Filters are evaluated once into an id set used by both search branches.
Adds tests for age/genre/mechanic filters with a search term, rank 2 vs 3
and null p_limit. approxNormalize uses NFKD. Perf script checks confirm_local
value and unix socket and re-analyzes after rollback. Adds
docs/specs/3-katalog-suche-umsetzung.md (measurements, log check incl. GET
finding), release checklist section 2a and the CSV-only note for alt titles.
Refs #3
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Adds the unchanged runbook queries (rolconfig, pg_available_extensions,
pg_settings) and asserts the settings that keep search terms out of logs.
katalog-pflege 8.2 now states that the three log settings depend on each other.
Refs #3
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
supabase/perf/catalog_search_perf.sql loads synthetic data in one rolled-back
transaction and measures search_catalog_games as anon with RLS (explain and
p50/p95/p99 per case). Lives outside supabase/tests so it never runs in CI.
Refs #3
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Rank 4 and the genre/mechanic filters ran a correlated subplan per game
(13 s at 10 000 games). Now: one set-based word coverage, filters as one hashed
IN set, 'like all' patterns, similarity only for hits. Adds pgTAP cases found
by mutation testing (paging by relevance, alt title choice, duplicate words,
mechanic duplicates) and a Deno case for mark characters.
Refs #3
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Parses de:Title;en:Title (';' forbidden in titles, errors with line number),
generates an idempotent reconcile for game_alt_titles, adds an info row to
checks.sql and attaches alternative titles to Testspiel B. Regenerated files.
Refs #3
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Stand von 2c (PR #57) und Fremdschlüssel-Wächter in 07 und 08,
normalize_game_title ausgeschrieben, Release-Checkliste als offener
PR #54 gekennzeichnet, Änderungsprotokoll nur ergänzt.
Refs #3
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Kandidatenschlüssel über title_search, gemeinsame Normalisierung
wörtlich aus Spec #3, pg_trgm aus 3a, Abhängigkeiten 3a/3b,
Spec-F6 korrigiert (nur 2c-Nachtrag, vor Merge von PR #57).
Refs #13
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
search_normalize für alle API-Rollen ausführbar, Symbole vor unaccent
entfernen, eine Basisfunktion gemeinsam mit #13, Reihenfolge 3a nach 2b,
Spec-F4 mit vorläufiger Messung ohne Staging, Indizes nicht als
Begründung, Grenzen und Hinweise aus dem Review.
Refs #3
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Tageslimit-Rechte (report_day), Normalisierung als Wrapper um
search_normalize aus Spec #3, Sperrreihenfolge und Rechte von
adopt_report_target, Indizes aus 2c, Testdatei 10, Edition-Hinweis
mit editions_cleared als Empfehlung.
Refs #13
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Existing team KSX6GJY9C5 (C1), defer CODE_SIGN_ENTITLEMENTS until
the team is confirmed, local mobile builds mandatory, email_change
needs KTO-11, verified facts separated from spike assumptions,
QUE-06 keyboard criterion, T2 depends on PR A, limited interception
protection on Android < 12 (R13, F11), login CSRF accepted (R14),
supersession of spec 1 E4 recorded as task W2.
Refs #49
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Idempotenz der alternativen Titel in pgTAP statt Deno, Auswirkung auf
06/00_schema, alternative Titel an Testspiel B, Merge-Reihenfolge
2b -> 2c -> 3a und Erweiterung des Fremdschlüssel-Wächters in 07.
Refs #3
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Spec for issue #49: sign-in prompt on the public home screen and
Universal Links / App Links for /auth/confirm on iOS and Android,
web fallback unchanged. Includes options, test strategy, task split
and tasks only Chris can do (Apple/Google accounts, domain files,
SITE_URL on prod).
Refs #49
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Replace tests that could not fail (manual_fields stability, no-op update), select runbook games by id, add grant matrix and forged-admin tests, assert vocabulary seeding and Deno checksum in CI, clarify the set local convention and API notes.
Refs #2
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Adds the checks.sql smoke step, the verified CLI result for generated seed content and the created_at/updated_at note for the API.
Refs #2
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Includes the checked CLI assumptions (sql_paths, begin/commit) and implementation notes in the spec.
Refs #2
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Deno script validates vocabulary.csv and startbestand.csv and generates the idempotent SQL, key list and checks.sql. startbestand.csv stays header-only; vocabulary.csv is a proposal pending approval (C0). seed.sql no longer inserts genres and mechanics.
Refs #2
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Sources, publishers, hidden/merged columns, manual_fields triggers, catalog_games view and column grants on games (KAT-05, PR 2a). Fixtures for the tests are generated from supabase/catalog/testdata.
Refs #2
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
