list: restore a valid keyboard cursor after search - #6
Draft
grishy wants to merge 2 commits into
Draft
Conversation
Search can replace results after the previous render cached its row counts. Validate the preferred cursor against current results at search start and completion, falling back to the first nonempty section. Resolve the preferred item by its committed value without committing the search cursor.
grishy
changed the title
grishy
added a commit
that referenced
this pull request
Sep 6, 2026
Validate the preferred cursor against current results after synchronous and asynchronous search. Clear stale deferred scrolling before scheduling the selected row. Keep committed Select values separate from the keyboard cursor. Includes the reviewed fork PR #6 implementation, regression tests and API docs.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
I made this change while developing my application and am now contributing it upstream.
This patch restores a valid keyboard cursor after list search results change, without changing a Select's committed value.
The current code chooses a cursor from the previous render's row cache before asynchronous results arrive. That can leave the cursor missing or pointing into an empty first section.
The list now checks the delegate's current item counts at search start and completion. An optional
preferred_selected_indexcan restore a specific item; an absent or invalid preference falls back to the first row in the first non-empty section. Empty results clear the cursor. Stale scroll requests are discarded.The searchable-list adapter finds the committed item by value, so filtering can move it to a different row. Existing delegates need no changes. This does not commit a selection or reset the query when a popup reopens.
How to Test
The focused tests cover asynchronous and synchronous results, invalid preferences, empty sections and results, and the deferred scroll target:
cargo test -p gpui-component --lib search_cursor_tests --lockedThe cursor regression failed before the fix and passes with it. The Select integration test starts with Go committed:
Tests and code checks
cargo test -p gpui-base -p gpui-component --lib --locked cargo clippy -p gpui-base -p gpui-component -p gpui-component-story --all-targets --locked -- -D warnings771 base and 420 component tests passed. Changed Rust files also passed formatting checks, and the contribution diff passed the whitespace check.
Native checks in
MTL_HUD_ENABLED=1 ./script/run-story-macos Listconfirmed that typing a query selects a real row after the empty first section, Down moves to the next row, and clearing restores the first row. The demo retains old results until PR #8; empty and asynchronous replacement are covered by the automated tests, not this demo.Windows and Linux runtime behavior has not been tested.
Checklist
AI assistance
OpenAI’s
gpt-6-astrahelped isolate the change and prepared the patch, tests, and PR description. The reported checks were run by the agent.