forgejo-lila/lila
1
0
Fork 0
Code Issues 50 Pull requests Projects 1 Releases Packages Wiki Activity Actions

updating docs

Browse source
This commit is contained in:
lila 2026-07-23 14:15:33 +02:00
parent 88b16a1ed7
commit 61fa032d6c
2 changed files with 1321 additions and 0 deletions
Download patch file Download diff file Expand all files Collapse all files

704
documentation/pipeline/roadmap.md Normal file
View file

@ -0,0 +1,704 @@
# Vocabulary Trainer — Roadmap: Data Pipeline & Schema Migration
> **Objective:** Replace the OpenWordNet/kaikki data source with a
> Gemini-powered pipeline that generates high-quality vocabulary data,
> backed by a normalized Postgres schema.
>
> **Author:** [Your Name]
> **Date:** July 2026
> **Companion doc:** `docs/schema-design.md`
---
## How to Read This Document
This roadmap has three zoom levels:
- **Part 1 — Overview:** The phases at a glance. Read this to understand the full scope in 30 seconds.
- **Part 2 — Phase Breakdown:** Goals, tasks, dependencies, and acceptance criteria per phase. Read this to plan your week.
- **Part 3 — Detailed Tasks:** Step-by-step instructions within each phase. Read this when you sit down to code.
---
# Part 1 — High-Level Overview
```
Phase 0 Preparation
Get wordlists, set up tooling, finalize the Gemini prompt.
Phase 1 Data Pipeline
Build the Gemini → validate → SQLite pipeline.
Produce a clean dataset of ~1000 nouns × 5 languages.
Phase 2 Schema & Migration
Create the new Drizzle schema (words, senses, translations).
Write the SQLite → Postgres import script.
Migrate the local dev database.
Phase 3 App Integration
Rewrite the game and distractor queries against the new schema.
Update the exercise-generation logic.
Test the full game flow in dev.
Phase 4 Production Deploy
Run the Drizzle migration on prod.
Import the dataset.
Verify the live app works end-to-end.
Phase 5 Extend POS
Run the pipeline for verbs, adjectives, adverbs.
No schema changes needed — new wordlists + adjusted prompts.
Phase 6 Future Features (out of scope for now)
Inflection tables, conjugation/declension exercises,
gender exercises, spaced-repetition scheduling.
```
**Dependency chain:**
```
Phase 0 → Phase 1 → Phase 2 → Phase 3 → Phase 4 → Phase 5
│
▼
Phase 6
```
Each phase depends on the previous one being complete. Do not skip
ahead.
---
# Part 2 — Phase Breakdown
---
## Phase 0: Preparation
**Goal:** Have everything you need before writing pipeline code.
**Tasks:**
- [ ] Acquire frequency-based noun lists for all 5 languages
- [ ] Clean and format the lists (one word per line, UTF-8)
- [ ] Set up local Postgres (Docker or native)
- [ ] Set up a SQLite database file for staging
- [ ] Write and test the Gemini prompt with 5 sample words
- [ ] Refine the prompt until the JSON output matches the contract
defined in `docs/schema-design.md` §6.3
**Dependencies:** None.
**Acceptance criteria:**
- You have 5 wordlist files in `data/wordlists/` (one per language).
- A Gemini call with 5 German nouns returns valid JSON matching the
contract, including definitions, examples, translations with gender,
and difficulty levels.
- Local Postgres is running and reachable from your app.
---
## Phase 1: Data Pipeline
**Goal:** A repeatable script that takes a wordlist, calls Gemini in
batches of 20, validates the output, and writes clean rows to SQLite.
**Tasks:**
- [ ] Write the validation module (see schema-design §6.4)
- [ ] Write the pipeline script: - Read wordlist file - Split into batches of 20 - Call Gemini API per batch - Parse JSON response - Validate each entry - Write valid entries to SQLite - Log invalid entries to a rejection file
- [ ] Create the SQLite schema (mirrors the Postgres schema)
- [ ] Run the pipeline for all 5 languages (nouns only)
- [ ] Review the rejection log, fix prompt issues, re-run failed batches
- [ ] Spot-check 50 random entries for correctness
**Dependencies:** Phase 0 complete.
**Acceptance criteria:**
- SQLite database contains ~1000 nouns × 5 languages with senses and
translations.
- Rejection rate is below 10%.
- Spot-checked entries have correct definitions, plausible examples,
correct genders, and reasonable difficulty levels.
- The pipeline is re-runnable (idempotent or with duplicate handling).
---
## Phase 2: Schema & Migration
**Goal:** The new normalized schema exists in Drizzle, the local
Postgres database is migrated, and the SQLite data is imported.
**Tasks:**
- [ ] Write the new Drizzle schema file: - `words` table (headword, language_code, pos) - `senses` table (word_id FK, sense_index, difficulty,
cefr_level, definitions TEXT[], examples TEXT[]) - `translations` table (sense_id FK, target_language_code,
translation, gender, difficulty) - All CHECK constraints, UNIQUE constraints, indexes
- [ ] Generate the Drizzle migration (`npx drizzle-kit generate`)
- [ ] Inspect the generated SQL file — verify it creates the right
tables, constraints, and indexes
- [ ] Apply the migration to local Postgres (`npx drizzle-kit migrate`)
- [ ] Write the import script (SQLite → Postgres): - Read all rows from SQLite - Insert into Postgres in dependency order:
words → senses → translations - Use batch inserts (not row-by-row) - Wrap in transactions (per batch of 20 words) - Handle duplicates gracefully (skip or upsert)
- [ ] Run the import script
- [ ] Verify row counts match between SQLite and Postgres
- [ ] Run 3–5 manual SQL queries against Postgres to sanity-check
the data
**Dependencies:** Phase 1 complete (SQLite has data).
**Acceptance criteria:**
- `npx drizzle-kit migrate` runs without errors on local Postgres.
- Import script completes without errors.
- Row counts in Postgres match SQLite (±rejection count).
- Manual query: "Give me 5 random German nouns with Spanish
translations at easy difficulty" returns sensible results.
---
## Phase 3: App Integration
**Goal:** The running app uses the new schema. Game rounds and
distractors work correctly for all language pairs.
**Tasks:**
- [ ] Rewrite `getGameTerms` query: - JOIN words → senses → translations - Filter: source language, pos, sense difficulty (ceiling),
target language, translation difficulty (exact) - ORDER BY RANDOM(), LIMIT rounds
- [ ] Rewrite `getDistractors` query: - Same JOINs and filters - Exclude: `sense_id != current`, `translation != correct` - ORDER BY RANDOM(), LIMIT 3
- [ ] Update the exercise-generation logic: - Pick one random definition from the `definitions` array - Pick one random example from the `examples` array - Assemble the 4 answer options (1 correct + 3 distractors) - Shuffle the options
- [ ] Remove or deprecate old schema references
(old `vocabulary_entries`, `entry_translations` tables)
- [ ] Test manually: - German → Spanish, nouns, easy, 10 rounds - Spanish → German, nouns, medium, 10 rounds - English → French, nouns, hard, 10 rounds - Italian → German, nouns, easy, 10 rounds - Verify: no duplicate answers, no same-sense synonyms as
distractors, definitions and examples are in the source
language, genders display correctly
- [ ] Test edge cases: - A difficulty/pos/language combo with very few words
(does the app handle < 4 available words gracefully?) - A word with multiple senses (does the correct sense appear?)
**Dependencies:** Phase 2 complete (Postgres has data, schema exists).
**Acceptance criteria:**
- Full game flow works in dev for at least 4 different language pairs.
- No same-sense synonyms appear as distractors.
- Definitions and examples are in the correct language.
- Gender is displayed where applicable.
- No console errors or unhandled query failures.
---
## Phase 4: Production Deploy
**Goal:** The live app runs on the new schema with the new data.
**Tasks:**
- [ ] Back up the production database
- [ ] Run the Drizzle migration on prod
- [ ] Run the import script against prod Postgres
- [ ] Verify row counts on prod
- [ ] Test the live app: - Play 2 full games on the deployed app - Check different language pairs and difficulties
- [ ] Monitor for errors (server logs, browser console) for 24h
- [ ] Remove old tables from prod (after confirming everything works)
**Dependencies:** Phase 3 complete (dev is fully working).
**Acceptance criteria:**
- Live app serves game rounds from the new schema.
- No errors in server logs for 24 hours.
- Old tables are dropped (or scheduled for removal).
- Rollback plan exists: the prod backup can be restored if needed.
---
## Phase 5: Extend POS
**Goal:** Verbs, adjectives, and adverbs are in the database and
usable in the app.
**Tasks:**
- [ ] Acquire frequency lists for verbs, adjectives, adverbs
(all 5 languages)
- [ ] Adjust the Gemini prompt per POS: - Verbs: may need different metadata (transitivity, etc.) - Adjectives: may need base form info - Adverbs: typically simpler metadata
- [ ] Run the pipeline for each POS
- [ ] Validate, import to Postgres (dev → prod)
- [ ] Test game flow with verbs, adjectives, adverbs
- [ ] Verify the POS filter in the app UI works for all types
**Dependencies:** Phase 4 complete.
**Acceptance criteria:**
- All 4 POS types are playable in the app.
- Pipeline is repeatable for future data additions.
---
## Phase 6: Future Features (out of scope)
Listed here for visibility. Not planned, not estimated.
- [ ] `inflection_forms` table + conjugation exercises (verbs)
- [ ] Adjective declension exercises (der grüne Mann, grüner Mann…)
- [ ] Gender exercises (pick the correct article)
- [ ] Spaced-repetition scheduling (track which words the user knows)
- [ ] User accounts and progress persistence
- [ ] Additional languages (if ever)
---
# Part 3 — Detailed Task Breakdown
---
## Phase 0: Preparation — Detailed
### 0.1 Acquire wordlists
- Search for frequency lists. Good starting points:
- "Leipzig Corpora Collection" (academic, per-language)
- Wiktionary frequency lists
- GitHub repos: search "german noun frequency list",
"spanish noun frequency list", etc.
- Tatoeba sentence counts as a proxy for word frequency
- Target: ~1000 nouns per language, sorted by frequency.
- Format: plain text, one word per line, UTF-8, no headers.
- Save to `data/wordlists/nouns_de.txt`, `nouns_es.txt`, etc.
- Clean the lists: remove duplicates, remove words with spaces
(multi-word expressions), remove proper nouns if desired.
### 0.2 Set up local Postgres
- Option A (Docker):
```
docker run --name vocab-dev \
-e POSTGRES_USER=dev \
-e POSTGRES_PASSWORD=dev \
-e POSTGRES_DB=vocab \
-p 5432:5432 \
-d postgres:16
```
- Option B (native install): install Postgres, create a `vocab`
database.
- Update your `.env` / `.env.local` with the connection string.
- Verify: connect with `psql` or a GUI client (TablePlus, DBeaver,
pgAdmin). Run `SELECT 1;`.
### 0.3 Write and test the Gemini prompt
- Start with 5 German nouns.
- The prompt should specify:
- The exact JSON structure expected (copy from schema-design §6.3)
- That definitions and examples must be in the word's language
- That translations are needed for all 4 other supported languages
- That gender must be provided for de/fr/es/it, null for en
- That difficulty must be one of: easy, medium, hard
- That CEFR level must be one of: A1, A2, B1, B2, C1, C2
- That multiple senses should be included for polysemous words
- Call the API. Inspect the JSON. Common issues to fix:
- Gemini wraps the JSON in markdown code fences → strip them
- Gemini returns "intermediate" instead of "medium" → normalize
- Gemini omits a language → re-prompt or reject
- Gemini returns gender "common" for German → reject
- Iterate until 3 consecutive batches of 20 return clean JSON.
---
## Phase 1: Data Pipeline — Detailed
### 1.1 Create the SQLite schema
- Create a file `pipeline/schema.sql` or define it in your pipeline
script.
- Tables mirror the Postgres schema:
```sql
CREATE TABLE words (
id TEXT PRIMARY KEY,
headword TEXT NOT NULL,
language_code TEXT NOT NULL,
pos TEXT NOT NULL,
created_at TEXT DEFAULT (datetime('now')),
UNIQUE(headword, language_code, pos)
);
CREATE TABLE senses (
id TEXT PRIMARY KEY,
word_id TEXT NOT NULL REFERENCES words(id),
sense_index INTEGER NOT NULL DEFAULT 0,
difficulty TEXT NOT NULL,
cefr_level TEXT,
definitions TEXT NOT NULL DEFAULT '[]', -- JSON array as text
examples TEXT NOT NULL DEFAULT '[]', -- JSON array as text
created_at TEXT DEFAULT (datetime('now')),
UNIQUE(word_id, sense_index)
);
CREATE TABLE translations (
id TEXT PRIMARY KEY,
sense_id TEXT NOT NULL REFERENCES senses(id),
target_language_code TEXT NOT NULL,
translation TEXT NOT NULL,
gender TEXT,
difficulty TEXT NOT NULL,
created_at TEXT DEFAULT (datetime('now')),
UNIQUE(sense_id, target_language_code, translation)
);
```
- Note: SQLite has no native TEXT[]. Store arrays as JSON text.
The import script will parse them into Postgres TEXT[] on import.
### 1.2 Write the validation module
- Create `pipeline/validate.ts` (or .js).
- Input: one parsed Gemini entry (the JSON object for one word).
- Checks (return a list of errors, empty = valid):
- `headword` is a non-empty string
- `language` is in ['en','de','it','fr','es']
- `pos` is in ['noun','verb','adjective','adverb']
- `senses` is a non-empty array
- Each sense has:
- `sense_index` is a non-negative integer
- `difficulty` is in ['easy','medium','hard']
- `cefr_level` is in ['A1','A2','B1','B2','C1','C2'] or null
- CEFR ↔ difficulty mapping is consistent
- `definitions` is a non-empty array of non-empty strings
- `examples` is a non-empty array of non-empty strings
- `translations` is a non-empty array
- Each translation has:
- `target_language` is in the supported list
- `target_language` != the word's own language
- `word` is a non-empty string
- `gender` is valid for the target language
(de: masculine/feminine/neuter/null,
fr/es/it: masculine/feminine/null,
en: null)
- `difficulty` is in ['easy','medium','hard']
- Output: `{ valid: boolean, errors: string[] }`
### 1.3 Write the pipeline script
- Create `pipeline/run.ts`.
- Pseudocode:
```
for each language in [de, en, es, fr, it]:
read wordlist file → array of words
split into batches of 20
for each batch:
call Gemini API with the batch
parse JSON response (strip markdown fences if present)
for each entry in response:
validate(entry)
if valid:
generate UUIDs for word, senses, translations
INSERT into SQLite (words, senses, translations)
else:
append to rejection log (data/rejections/{lang}_{pos}.jsonl)
log progress: "Batch 12/50 done. 238 words imported, 2 rejected."
sleep 1s (rate limiting)
```
- Use `better-sqlite3` for SQLite access (synchronous, simple).
- Generate UUIDs with `crypto.randomUUID()`.
- Store definitions/examples as JSON strings in SQLite
(`JSON.stringify(arr)`).
### 1.4 Run and review
- Run the pipeline for all 5 languages.
- Check the rejection log. If rejection rate > 10%, fix the prompt
and re-run failed batches.
- Spot-check: open the SQLite DB, run:
```sql
SELECT w.headword, s.definitions, t.translation, t.gender
FROM words w
JOIN senses s ON s.word_id = w.id
JOIN translations t ON t.sense_id = s.id
WHERE w.language_code = 'de' AND w.pos = 'noun'
ORDER BY RANDOM()
LIMIT 20;
```
- Read the definitions. Are they in German? Do they make sense?
Are the genders correct? Are the difficulties reasonable?
---
## Phase 2: Schema & Migration — Detailed
### 2.1 Write the Drizzle schema
- Create or update `src/db/schema.ts`.
- Define the three tables using Drizzle's `pgTable` builder.
- Include all columns, types, constraints, and indexes as specified
in `docs/schema-design.md` §3.
- Keep the old tables in the file for now (commented out or in a
separate file) so the app doesn't break during development.
### 2.2 Generate and inspect the migration
- Run: `npx drizzle-kit generate`
- Open the generated SQL file in `drizzle/` (or wherever your
config puts it).
- Read it. Verify:
- Three CREATE TABLE statements
- CHECK constraints match your schema
- UNIQUE constraints are present
- Three CREATE INDEX statements
- Foreign keys reference the correct tables with ON DELETE CASCADE
- If something looks wrong, fix the schema file and regenerate.
### 2.3 Apply the migration locally
- Run: `npx drizzle-kit migrate`
- Connect to local Postgres and verify:
```sql
\dt -- list tables
\d words -- describe words table
\d senses -- describe senses table
\d translations -- describe translations table
```
### 2.4 Write the import script
- Create `scripts/import-to-postgres.ts`.
- Pseudocode:
```
open SQLite database (read-only)
connect to Postgres via Drizzle
read all words from SQLite
for each batch of 100 words:
begin transaction
INSERT words into Postgres
for each word:
read its senses from SQLite
INSERT senses into Postgres
for each sense:
read its translations from SQLite
parse definitions/examples from JSON string → TEXT[]
INSERT translations into Postgres
commit transaction
log progress
log final counts: words, senses, translations
```
- Parse `definitions` and `examples` from JSON strings (SQLite)
into actual arrays (Postgres TEXT[]).
- Use `ON CONFLICT DO NOTHING` to handle re-runs gracefully.
### 2.5 Run and verify
- Run the import script.
- Compare counts:
```sql
-- In SQLite
SELECT COUNT(*) FROM words;
SELECT COUNT(*) FROM senses;
SELECT COUNT(*) FROM translations;
-- In Postgres
SELECT COUNT(*) FROM words;
SELECT COUNT(*) FROM senses;
SELECT COUNT(*) FROM translations;
```
- Counts should match (minus any rows that failed validation).
- Run the game query manually in Postgres:
```sql
SELECT w.headword, s.definitions, s.examples,
t.translation, t.gender
FROM words w
JOIN senses s ON s.word_id = w.id
JOIN translations t ON t.sense_id = s.id
WHERE w.language_code = 'de'
AND w.pos = 'noun'
AND s.difficulty IN ('easy', 'medium')
AND t.target_language_code = 'es'
AND t.difficulty = 'medium'
ORDER BY RANDOM()
LIMIT 5;
```
- Verify the results make sense.
---
## Phase 3: App Integration — Detailed
### 3.1 Rewrite getGameTerms
- Open the file containing `getGameTerms`.
- Replace the old query (2-table join on `vocabulary_entries` +
`entry_translations`) with the new 3-table join
(words → senses → translations).
- Parameters: sourceLanguage, targetLanguage, pos, difficulty, rounds.
- Difficulty filter:
- `senses.difficulty IN (all levels up to and including selected)`
— ceiling logic
- `translations.difficulty = selected` — exact match
- Return: word_id, headword, sense_id, definitions, examples,
translation, gender.
### 3.2 Rewrite getDistractors
- Same 3-table join.
- Additional filters:
- `t.sense_id != :currentSenseId`
- `t.translation != :correctAnswer`
- LIMIT 3.
- If fewer than 3 distractors are found (small data pool), handle
gracefully: reduce the number of options or log a warning.
### 3.3 Update exercise generation
- In the function that assembles a game round:
- Pick one random definition:
`definitions[Math.floor(Math.random() * definitions.length)]`
- Pick one random example:
`examples[Math.floor(Math.random() * examples.length)]`
- Combine 1 correct translation + 3 distractors.
- Shuffle the 4 options (Fisher-Yates or similar).
- Attach gender to each option for display.
### 3.4 Test matrix
Run through this matrix manually in the dev app:
| Source | Target | POS | Difficulty | Rounds | Pass? |
| ------ | ------ | ---- | ---------- | ------ | ----- |
| de | es | noun | easy | 10 | |
| es | de | noun | medium | 10 | |
| en | fr | noun | hard | 10 | |
| it | de | noun | easy | 10 | |
| fr | en | noun | medium | 10 | |
For each: verify definitions are in the source language, translations
are in the target language, genders are shown, no same-sense synonyms
appear as distractors, no duplicate options.
---
## Phase 4: Production Deploy — Detailed
### 4.1 Pre-deploy checklist
- [ ] All Phase 3 acceptance criteria pass
- [ ] `git status` is clean, all changes committed
- [ ] The Drizzle migration file is committed to the repo
- [ ] You know your prod database connection string
- [ ] You have a backup method for prod (pg_dump, hosting provider
snapshot, etc.)
### 4.2 Deploy
- Back up prod:
```
pg_dump -h <host> -U <user> -d <db> > backup_$(date +%Y%m%d).sql
```
- Run migration on prod:
```
DATABASE_URL=<prod-url> npx drizzle-kit migrate
```
- Run import script against prod:
```
DATABASE_URL=<prod-url> npx tsx scripts/import-to-postgres.ts
```
- Verify counts on prod.
### 4.3 Post-deploy verification
- Open the live app. Play 2 full games with different settings.
- Check server logs for errors.
- Wait 24h. Check logs again.
- If everything is clean, drop old tables:
```sql
DROP TABLE IF EXISTS entry_translations;
DROP TABLE IF EXISTS vocabulary_entries;
```
(Or keep them for another week if you want a safety net.)
### 4.4 Rollback plan
If something goes wrong:
- Restore the backup:
```
psql -h <host> -U <user> -d <db> < backup_20260722.sql
```
- Revert the code to the previous commit.
- Redeploy.
---
## Phase 5: Extend POS — Detailed
### 5.1 Per POS
Repeat the Phase 1 pipeline for each new POS:
- Acquire wordlists (verbs, adjectives, adverbs) for all 5 languages.
- Adjust the Gemini prompt:
- Verbs: ask for transitivity, common prepositions, or other
verb-specific metadata if needed for future exercises.
- Adjectives: ask for the base form. Note that adjective metadata
differs from noun metadata (no gender on the adjective itself
in the same way — gender applies to the noun it modifies).
- Adverbs: typically simpler. May not need gender at all.
- Run pipeline → validate → SQLite → import → Postgres.
- Test in the app with the POS filter set to the new type.
### 5.2 No schema changes
The `pos` column already supports all four types. The `gender`
column is nullable and simply won't be populated for adverbs or
English words. No migration needed.
---
# Risks & Mitigations
| Risk | Likelihood | Impact | Mitigation |
| ------------------------------------------------------ | ----------------- | ------------------------- | ------------------------------------------------------------------------------------------------- |
| Gemini returns malformed JSON | Medium | Pipeline stalls | Strip markdown fences, wrap parsing in try/catch, log and skip bad batches |
| Gemini produces wrong genders or difficulties | Medium | Bad exercise data | Validation module rejects invalid entries. Spot-check 50+ entries per language |
| Too few words at a given difficulty/pos/language combo | Medium | Game can't fill 4 options | Graceful fallback: reduce options or show a "not enough words" message. Log which combos are thin |
| SQLite → Postgres import fails mid-way | Low | Partial data in Postgres | Transactions per batch. Re-run with ON CONFLICT DO NOTHING |
| ORDER BY RANDOM() gets slow at scale | Low (not at 500k) | Slow game load | Add a TODO comment. Optimize with random() pre-filter when needed |
| Prod migration breaks the live app | Low | Downtime | Backup before migrating. Rollback plan documented. Deploy during low-traffic hours |
---
# File / Folder Structure (new files)
```
project/
├── docs/
│ ├── schema-design.md ← companion design doc
│ └── roadmap.md ← this document
├── data/
│ ├── wordlists/
│ │ ├── nouns_de.txt
│ │ ├── nouns_en.txt
│ │ ├── nouns_es.txt
│ │ ├── nouns_fr.txt
│ │ └── nouns_it.txt
│ ├── rejections/ ← invalid Gemini entries (for review)
│ └── staging.db ← SQLite staging database
├── pipeline/
│ ├── run.ts ← main pipeline script
│ ├── validate.ts ← validation module
│ ├── prompt.ts ← Gemini prompt template
│ └── schema.sql ← SQLite schema definition
├── scripts/
│ └── import-to-postgres.ts ← SQLite → Postgres import
├── src/
│ └── db/
│ └── schema.ts ← Drizzle schema (updated)
└── drizzle/ ← generated migration files
```