6 Commits

Author SHA1 Message Date
prosolis
704c06f41d Add regional variant tagging (us/gb) for English words
- SCOWL loader now loads american-words and british-words alongside
  english-words, tagging each with variant "us", "gb", or NULL (common)
- Words appearing in both American and British lists get NULL variant
- Add variant column to words table with schema migration
- API: /define returns variant field for English, /random and /words
  accept ?variant=us|gb filter
- Add total word count to server status line
- Fix rows.Close leak in PopulateCMUTails, add rows.Err() check
- Add tests for PopulateCMUTails, cmuTail, TotalWordCount
- Expand test seed data and assertions to cover pt-PT definitions,
  synonyms, and validity checks
- Update README with variant docs, /difficulty, /words, /rhyme
  endpoints, and current database statistics

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-04-04 15:21:27 -07:00
prosolis
56e073f753 Add bearer token auth and clean up debug logging
- Bearer token auth via --token flag or DREAMDICT_TOKEN env var
- Timing-safe comparison with crypto/subtle
- /health exempt for monitoring without credentials
- 401 response for missing/invalid tokens
- Auth disabled when no token configured (backwards compatible)
- Remove debug logging from cetempublico loader

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-04-03 08:54:40 -07:00
prosolis
25ed087c3e Add /difficulty and /rhyme endpoints, CMU tail migration for rhyme matching
New /difficulty endpoint returns word difficulty scores (0.0-1.0).
New /rhyme endpoint uses CMU phoneme tail matching to find rhyming words,
sorted by frequency. Adds cmu_tail column migration with batch population
at startup, and index for efficient reverse lookups.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-04-02 06:45:29 -07:00
prosolis
207086da08 Add Phase 2+3 features: antonyms, backing, pronunciation, etymology, difficulty, affix expansion
New endpoints: /antonyms, /backing, /pronunciation, /etymology with difficulty
scoring on /random. Cross-language synset backing links French/Portuguese words
to English equivalents via WordNet 3.0 synset IDs (matching WOLF and OMW offsets).

New loaders: Hunspell affix expansion (fr, pt-PT), English programmatic inflector,
CMU Pronouncing Dictionary, SUBTLEX-US frequency, CETEMPúblico frequency,
Open Multilingual Wordnet (Portuguese), and SQL-based difficulty scoring.

Schema v2 adds tables: antonyms, synsets, word_synsets, pronunciations, etymology
with migration support for existing databases.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-04-02 01:01:06 -07:00
prosolis
d5e4ce0d36 Add live status line and frequency batch endpoint
Display query count, error count, req/s, uptime, and last endpoint
hit on a single updating line to stdout. Also wire up the
/frequency/batch route that was missing from the mux.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-31 20:01:51 -07:00
prosolis
eb471b6f6d Add word frequency endpoint and SCOWL frequency data
SCOWL loader now sets frequency based on tier (10=1000 most common,
70=50 rare). New GET /frequency?word=X&lang=Y endpoint returns frequency
score for a word. Enables downstream consumers to identify rare/
sophisticated vocabulary from dictionary data rather than LLM guessing.

Also fix .gitignore matching cmd/server/ directory instead of just the
server binary.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-30 21:27:45 -07:00