- Provision 872 verified Ingglish phonetic entries (253 Tier-1 lexicon/bridge items + 619 Tier-2 multi-scale 333/666/777/Numogram equations) formatted with native `[Ingglish: <English Gloss>]` comment brackets. - Guard short Ingglish syllables (`di`, `li`, `mi`, `vi`, `iv`) against Roman-numeral false positives in `tools/normalize_caps.py`. - Add newcomer onboarding documentation (`docs/ONBOARDING-INGGLISH-AQ.md`). - Bump `databaseVersion` in `config.default.js` to cache-bust precomputed JSON. - Verified 100% pass rate across all 4 CI gates (0 normalize mutations, 100.0% classify_candidates keep rate, 0 comment-strip/AQ mismatches). |
||
|---|---|---|
| .. | ||
| all_ciphers.py | ||
| classify_candidates.py | ||
| generate_db_json.py | ||
| normalize_caps.py | ||
| README.md | ||
Tools
Rebuilding the precomputed database (db.json)
The calculator works out of the box in live mode — it loads db.txt (one phrase per
line) and computes cipher values on the fly. For instant matching (~100ms vs 2–5s per
query), it can instead load a precomputed db.json containing all 69 cipher values per entry.
db.json is a build artifact derived from db.txt + calc/ciphers.js. It is not committed
to the repo (it's large). Two ways to get it:
-
Download the prebuilt
db.jsonfrom the latest GitHub Release. Place it next toindex.html. -
Regenerate it yourself:
python3 tools/generate_db_json.py db.txt db.jsonRequires Python 3. Reads cipher definitions from
calc/ciphers.js, scores every entry indb.txtagainst all 69 English-applicable ciphers, and writesdb.json.
The calculator auto-detects db.json (see databaseMode in config.js, default "auto"):
it tries db.json first and falls back to db.txt if absent.
Files
generate_db_json.py— buildsdb.jsonfrom a flatdb.txt.all_ciphers.py— parsescalc/ciphers.jsand scores text against all English ciphers (used by the generator). Handles case-sensitive ciphers and strips[...]display annotations before scoring, matching the live engine.