Adds a `--repeat` flag to the zeta eval that runs each example as many
times as specified. Also makes the output nicer in a few ways.
Release Notes:
- N/A
---------
Co-authored-by: Ben Kunkle <ben@zed.dev>
Co-authored-by: Michael <michael@zed.dev>
The existing sorting approach when faced with `Dir1`, `dir2`, `Dir3`,
would only get as far as comparing the stems without numbers (`dir` and
`Dir`), and then the lowercase-first tie breaker in that function would
determine that `dir2` should come first, resulting in an undesirable
order of `dir2`, `Dir1`, `Dir3`.
This patch defers tie-breaking until it's determined that there's no
other difference in the strings outside of case to order on, at which
point we tie-break to provide a stable sort.
Natural number sorting is still preserved, and mixing different cases
alphabetically (as opposed to all lowercase alpha, followed by all
uppercase alpha) is preserved.
Closes#41080
Release Notes:
- Fixed: ProjectPanel sorting bug
Screenshots:
Before | After
----|---
<img width="237" height="325" alt="image"
src="https://github.com/user-attachments/assets/6e92e8c0-2172-4a8f-a058-484749da047b"
/> | <img width="239" height="325" alt="image"
src="https://github.com/user-attachments/assets/874ad29f-7238-4bfc-b89b-fd64f9b8889a"
/>
I'm having trouble reasoning through what was previously going wrong
with `docs` in the before screenshot, but it also seems to now appear
alphabetically where you'd expect it with this patch
---------
Co-authored-by: Smit Barmase <heysmitbarmase@gmail.com>
This adds additional comments which were left out from #42494 by
accident. Namely, it describes why we have additional custom
highlighting in `highlights.scm` for the Typescript grammar.
Release Notes:
- N/A
Closes#7844
This change uses tree-sitter highlights as a method of showing
typescript errors prettily, keeping regex as simple as possible:
<img width="832" height="446" alt="Screenshot 2025-11-11 at 3 40 24 PM"
src="https://github.com/user-attachments/assets/0b3b6cf1-4d4d-4398-b89b-ef5ec0df87ec"
/>
It covers three main areas:
1. Diagnostics
Diagnostics are now rendered with language-aware typescript, by
providing the project's language registry.
2. Vtsls
The LSP provider for typescript now implements the
`diagnostic_message_to_markdown` function in the `LspAdapter` trait, so
as to provide Diagnostics with \`\`\`typescript...\`\`\`-style code
blocks for any selection of typescript longer than one word. In the
single-word case, it simply wraps with \`\`
3. Typescript's `highlights.scm`
`vtsls` doesn't provide strictly valid typescript in much of its
messaging. Rather, it returns a message with snippets of typescript
values which are invalid. Tree-sitter was not properly highlighting
these snippets because it was expecting key-value formats. For instance:
```
type foo = { foo: string; bar: string; baz: number[] }
```
is valid, whereas simply
```
{ foo: string; bar: string; baz: number[] }
```
is not.
Therefore, highlights.scm needed to be adjusted in order to
pattern-match on literal values that might be returned from the vtsls
diagnostics messages. This was done by a) identifying arrow functions on
their own, and b) augmenting the `statment_block` pattern matching in
order to match on values which were clearly object literals.
This approach may not be exhaustive - I'm happy to work on any
additional cases we might identify from `vtsls` here - but hopefully
demonstrates an extensible approach to making these messages look nice,
without taking on the technical burden of extensive regex.
Release Notes:
- Show pretty TypeScript errors with language-aware Markdown.
- Improves the determinism of the search step for better cache
reusability
- Adds a `--cache force` mode that refuses to make any requests or
searches that aren't cached
- The structure of the `zeta-*` directories under `target` has been
rethought for convenience
Release Notes:
- N/A
---------
Co-authored-by: Agus <agus@zed.dev>
Closes #ISSUE
Improves error reporting for various failure modes of zeta2, including
failing to parse the `<old_text>`/`<new_text>` pattern, and the contents
of `<old_text>` failing to match.
Additionally, makes it so that evals are checked out into a worktree
with the _repo_ name instead of the _example_ name, in order to make
sure that the eval name has no influence on the models prediction. The
repo name worktrees are still namespaced by the example name like
`{example_name}/{repo_name}` to ensure evals pointing to the same repo
do not conflict.
Release Notes:
- N/A *or* Added/Fixed/Improved ...
---------
Co-authored-by: Agus <agus@zed.dev>
This PR introduces `Nearest` scrolling strategy to `UniformList`. This
is now used in completions menu and the picker to choose the appropriate
scrolling strategy depending on movement direction. Previously,
selecting the next element after the last visible item caused the menu
to scroll with `ScrollStrategy::Top`, which scrolled the whole page and
placed the next element at the top. This behavior is inconsistent,
because using `ScrollStrategy::Top` when moving up only scrolls one
element, not the whole page.
https://github.com/user-attachments/assets/ccfb238f-8f76-4a18-a18d-bbcb63340c5a
The solution is to introduce the `Nearest` scrolling strategy which will
internally choose the scrolling strategy depending on whether the new
selected item is below or above currently visible items. This ensures a
single-item scroll regardless of movement direction.
https://github.com/user-attachments/assets/8502efb8-e2c0-4ab1-bd8d-93103841a9c4
I also noticed that some functions in the file have different logic
depending on `y_flipped`. This appears related to reversing the order of
elements in the list when the completion menu appears above the cursor.
This was a feature suggested in #11200 and implemented in #23446. It
looks like this feature was reverted in #27765 and there currently seem
to be no way to have `y_flipped` to be set to `true`.
My understanding is that the opposite scroll strategy should be used if
`y_flipped`, but since there is no way to enable this feature to test it
and I don't know if the feature is ever going to be reintroduced I
decided not to include it in this PR.
Release Notes:
- gpui: Add 'Nearest' scrolling strategy to 'UniformList'
Closes#35142
In addition to cleaning up the terminals, also stops the conversation.
Release Notes:
- Restoring a checkpoint now stops the agent conversation.
Reverts zed-industries/zed#42480
This panics on Nightly in cases where anchor might not be valid for that
snapshot. Taking it back before the cutoff.
Release Notes:
- N/A
Update `Vim.change_surround` in order to ensure that there's no
overlapping edits by keeping track of where the open string range ends
and ensuring that the closing string range start does not go lower than
the open string range end.
Closes#42316
Release Notes:
- Fix vim's change surrounds `cs` inserting spaces with quotes by
preventing overlapping edits
---------
Co-authored-by: dino <dinojoaocosta@gmail.com>
- switch the bot from looking at the `bug/crash` labels which we don't
use anymore to the Bug/Crash issue types which we do use
- shorten the period of time after which a bug is suspected to be stale
(with our pace they can indeed be outdated in 60 days)
- extend the grace period for someone to come around and say nope, this
problem still exists (people might be away for a couple of weeks).
Release Notes:
- N/A
This reverts commit 082b80ec89.
This broke clicking, e.g. in snippets like
```rs
let x = vec![
1, 2, //
3,
];
```
clicking between `2` and `,` is quite off now.
Release Notes:
- N/A
Fixes a regression caused by #42135 where LSP and DAP binaries weren't
being used from `PATH` env var
Now we absolutize the path if (path is relative AND (path has multiple
components OR path exists in worktree)).
- Relative paths with multiple components might not exist in the
worktree because they are ignored. Paths with a single component will at
least have an entry saying that they exist and are ignored.
- Relative paths with multiple components will never use the `PATH` env
var, so they can be safely absolutized
Release Notes:
- N/A
<img width="395" height="444" alt="Screenshot 2025-11-11 at 4 04 57 PM"
src="https://github.com/user-attachments/assets/8da68721-6e33-4d01-810d-4aa1e2f3402d"
/>
Discussed with @danilo-leal and we're going with the "it's checked in
both places" design!
Closes#40910
Release Notes:
- Recommended AI models now still appear in their normal category in
addition to "Recommended:"
