleting them.
- **`HighlightedMatchWithPaths`**, for both the match label and its
paths — it backs the tab switcher and the outline panel, among others.

### `HighlightedLabel` needs its offsets moved, not just its text

Its highlight indices are byte offsets, and every stand-in is wider in
bytes than the character it replaces, so the offsets have to move with
the text. Left alone they index into the middle of a character, and
`highlight_ranges` slices the string at exactly those offsets — which
panics in release builds too. The constructor's `debug_panic!` does not
cover this, since `single_line` runs after construction.

So `replace_control_characters_remapping_offsets` substitutes and remaps
in one pass, over an old-to-new byte offset table built the way
`ensure_uniform_list_compatible_label` does it in `lsp_store`. This also
fixes the call sites that already asked for a single line, including the
branch picker and the tabular column filter values.

### Deleting the git_ui overrides

Nine files replicated the default `Item::tab_content` verbatim except
for the colour, hardcoding `selected ? Default : Muted` instead of
`params.text_color()`. Deleting them lets those tabs inherit the fixed
default, and fixes a second bug along the way: none of them dimmed when
the pane lost focus. Titles and icons are unaffected — they come from
`tab_content_text` and `tab_icon`, which the default calls.
`git_graph.rs` already did it this way.

### Tabular column headers

These never went through a `Label` at all: the raw `SharedString` was a
child of a `div`. They now display stand-ins, while right-click-copy
still yields the real column name.

Two notes for reviewers:

- Both helpers avoid allocating when there is nothing to replace, so the
common path is untouched. The previous `self.label.replace('\n', "⏎")`
allocated a `String` on every render for every label — including every
visible project panel entry, every frame.
- Every stand-in is exactly one character, so character offsets are
preserved and the existing `truncate_and_trailoff` math on tab titles
stays correct.

## Testing

- Unit tests cover both helpers: each control character's substitution,
repeated occurrences, multi-byte characters (accents, emoji), the
borrow-vs-own behaviour, the preserved character count, C1 passthrough,
and — for the remapping — offsets before, at and after a replacement,
offsets at and past the end, and the invariant that every remapped
offset lands on a character boundary.
- Unit tests cover `Label::single_line` and
`HighlightedLabel::single_line`, the entry points every call site uses.
`highlighted_label.rs` had no tests before.
- `cargo test -p ui -p file_finder -p workspace -p project_panel -p
git_ui -p git_ui_core -p terminal_view -p tabular_data_preview` passes
(774 tests), as do `cargo clippy` and `cargo fmt --check`.
- No existing test changes behaviour: tests assert on
`tab_content_text`, which is untouched; only rendering paths changed.
- Verified manually on macOS (aarch64) — see Showcase.

## Self-Review Checklist:

- [x] I've reviewed my own diff for quality, security, and reliability
- [x] Unsafe blocks (if any) have justifying comments — none added
- [x] The content adheres to Zed's UI standards
- [x] Tests cover the new/changed behavior
- [x] Performance impact has been considered and is acceptable

## Showcase

A directory holding `report<TAB>table.csv`, `notes<LF>draft.md`,
`carriage<CR>return.txt` and a normally named `normal.txt`, all four
open as tabs.

**Before**


![before](https://raw.githubusercontent.com/Taqinou/zed/0d5ee2b4f94db7bcb713298c38cf411458b75239/62664/before.png)

**After**


![after](https://raw.githubusercontent.com/Taqinou/zed/0d5ee2b4f94db7bcb713298c38cf411458b75239/62664/after.png)

Three things change:

1. **The `notes<LF>draft.md` tab** is the clearest one. Before, the
literal newline splits the tab across two lines — `notes` above
`draft.md` — distorting the whole tab bar. After, it stays on one line.
2. **`carriage<CR>return.txt` in the project panel** renders as
`carriagereturn.txt` before: the `\r` is completely invisible, so the
name reads as one word and there is no way to tell a character is there.
After, it renders as `carriage␍return.txt`.
3. **`report<TAB>table.csv`** shows a bare gap before, indistinguishable
from a space. After, it renders as `report␉table.csv`.

Note that `notes<LF>draft.md` looks the same in the project panel in
both shots: that entry already called `single_line`, which already
handled `\n`. That is precisely the asymmetry this PR removes.

### File finder

Before, the newline also breaks the list itself — the
`notes<LF>draft.md` row grows to two lines and overlaps the row beneath
it.

**Before**

![file finder
before](https://raw.githubusercontent.com/Taqinou/zed/0d5ee2b4f94db7bcb713298c38cf411458b75239/62664/finder-before.png)

**After**

![file finder
after](https://raw.githubusercontent.com/Taqinou/zed/0d5ee2b4f94db7bcb713298c38cf411458b75239/62664/finder-after.png)

Searching for `table` exercises the offset remapping, the match falling
after the control character:

**Before**

![match
before](https://raw.githubusercontent.com/Taqinou/zed/0d5ee2b4f94db7bcb713298c38cf411458b75239/62664/match-before.png)

**After**

![match
after](https://raw.githubusercontent.com/Taqinou/zed/0d5ee2b4f94db7bcb713298c38cf411458b75239/62664/match-after.png)

One thing this PR does **not** cover: the breadcrumb below the tab bar
still renders the raw tab character, as it goes through a different
path.

## Release Notes:

- Fixed file names containing control characters, such as tabs and
newlines, rendering unreadably in tabs, the project panel, the file
finder, the tab switcher and terminal tab titles.

---------

Co-authored-by: Kirill Bulatov <kirill@zed.dev>
2026-08-19 21:10:56 +00:00
tidely 12f3994269 git_ui: Dismiss askpass prompts when requests end (#61292)
# Objective

Fixes #47623

- When authenticating git commands using a security key through
`askpass`. The modal which asks for user presence does not get dismissed
even after the git command finishes successfully.

## Solution

Add a cancellation task to the `AskPassModal`, which gets dropped when
the requested operation completes. This cancellation task then dismisses
the modal.

## Testing

- I've tried authentication through `askpass` using my own security key.
Testing both successful and failed authentication.
- I've added tests which confirm that the modal gets dismissed when a
task is cancelled, and that the cancellation is triggered when
`ask_password` Task gets dropped.

Willing to pair on review, message me on Slack. Showcase video left out
because it would leak private information.

## Self-Review Checklist:

- [x] I've reviewed my own diff for quality, security, and reliability
- [x] Unsafe blocks (if any) have justifying comments
- [x] The content adheres to Zed's UI standards
([UX/UI](https://github.com/zed-industries/zed/blob/main/CONTRIBUTING.md#uiux-checklist)
and
[icon](https://github.com/zed-industries/zed/blob/main/crates/icons/README.md)
guidelines)
- [x] Tests cover the new/changed behavior
- [x] Performance impact has been considered and is acceptable

Release Notes:

- Fixed authentication prompts not dismissing automatically when using
security keys with ssh
2026-08-18 10:02:20 +00:00
Piotr Osiewicz ff50948537 deps: Bump pathfinder_simd & fix upcoming rustc warnings (#62170)
While building Zed with nightly rustc I've noticed it doesn't compile
because of good old pathfinder_simd. It also emits a bunch of warnings
about use of f64 literals where f32 is expected, so I've fixed them - it
should make future upgrades more straightforward.
2026-08-07 10:16:57 +00:00
Ben Kunkle 81db8c68b2 workspace: Choose the replacement workspace inside MultiWorkspace::remove (#62141)
# Objective

Follow-up to #62028. When archiving removed a thread's workspace, the
replacement
was derived from whichever sidebar row sat adjacent to the archived one.
Rows are
a flat list with project headers as separators, so archiving the oldest
thread in
a project selected the next project's newest thread and moved the user
out of the
project they were working in — and when that neighbor was remote,
connected to
its host just to pick a fallback.

# Solution

- `MultiWorkspace::remove` now chooses the replacement itself from a
`RemovalIntent` (keep the project vs close it): a live workspace in the
same
project, the project's own roots when kept, the nearest retained
neighbor,
an adjacent local project, then an empty workspace. The workspaces being
  removed are excluded from that search in one place instead of five.
- Sidebar neighbor selection stays within the entry's project section.
- The three archive paths now share one removal orchestration, and the
three
  "open the closed workspace first" helpers collapse into one.

# Testing

New tests cover: the keep-project fallback, neighbor selection staying
inside the
project section, and archiving with a mock-remote neighboring project
never
connecting to it. Each was verified to fail against the previous
behavior.

## Self-Review Checklist:

- [x] I've reviewed my own diff for quality, security, and reliability
- [x] Unsafe blocks (if any) have justifying comments
- [x] The content adheres to Zed's UI standards
([UX/UI](https://github.com/zed-industries/zed/blob/main/CONTRIBUTING.md#uiux-checklist)
and
[icon](https://github.com/zed-industries/zed/blob/main/crates/icons/README.md)
guidelines)
- [x] Tests cover the new/changed behavior
- [x] Performance impact has been considered and is acceptable

---

Release Notes:

- Fixed archiving a thread switching the window to a different project.
- Fixed a possible crash when archiving a thread while its window was
closing.
2026-08-05 02:48:46 +00:00
Kirill Bulatov 01961bef59 Speed up cargo build -p zed compilation (#62059)
Splits the crate graph.
Before, the compilation graph: `editor → picker_preview → search →
project_panel → open_path_prompt → recent_projects → title_bar →
collab_ui → zed`
After, the compilation graph: `editor → picker_preview → search →
agent_ui → sidebar → zed`

With sccache disabled and project fully built, `touch
crates/editor/src/editor.rs` and `cargo build -p zed` took time
Before (0500fd9ee9): 13.19s
After: 11.85s (-10.2% speed up)

Each commit contains a separate compilation instructions, and a test in
the final commit.
The main approach is to split coupled crates and replace them with
`zed_actions` and move some shared functionality into `git_ui_core` new,
shared module.

I've also tried to add a test to prevent common pitfalls, but not sure
I'm happy with the end result — can remove it if it looks too synthetic.

Release Notes:

- N/A
2026-08-01 19:42:37 +00:00