Skip to content

test(switch): regen alt-l picker snapshot for the new Remote⇅ column - #3227

Merged
max-sixty merged 1 commit into
mainfrom
picker-alt-l-gutter
Jun 25, 2026
Merged

max-sixty merged 1 commit into
mainfrom
picker-alt-l-gutter

Conversation

@max-sixty

Copy link
Copy Markdown
Owner

Follow-up to #3226. The picker list gained a Remote⇅ column and dropped Age (from concurrent picker work) in the window between when test_switch_picker_alt_l_does_not_hscroll's snapshot was generated and when #3226 merged. Both PRs were green independently, but main ended up with the new columns plus the old-column snapshot, so the test fails on main.

This regenerates the snapshot for the current layout. The assertion is unchanged in spirit — the worktree-status gutter sigils (@/+) stay intact after alt-l/alt-h, proving the keys still don't horizontally scroll the list; only the surrounding column header/data moved.

This was written by Claude Code on behalf of max

The picker list gained a Remote⇅ column and dropped Age between when
test_switch_picker_alt_l_does_not_hscroll's snapshot was generated and
when #3226 merged, so the committed frame no longer matched the rendered
list (the test failed on main). Regenerate it — the gutter sigils stay
intact after alt-l/alt-h, which is what the test asserts.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
@worktrunk-bot

Copy link
Copy Markdown
Collaborator

Approval stands — the one red check is an unrelated Windows flake, not this PR.

test (windows) failed on test_switch_picker_alt_r_keeps_cursor_sticky, a different test from the one this PR regenerates (test_switch_picker_alt_l_does_not_hscroll, which passed on all three platforms). The failure is a PTY content-wait timeout, not a snapshot mismatch:

Timed out after 30s waiting for expected content "wt-keep has no uncommitted changes" to appear on screen.

This PR only edits a snapshot data file and changes no production code, so it can't affect the interactive picker's runtime behavior. The interactive picker tests were only just enabled on Windows in #3221, so this is a first-seen timing flake on a loaded Windows runner — the preview pane hadn't repainted wt-keep's status within the 30s window. A rerun of the failed job should clear it; leaving the approval in place per the don't-self-dismiss-over-unrelated-flakes policy.

@max-sixty
max-sixty merged commit 2b245fa into main Jun 25, 2026
49 of 50 checks passed
@max-sixty
max-sixty deleted the picker-alt-l-gutter branch June 25, 2026 05:30
max-sixty added a commit that referenced this pull request Jun 25, 2026
…rees

Merge: pulls in #3231, #3223, #3229, #3227, #3157. #3223 (unify preview tabs
+ dim empty diff tabs) added `comments_fetched` / `local_content_slots` to
`PickerHandler`; threaded both through `PipelineFactory::spawn`'s handler
construction and combined the `items` import with this branch's shortcut
types. Regenerated the `empty_all_local_diffs` preview snapshot (#3223's new
snapshot captured the pre-shortcut controls line).

Fix (review feedback from worktrunk-bot on #3233): `alt-y` on a detached
worktree row copied the literal `"(detached)"` — the `branch_name()` fallback
the side table stored. `RowShortcutData.branch` is now `Option<String>`,
carried from the row's raw `ListItem.branch` (and `Some(head_branch)` for
`--prs` rows); the `alt-y` callback `.and_then`s it, so a detached row no-ops
just as `alt-o` no-ops on a row with no PR. Covered by
`shortcut_table_branch_is_none_for_detached_worktree`.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants