merge-vocab-dict.py

# Augment dict-<lang>.json with vocabulary metadata (JLPT readings/meanings, Yonsei zh/en)# so every game word has at least a minimal dictionary entry.import json, osLANG = "D:/gw-dl/lang"for lang, src_key in (("ja", "jlpt"), ("ko", "yonsei")):    dpath = f"{LANG}/dict-{lang}.json"    d = json.load(open(dpath, encoding="utf-8")) if os.path.exists(dpath) else {}    vocab = json.load(open(f"{LANG}/vocab-{lang}.json", encoding="utf-8"))    pack = json.load(open(f"{LANG}/pack-{lang}.json", encoding="utf-8"))    ingame = set(pack["words"].split("|"))    added = enriched = 0    for c in vocab:        w = c["w"]        if w not in ingame or c["src"] != src_key:            continue        e = d.get(w)        if e is None:            senses = []            if c.get("en"):                senses.append({"d": c["en"]})            if c.get("zh"):                senses.append({"d": "中文: " + c["zh"]})            if senses:                e = {"p": [{"p": c.get("pos") or "—", "s": senses}]}                if c.get("reading"):                    e["rd"] = c["reading"]                d[w] = e                added += 1        else:            if c.get("reading") and not e.get("rd"):                e["rd"] = c["reading"]            if c.get("zh"):                if not any(s.get("d", "").startswith("中文:") for g in e["p"] for s in g["s"]):                    e["p"].append({"p": "中文", "s": [{"d": "中文: " + c["zh"]}]})            enriched += 1    json.dump(d, open(dpath, "w"), ensure_ascii=False)    print(f"{lang}: +{added} from vocab metadata, {enriched} enriched | total {len(d)} | {os.path.getsize(dpath)//1024} KB")