(.*)', html) if not m: return fallback body = m.group(1) for marker in ('id="sidebar', 'id="footer', "sharedaddy", "yarpp"): i = body.find(marker) if i > 200: body = body[:i] break text = strip_tags(body) text = re.split(r"(?i)Смотрите также", text)[0] text = re.split(r"(?i)Рекомендуем по этой теме", text)[0] text = re.sub(r"(?im)^Карты Ленорман .*$", "", text) text = re.sub(r"\n{3,}", "\n\n", text).strip() return text if len(text) > 80 else fallback def preview_of(readings: list[dict]) -> str: for src in SOURCE_ORDER: for r in readings: if r.get("source") == src and r.get("text"): t = r["text"].replace("\n", " ").strip() return t if len(t) <= 160 else t[:157].rstrip() + "…" return "" def main() -> None: corpus = json.loads(CORPUS.read_text(encoding="utf-8")) canon = json.loads(CANON.read_text(encoding="utf-8")) unified = {} if UNIFIED.exists(): unified = json.loads(UNIFIED.read_text(encoding="utf-8")) stations = canon["lenormand"]["stations"] by_n: dict[int, dict] = {} for st in stations: sid = st.get("id", "") if sid.startswith("card_"): n = int(sid.split("_")[1]) by_n[n] = st cards_out = [] for n in range(1, 37): cc = corpus["cards"][str(n)] st = by_n.get(n, {}) meanings = [] magya_short = "" for m in cc.get("meanings", []): src = m.get("source", "") text = (m.get("text") or "").strip() if src == "magya": text = clean_magya_meaning(text) mm = re.search(r"Краткое значение:\s*(.+)", text) if mm: magya_short = mm.group(1).split("\n")[0].strip(" .") elif src == "ezoplaneta": text = clean_ezo_meaning(text) elif src == "karusel": text = clean_karusel_from_cache(n, text) if len(text) < 40: continue meanings.append( { "source": src, "label": SOURCE_LABELS.get(src, src), "text": text, } ) pairs = [] for other in range(1, 37): if other == n: continue key = f"{min(n, other):02d}-{max(n, other):02d}" bucket = corpus["pairs"].get(key) if not bucket: continue readings = [] for r in sort_readings(bucket.get("readings", [])): readings.append( { "source": r.get("source"), "label": SOURCE_LABELS.get(r.get("source", ""), r.get("source", "")), "left": r.get("left"), "right": r.get("right"), "text": (r.get("text") or "").strip(), } ) if not readings: continue ost = by_n.get(other, {}) up = (unified.get("pairs") or {}).get(key) or {} utext = (up.get("text") or "").strip() pairs.append( { "n": other, "id": f"card_{other}", "name": corpus["cards"][str(other)]["name"], "symbol": ost.get("symbol", ""), "preview": (utext[:157] + "…") if len(utext) > 160 else (utext or preview_of(readings)), "text": utext, "readings": [] if utext else readings, } ) ucard = (unified.get("cards") or {}).get(str(n)) or {} umeaning = (ucard.get("meaning") or "").strip() ucore = (ucard.get("core") or "").strip() cards_out.append( { "n": n, "id": f"card_{n}", "name": cc["name"], "symbol": st.get("symbol", ""), "lore": ucore or magya_short or st.get("lore", ""), "group": st.get("group", f"Карта {n}"), "meaning": umeaning, "shadow": (ucard.get("shadow") or "").strip(), "meanings": [] if umeaning else meanings, "pairs": pairs, } ) names = {c["n"]: c["name"] for c in cards_out} triads = [] uni_triads = unified.get("triads") or [] if uni_triads: for t in uni_triads: nums = [int(x) for x in t.get("cards", [])] if len(nums) != 3: continue triads.append( { "cards": nums, "names": [names.get(x, str(x)) for x in nums], "text": (t.get("text") or "").strip(), } ) else: for t in corpus.get("triads", []): nums = [int(x) for x in t.get("cards", []) if str(x).isdigit()] if len(nums) != 3: continue triads.append( { "cards": nums, "names": [names.get(x, str(x)) for x in nums], "text": (t.get("text") or "").strip(), } ) extras = [] for e in corpus.get("extras", []): title = (e.get("title") or "").strip() text = (e.get("text") or "").strip() if len(text) < 80: continue extras.append( { "source": e.get("source", ""), "title": title, "text": text, } ) blob = { "system": "lenormand", "source_labels": SOURCE_LABELS, "card_count": 36, "pair_count": sum(len(c["pairs"]) for c in cards_out) // 2, "cards": cards_out, "triads": triads, "extras": extras, } text = json.dumps(blob, ensure_ascii=False, indent=2) + "\n" OUT_ASSET.parent.mkdir(parents=True, exist_ok=True) OUT_ASSET.write_text(text, encoding="utf-8") OUT_DATA.write_text(text, encoding="utf-8") # Supplement station lore in traditions canon (short line on the card). lore_by_n = {c["n"]: c["lore"] for c in cards_out if c.get("lore")} for canon_path in (CANON, ASSET_CANON): data = json.loads(canon_path.read_text(encoding="utf-8")) ln = data["lenormand"] ln["description"] = ( "36 карт французского оракула мадемуазель Ленорман: значения с трёх источников " "и 630 парных сочетаний (карусель, эзопланета, magya) — источники не схлопываются." ) for st in ln["stations"]: sid = st.get("id", "") if not sid.startswith("card_"): continue n = int(sid.split("_")[1]) if lore_by_n.get(n): st["lore"] = lore_by_n[n] canon_path.write_text( json.dumps(data, ensure_ascii=False, indent=2) + "\n", encoding="utf-8" ) if HTML_VIEWER.exists(): html = HTML_VIEWER.read_text(encoding="utf-8") start = html.find(" const TRADITIONS_CANON = ") end = html.find(" const WORDS_DATA = ") if start > 0 and end > start: data = json.loads(CANON.read_text(encoding="utf-8")) blob_js = ( " const TRADITIONS_CANON = " + json.dumps(data, ensure_ascii=False, indent=2) + ";\n\n" ) HTML_VIEWER.write_text(html[:start] + blob_js + html[end:], encoding="utf-8") print("patched", HTML_VIEWER.relative_to(ROOT)) print( f"cards={len(cards_out)} pairs_per_card={len(cards_out[0]['pairs'])} " f"triads={len(triads)} extras={len(extras)} bytes={len(text)}" ) print("wrote", OUT_ASSET.relative_to(ROOT)) if __name__ == "__main__": main()