- Add double join on translations for source/target languages
- Left join term_glosses for optional source-language glosses
- Filter difficulty on target side only (intentionally asymmetric:
a word's difficulty can differ between languages, and what matters
is the difficulty of the word being learned)
- Return neutral field names (sourceText, targetText, sourceGloss)
instead of quiz semantics; service layer maps to prompt/answer
- Tighten term_glosses unique constraint to (term_id, language_code)
to prevent the left join from multiplying question rows
- Add TODO for ORDER BY RANDOM() scaling post-MVP
- Add GameRequestSchema and derived types to packages/shared
- Add SupportedLanguageCode, SupportedPos, DifficultyLevel type exports
- Add getGameTerms() model to packages/db with pos/language/difficulty/limit filters
- Add prepareGameQuestions() service skeleton in apps/api
- Add createGame controller with Zod safeParse validation
- Wire POST /api/v1/game/start route
- Add scripts/gametest/test-game.ts for manual end-to-end testing
- Add extractors for Italian sources: it_m3.xls and italian.json
- Add comparison script (compare-italian.py) to report source overlaps and conflicts
- Add merge script (merge-italian-json.py) with priority order ['italian', 'it_m3']
- Output authoritative dataset to datafiles/italian-merged.json
- Update README to document both English and Italian pipelines
- add deck_terms to schema imports
- add addTermsToDeck — diffs source term IDs against existing deck_terms,
inserts only new ones, returns count of inserted terms
- add updateValidatedLanguages — recalculates and persists validated_languages
on every run so coverage stays accurate as translation data grows
- wire both functions into main with isNewDeck guard to avoid redundant
validated_languages update on deck creation
- add final summary report
- fix possible undefined on result[0] in createDeck
- tick off remaining roadmap items
Implements packages/db/src/seed.ts — reads all JSON files from
scripts/datafiles/, validates filenames against supported language
codes and POS, and upserts synsets into and
via onConflictDoNothing. Safe to re-run; produces 0 writes on
a duplicate run.
- terms, translations, term_glosses with cascade deletes and pos check constraint
- language_pairs with source/target language check constraints and no-self-pair guard
- users with openauth_sub as identity provider key
- decks and deck_terms with composite PK and position ordering
- indexes on all hot query paths (distractor generation, deck lookups, FK joins)
- SUPPORTED_POS and SUPPORTED_LANGUAGE_CODES as single source of truth in @glossa/shared
Database Schema:
- Add decks table for curated word lists (A1, Most Common, etc.)
- Add deck_terms join table with position ordering
- Link rooms to decks via rooms.deck_id FK
- Remove frequency_rank from terms (now deck-scoped)
- Change users.id to uuid, add openauth_sub for auth mapping
- Add room_players.left_at for disconnect tracking
- Add rooms.updated_at for stale room recovery
- Add CHECK constraints for data integrity (pos, status, etc.)
Extraction Script:
- Rewrite extract.py to mirror complete OMW dataset
- Extract all 25,204 bilingual noun synsets (en-it)
- Remove frequency filtering and block lists
- Output all lemmas per synset for full synonym support
- Seed data now uncurated; decks handle selection
Architecture:
- Separate concerns: raw OMW data in DB, curation in decks
- Enables user-created decks and multiple difficulty levels
- Rooms select vocabulary by choosing a deck
- Configure PostgreSQL 18 and Valkey 9.1 services
- Create multi-stage Dockerfiles for API and Web apps
- Set up pnpm workspace support in container builds
- Configure hot reload via volume mounts for both services
- Add healthchecks for service orchestration
- Support dev/production stage targets (tsx watch vs compiled)
