marcel/familienarchiv
1
0
Fork 0
Code Issues 105 Pull Requests 1 Actions Packages Projects Releases Wiki Activity

test(transcription): restore strong one-fetch regression guard

Browse Source 
Sara on PR #629 round 3: the round-2 fix captured the fetch count AFTER
typing '@', so a regression that re-introduced the legacy per-keystroke
items() callback would have its '@'-keystroke fetch silently absorbed
into the baseline. Drop the baseline subtraction and count every
/api/persons fetch since render — typing '@' + fill('Walter') must
total exactly one fetch.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
This commit is contained in:
Marcel
2026-05-20 00:01:54 +02:00
committed by marcel
parent 9886f2bcac
commit 4f469db02e
1 changed files with 16 additions and 10 deletions
Download Patch File Download Diff File Expand all files Collapse all files

26
frontend/src/lib/shared/discussion/PersonMentionEditor.svelte.spec.ts
View File

@@ -219,30 +219,36 @@ describe('PersonMentionEditor — AC-2/3: search input drives fetch', () => {
expect(fetchesAfterSearch).toBe(1);
});
it('fires exactly one /api/persons fetch when the user searches for Walter (debounced)', async () => {
it('fires exactly one /api/persons fetch when the user searches for Walter (regression guard)', async () => {
// Regression guard: a previous version of PersonMentionEditor had a
// duplicated `items()` callback in the Tiptap suggestion config that
// fetched per-keystroke in addition to the debounced search-input fetch
// (Markus & Felix round-1). To catch that regression, we must NOT
// subtract any baseline — every fetch from render onwards counts.
// Sara on PR #629 round 3.
const fetchMock = vi
.fn()
.mockResolvedValue({ ok: true, json: vi.fn().mockResolvedValue([AUGUSTE]) });
vi.stubGlobal('fetch', fetchMock);
renderHost();
// Open the dropdown first so the search input is reachable. `fill` then
// drives the searchbox in one input event — sidesteps per-keystroke
// debounce timing on CI that Sara flagged on PR #629 round 2.
// Open the dropdown, then drive the search input via fill() — sidesteps
// per-keystroke timing of userEvent.type that Sara flagged round 2.
await userEvent.type(page.getByRole('textbox'), '@');
await vi.waitFor(async () => {
await expect.element(page.getByRole('searchbox')).toBeVisible();
});
const fetchesBeforeSearch = fetchMock.mock.calls.length;
await page.getByRole('searchbox').fill('Walter');
await new Promise((r) => setTimeout(r, SEARCH_DEBOUNCE_MS + POST_DEBOUNCE_SLACK_MS));
const personsFetches = fetchMock.mock.calls
.slice(fetchesBeforeSearch)
.filter(([url]) => typeof url === 'string' && url.startsWith('/api/persons'));
// No baseline subtraction — count ALL /api/persons fetches since render.
// If the legacy per-keystroke items() callback returns, typing `@` alone
// would already produce one fetch and `fill('Walter')` another, breaking
// this assertion.
const personsFetches = fetchMock.mock.calls.filter(
([url]) => typeof url === 'string' && url.startsWith('/api/persons')
);
expect(personsFetches.length).toBe(1);
});