test(pty): cover add-note context rows

[benvinegar]

Summary

• add Tuistory coverage showing context/unchanged rows expose [+]
• verify a clicked context-row draft can be saved with Ctrl-S

Tests

• bun test test/pty/ui-integration.test.ts -t "context-row add-note" ×5
• bun test test/pty/ui-integration.test.ts -t "add-note|add-note drafts"
• bun run typecheck
• bun run lint

[greptile-apps]

Greptile Summary

This PR adds a single PTY integration test verifying that context/unchanged rows in a deletion-only diff expose the [+] add-note affordance and that a note drafted from such a row can be saved with Ctrl-S.

• The new test mirrors the existing deletion-row and stack-row coverage, reusing createDeletionOnlyFilePair and targeting the unchanged keep = true line as the context row under test.
• The fixture and session parameters (cols: 120, rows: 16) are identical to the adjacent deletion-only test, keeping the setup consistent.

Confidence Score: 4/5

Safe to merge — change is test-only and introduces no production code paths.

The test correctly covers the intended context-row scenario and follows established patterns throughout the file. The one concern is that revealAddNoteNear's row-fallback logic could allow the test to pass even if the context-row affordance is broken, since the adjacent deletion row would satisfy the [+] check. This doesn't affect the production build, but leaves the new coverage weaker than it appears.

test/pty/ui-integration.test.ts — specifically the hover-fallback logic in the new test block.

Important Files Changed

Filename	Overview
test/pty/ui-integration.test.ts	Adds one new PTY integration test targeting context/unchanged rows in a deletion-only diff; follows the established pattern of nearby tests closely with no structural defects.

Sequence Diagram

sequenceDiagram
    participant T as Test
    participant S as PTY Session (hunk)
    participant H as Harness

    T->>H: createDeletionOnlyFilePair()
    T->>H: launchHunk(diff, split, 120x16)
    H-->>T: session

    T->>S: "waitForText(/keep = true/, 15s)"
    S-->>T: initial snapshot

    T->>T: "lineIndexOf(initial, "keep = true") → targetRow"

    loop revealAddNoteNear(targetRow)
        T->>S: moveMouse(x, y) near targetRow
        S-->>T: waitForText(/[+]/, 200ms)
    end

    T->>S: click(/[+]/)
    S-->>T: waitForText(/Draft note/)

    T->>S: type("Save this context draft.")
    T->>S: press(ctrl+s)
    S-->>T: waitForText(/Your note/)

    T->>T: expect(saved).toContain("Save this context draft.")
    T->>S: session.close()

Loading

Prompt To Fix All With AI

Fix the following 1 code review issue. Work through them one at a time, proposing concise fixes.

---

### Issue 1 of 1
test/pty/ui-integration.test.ts:742-768
**Hover fallback may mask a broken context-row `[+]`**

`revealAddNoteNear` probes the exact `targetRow` first, but if no `[+]` appears there it silently falls through to `targetRow ± 1` — which in this fixture is the `removeMe` deletion row, a row that is already known to surface `[+]` from the previous test. If the context-row affordance were broken (i.e., hovering `keep = true` never shows `[+]`), the helper would succeed against the adjacent deletion row, the test would still pass, and the regression would go undetected. Consider either (a) asserting that the `[+]` was found without falling back, or (b) calling `moveMouse` directly at the `targetRow` and asserting `[+]` appears before proceeding to `session.click`.

_{Reviews (1): Last reviewed commit: "test(pty): cover add-note context rows" | Re-trigger Greptile}

[greptile-apps]

Hover fallback may mask a broken context-row [+]

revealAddNoteNear probes the exact targetRow first, but if no [+] appears there it silently falls through to targetRow ± 1 — which in this fixture is the removeMe deletion row, a row that is already known to surface [+] from the previous test. If the context-row affordance were broken (i.e., hovering keep = true never shows [+]), the helper would succeed against the adjacent deletion row, the test would still pass, and the regression would go undetected. Consider either (a) asserting that the [+] was found without falling back, or (b) calling moveMouse directly at the targetRow and asserting [+] appears before proceeding to session.click.

Prompt To Fix With AI

This is a comment left during a code review.
Path: test/pty/ui-integration.test.ts
Line: 742-768

Comment:
**Hover fallback may mask a broken context-row `[+]`**

`revealAddNoteNear` probes the exact `targetRow` first, but if no `[+]` appears there it silently falls through to `targetRow ± 1` — which in this fixture is the `removeMe` deletion row, a row that is already known to surface `[+]` from the previous test. If the context-row affordance were broken (i.e., hovering `keep = true` never shows `[+]`), the helper would succeed against the adjacent deletion row, the test would still pass, and the regression would go undetected. Consider either (a) asserting that the `[+]` was found without falling back, or (b) calling `moveMouse` directly at the `targetRow` and asserting `[+]` appears before proceeding to `session.click`.

How can I resolve this? If you propose a fix, please make it concise.

[benvinegar]

Good catch. I added a same-row-only helper for this context-row case, so the test can try nearby columns but no longer falls back to adjacent rows and cannot pass by hitting the deletion row. Re-ran the context test 5x plus the add-note PTY subset, typecheck, and lint.

_{Responded by pi using GPT-5.}