All 30+ journey_* message keys added to de/en/es.json. Four new ErrorCode
values for journey item operations wired into errors.ts + getErrorMessage().
Interlude CSS primitives (--c-interlude-bg/border/label) defined for light
and dark themes so JourneyItemRow can reference them via semantic aliases.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Moves Status + Persons sections into a shared component so both
GeschichteEditor (STORY) and the upcoming JourneyEditor (JOURNEY) can
use the same sidebar without duplicating markup. Adds <details> mobile
collapsibles with 44px summary hit areas.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Removes the hard-typed TranscriptionBlockData constraint so JourneyEditor
can reuse the pointer-drag module without importing transcription types.
Selector contract (data-block-wrapper / data-drag-handle) unchanged.
Adds type-regression guard test verified via tsc --noEmit.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Drop stale JourneyItem/JourneyItemCreateDTO schemas — removed in base
branch when api.ts was regenerated; neither type is referenced in
frontend code (JourneyItemView is the read model used instead).
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Importing layout.css in test-setup.ts activated Tailwind's responsive
breakpoint classes (hidden lg:flex, hidden md:block, etc.), making
42 elements invisible at the default narrow Playwright test viewport.
Revert the CSS import. Instead, add inline style attributes to the three
components whose tests measure computed properties (min-height, font-size)
— these values match what the Tailwind classes produce, so the real app
appearance is unchanged.
Also fix goto mock leakage in the geschichten/[id] delete-failure test:
the delete-success test's goto('/geschichten') call was not cleared before
the failure test ran. Add beforeEach(vi.clearAllMocks) to reset mock state.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- Import layout.css in test-setup so Tailwind utilities (text-xs,
min-h-[44px]) apply in vitest-browser — fixes computed-style assertions
for badge font-size and touch-target height
- radioGroupNav: write aria-checked directly on radio buttons on arrow-key
navigation, not only via the optional onChangeFn callback
- DashboardNeedsMetadata spec: tighten footer-link matcher from /50/ to
/Alle 50/ — avoids strict-mode collision with row link whose relative
time text also contains "50" (uploadedAt is exactly 50 days ago today)
- geschichten/[id] page spec: add missing await on userEvent.click before
confirmService.settle() in both delete tests
- TypeSelector spec: replace storyCard.focus() (not on vitest-browser
Locator) with userEvent.click(); force-dispatch aria-disabled Weiter
click via element.click() to bypass Playwright actionability check
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
The inline publishedAt $derived.by() duplicated the exact logic that
formatPublishedAt() in utils.ts encapsulates. Replace it with the
shared helper and drop the now-unused formatDate import.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Without a landmark or widget role, aria-label on a generic <div> is
silently ignored by most screen readers (ARIA spec). Adding role="note"
gives the element an ARIA role that accepts an accessible name, making
the interlude label actually announced.
Also adds a test asserting role="note" and the matching aria-label are
both present on the same element.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Mirrors the getBoundingClientRect pattern from JourneyItemCard.svelte.spec.ts.
Tests actual rendered height rather than presence of a CSS class string.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
The action was writing aria-checked directly and then firing onChange,
which also triggered Svelte's own aria-checked={selected === type} binding.
Double-ownership: action now only calls focus() + onChange(value);
Svelte owns the attribute update.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Function names already communicate intent. Comments that restate the
function name add noise without explaining why.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
JourneyReader filters items to only those where document != null before
passing them here — the ! assertion is valid by caller invariant.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
CSS class string assertion was fragile — class names can change without
breaking the actual layout. DOM measurement via getBoundingClientRect is the
correct way to verify computed height meets WCAG 2.2 minimum.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
13 tests covering null/undefined inputs, partial names, email fallback,
and TZ-safe date slicing for formatPublishedAt.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Moves the confirm-then-delete flow out of StoryReader and JourneyReader into
the single [id]/+page.svelte owner. Both reader components gain an optional
ondelete prop — the delete button calls ondelete?.() so the handler is opt-in
and never duplicated. Tests verify the prop is called on click.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Replaces the 3-line inline join with the shared formatAuthorName helper from
utils.ts. Test switches from CSS class string assertion to getComputedStyle
for the badge font-size check.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Replaces aria-label="Kuratorennotiz" with m.journey_interlude_aria_label()
so screen readers get the correct label in all three supported locales.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Self-check: GeschichteView.items present; type emitted as 'STORY'|'JOURNEY' union literal.
List endpoint returns GeschichteSummary[]; detail endpoint returns GeschichteView.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Adds JOURNEY_ITEM_NOT_FOUND and JOURNEY_ITEM_POSITION_CONFLICT to the frontend
ErrorCode union and getErrorMessage() switch. Adds de/en/es translations.
Regenerates api.ts from the current OpenAPI spec (needs a second run once the
backend is restarted with the new endpoints compiled in).
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- GeschichtenCard.svelte: use GeschichteSummary instead of Geschichte
(list endpoint returns summaries; no items/createdAt/updatedAt needed)
- GeschichtenCard.svelte.test.ts: factory returns GeschichteSummary with
lean author shape; drop Geschichte-only fields (createdAt, groups, etc.)
- geschichten/[id]/+page.svelte: add focus:outline-none focus-visible:ring-2
focus-visible:ring-focus-ring to journey item document links (WCAG 2.4.7)
- page.svelte.test.ts ([id]): replace stale documents[] factory field with
items[]; test now checks placeholder text + note caption
- page.svelte.test.ts (new): remove removed initialDocuments from baseData;
rename test to reflect that only initialPersons is passed through
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- GeschichteEditor.svelte.spec.ts: remove docFactory + initialDocuments test;
rename documentIds test to personIds-only; add familyMember+provisional to
personFactory (were pre-existing omissions)
- GeschichtenCard.svelte.spec.ts: add type:'STORY', replace documents:[] with
items:[], change body null→undefined to match Geschichte schema
- GeschichtenCard.svelte.test.ts: add status/type/createdAt/updatedAt to factory;
cast result as Geschichte to avoid spread-widening type inference
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- api.ts: add GeschichteType, JourneyItem, GeschichteSummary schemas;
remove documentId param from list endpoint; change list response to
GeschichteSummary[]; add type + items to Geschichte; remove documents field
- GeschichteEditor: remove DocumentMultiSelect + documentIds from payload
(journey items are managed via the future Lesereisen editor, not here)
- GET /geschichten page: remove documentId filter from server load + URL logic
- geschichten/new: remove documentId pre-population from server load
- geschichten/[id]: replace g.documents with g.items (document-backed JourneyItems)
- geschichten/new + [id]/edit: remove documentIds from submit payload type
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Manual update since Docker compose backend runs old build; regenerate with
npm run generate:api once new backend is deployed.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Strengthen one renderTranscriptionBody case into the AC-6 contract: a
@DisplayName with an empty mentionedPersons array (the deleted-person case
V71 produces) must render as plain readable text with no <a>, person-mention
class, data-person-id, or href. Guards against a future renderer refactor
silently reintroducing the dead-link-on-deleted-person degradation.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The #718 keyboard-tab-order test hardcoded the visual order
['Eugenie','Walter','Clara','Hans'] on the assumption that buildLayout
sorts each generation alphabetically. #724 replaced that with the
tidy-tree layout, which orders a couple's run by structural ownership
(earliest birth year, then a deterministic id tie-break) — so Walter
(id …a1) now owns the run and Eugenie renders to his right.
Both PRs were green independently; the stale assertion only surfaced
once #718 and #724 landed together on main. Correct the expected reading
order to ['Walter','Eugenie','Clara','Hans'] and refresh the now-wrong
'alphabetical' comment. The companion self-validating test (DOM order ==
sorted by y,x) already guarded the real property, so only the hardcoded
assertion needed updating.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Review follow-up (Sara, fast-follow): the t no-active-region guard and the
draw-cue arm/disarm rule lived inline in the page with no direct coverage.
Extracted to pure resolveTrainingMark() (no-op when no region; recognition
enrolled flip) and canArmDraw()/shouldDisarmDraw(), each with unit tests
(10 cases total). The page now arms the draw cue only via canArmDraw and
disarms via shouldDisarmDraw, and routes t through resolveTrainingMark.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Review follow-up (Leonie, Felix, Markus): bump cheatsheet key caps to text-sm
for the 60+ audience, add a focus-visible ring to the close button, simplify
the draw-hint guard to {#if drawArmed} (the $effect already clears it outside
edit mode), and document why the transcribeShortcuts action ignores its node
and binds to window.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Review follow-up (Sara): the prior single-owner evidence was two separate
unit facts against an inert DOM stub. This renders a real AnnotationShape,
attaches the live transcribeShortcuts action, focuses the region, and presses
Delete once — asserting deleteCurrentRegion fires exactly once. A genuine
integration guard against re-introducing a double-bind.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Review follow-up (Sara): j/k wrap-around and fresh-entry had no direct
coverage — the logic lived inline in the page where the action spec only
mocks the callbacks. Extracted to a pure stepRegion() with 9 unit tests
(empty list, forward/back, both wraps, fresh-entry null + unknown id,
length-1). Also replaces the inline nested ternary Felix flagged.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Review follow-up (Requirements Engineer, Leonie) — closes the unmet
acceptance row. The coach card's "press ?" tip rendered unconditionally, so
a touch-only tablet transcriber (no hardware keyboard) was told to press a
key they don't have. The hint is now gated behind a fine-pointer media
query ([@media(pointer:coarse)]:hidden); the cheatsheet itself only opens
via the "?" key, so it already never surfaces without a keyboard. Also bumps
the key cap from 11px to text-xs for the 60+ audience.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Review follow-up (Leonie, Requirements Engineer): the Delete key cap was a
hardcoded German "Entf" shown to EN/ES users — now driven by key_cap_delete
(Entf/Del/Supr). The annotation read-only aria-label was a hardcoded German
"Block anzeigen" in all locales — now annotation_view_label. Renamed the Esc
row label from "Bereich schließen" to "Panel schließen" so it no longer
collides with "Bereich" (= region) used elsewhere in the cheatsheet.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Adds a secondary keyboard hint to the existing coach footer row pointing
transcribers at the "?" cheatsheet, with a semantic <kbd>. Cross-references
the shortcuts introduced for the empty-state coach (#320).
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Attaches the transcribeShortcuts action to the document page and wires every
command to existing context setters: j/k walk the sortOrder-sorted regions
and set activeAnnotationId, e toggles read/edit, n arms a draw cue (edit
only), Delete routes to the existing confirm path, ? opens the cheatsheet,
and Esc is now owned solely by the action — the inline onMount Esc listener
is removed (decision B1). Renders ShortcutCheatsheet and a draw-armed hint.
"t" toggles the document-level KURRENT_RECOGNITION training enrollment (the
only training surface that exists; there is no per-region flag yet — see
#321) and no-ops unless a region is active. Also reconciles annotation
Delete: the shape no longer self-handles the key, with onfocus syncing the
active region so the action deletes exactly once.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Native <dialog aria-modal> cheatsheet: showModal()/close() bridge, close
button focused on open, eight grouped <kbd> rows (nav/edit/utility), an
autosave footer line, and a reduced-motion-guarded fade. Closes on Esc,
backdrop click, and the close button; "?" while open is a no-op. Adds the
shortcut_close_panel i18n key. 8 component tests.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Single-owner window keydown action for the Transcribe panel: j/k region
nav, e mode toggle, n draw (edit only), t training mark, Delete, ? cheat-
sheet, and the Esc precedence ladder (cheatsheet → editable no-op → close
panel). Pure input-to-callback translator with a focus guard that exempts
only "?"; removes its listener on destroy. 20 unit tests cover every key,
the panel/focus guards, the Esc matrix, and teardown.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Bumps the title helper from text-xs (12px) to text-sm (14px) for the 60+ audience (FR-005
prefers a larger size than the field hints) and tightens the component test to assert the
actual localized string and the 14px class — addresses Leonie's and Sara's review notes.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Adds the FR-TITLE-005 helper line under the title input in DescriptionSection, shown only
on the single-document edit form via a new showTitleHelp prop (off for the new-document and
bulk-edit forms). It is wired to the input with aria-describedby and uses text-ink-3 (WCAG AA
on bg-surface). New Paraglide key form_helper_title_autogenerated in de/en/es. Adds a
component test for the helper + aria wiring and an end-to-end pass: create an auto-titled doc,
edit its date, and see the title follow on the detail page.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Review follow-ups:
- Leonie/UX: 0.55 navy on the sand canvas was ~2.6:1, under the WCAG 1.4.11 3:1
non-text floor for senior readers; 0.7 clears it.
- Sara/QA: add a browser test that actually renders a cross-level link and
asserts the distinct 2 6 dash, and that a non-cross-link parent edge stays
solid — the cadence was previously only validated via the structural
crossLinks array, never where it renders.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Review follow-up (Sara/QA): the empty graph (fresh /stammbaum before data loads)
exercised the positions.size===0 viewBox fallback and the roots.length===0 early
return, both previously untested. Assert no NaN in the viewBox and MIN dimensions,
plus a single isolated node placed once at rank 0.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
