267 lines
11 KiB
Python
267 lines
11 KiB
Python
#!/usr/bin/env python3
|
|
"""Expand rune clusters, restore ghost primaries, stamp [Руны: …] on the 1000."""
|
|
from __future__ import annotations
|
|
|
|
import json
|
|
import re
|
|
from collections import defaultdict
|
|
from pathlib import Path
|
|
|
|
ROOT = Path(__file__).resolve().parents[1]
|
|
CANON = ROOT / "oracle_enricher" / "data" / "traditions_canon.json"
|
|
ASSET = ROOT / "assets" / "data" / "traditions_canon.json"
|
|
LEX = ROOT / "oracle_enricher" / "data" / "lexicon.json"
|
|
DART = ROOT / "lib" / "data" / "words_data.dart"
|
|
HTML = ROOT / "lexicon_viewer.html"
|
|
MD = ROOT / "words_alphabetical.md"
|
|
|
|
RUNE_TAG_RE = re.compile(r"\[Руны:[^\]]*\]\s*")
|
|
|
|
# Extra lexicon words on top of the station primary.
|
|
CLUSTER_OVERRIDE = {
|
|
"fehu": ["изобилие", "полнота", "успех", "благословение", "рост", "начало"],
|
|
"uruz": ["сила", "тело", "жизнь", "воля", "желание", "кровь"],
|
|
"jera": ["осень", "цикл", "год", "труд", "семя", "плод", "хлеб"],
|
|
"perthro": ["тайна", "игра", "случай", "судьба", "завеса", "сосуд", "жребий", "кость"],
|
|
"tiwaz": ["отец", "клятва", "закон", "мужчина", "меч", "копьё", "судья"],
|
|
"laguz": ["вода", "глубина", "волна", "девушка", "море", "река", "родник", "озеро", "ручей"],
|
|
}
|
|
|
|
# Viewer/lexicon name -> live dart name
|
|
WORD_REMAP = {
|
|
"солнцестояние": "солнцеворот",
|
|
}
|
|
|
|
# Live dart still has frequency-list verbs on cells whose canon primary is a sacred noun.
|
|
DART_RESTORE = {
|
|
"глядеть": "золото",
|
|
"забыть": "тёрн",
|
|
"возникнуть": "град",
|
|
"вспомнить": "нужда",
|
|
"возвращаться": "урожай",
|
|
}
|
|
|
|
RESTORE_META = {
|
|
"золото": (
|
|
"существительное",
|
|
"Священный Предмет",
|
|
"Совершенный солнечный металл алхимии, нетленный свет, царский венец и мера высшей ценности. "
|
|
"Золото не ржавеет в земле и не тускнеет от времени, знаменуя победу духа над распадом материи. "
|
|
"В раскладе это знак высшего успеха, обретения нетленного сокровища мудрости и кристальной чистоты намерения.",
|
|
),
|
|
"тёрн": (
|
|
"существительное",
|
|
"Сакральный Символ",
|
|
"Черный шип неизбежности, страж предела, колючая ограда рока, через суровость ведущая к очищению духа.",
|
|
),
|
|
"град": (
|
|
"существительное",
|
|
"Образ / Архетип",
|
|
"Стихийное ледяное зерно с небес, внезапный сокрушительный удар бури, ломающий старые формы ради очищения.",
|
|
),
|
|
"нужда": (
|
|
"существительное",
|
|
"Образ / Архетип",
|
|
"Священный крест необходимости, стеснение рока, из которого высекается внутренний огонь спасения и несгибаемая стойкость.",
|
|
),
|
|
"урожай": (
|
|
"существительное",
|
|
"Образ / Архетип",
|
|
"Благодатный сбор плодов долгого труда, венчание годового цикла, щедрость земли и торжество плодородия.",
|
|
),
|
|
}
|
|
|
|
|
|
def tag_for(st: dict) -> str:
|
|
name = st["name"]
|
|
symbol = st.get("symbol", "")
|
|
group = st.get("group", "")
|
|
if " (" in name:
|
|
ru, rest = name.split(" (", 1)
|
|
en = rest.rstrip(")")
|
|
return f"[Руны: {group} {ru} ({en} / {symbol})]"
|
|
return f"[Руны: {group} {name}]"
|
|
|
|
|
|
def restore_dart(text: str) -> str:
|
|
for old, new in DART_RESTORE.items():
|
|
pos, cat, desc = RESTORE_META[new]
|
|
desc_esc = desc.replace("\\", "\\\\").replace('"', '\\"')
|
|
|
|
def repl(m: re.Match, new=new, pos=pos, cat=cat, desc_esc=desc_esc) -> str:
|
|
return (
|
|
f"{m.group(1)}{new}{m.group(2)}"
|
|
f"{m.group(3)}{pos}{m.group(4)}"
|
|
f"{m.group(5)}{cat}{m.group(6)}"
|
|
f"{m.group(7)}{desc_esc}{m.group(8)}"
|
|
)
|
|
|
|
pat = re.compile(
|
|
rf'(word: "){re.escape(old)}(",\s*\n)'
|
|
rf'(\s*partOfSpeech: ")[^"]+(",\s*\n)'
|
|
rf'(\s*frequency: [^\n]+\n\s*category: ")[^"]+(",\s*\n)'
|
|
rf'(\s*description: ")[^"]*(")',
|
|
)
|
|
if f'word: "{new}",' in text and f'word: "{old}",' not in text:
|
|
continue
|
|
text2, n = pat.subn(repl, text, count=1)
|
|
if n != 1:
|
|
raise SystemExit(f"failed to restore {old} -> {new} (n={n})")
|
|
text = text2
|
|
return text
|
|
|
|
|
|
def payload_from(word: str, dart_id: int, lex_by_word: dict, old: dict | None) -> dict:
|
|
lex = lex_by_word.get(word) or {}
|
|
old = old or {}
|
|
return {
|
|
"id": dart_id,
|
|
"word": word,
|
|
"emoji": lex.get("emoji") or old.get("emoji") or "",
|
|
"pos": lex.get("pos") or old.get("pos") or "существительное",
|
|
"category": lex.get("category") or old.get("category") or "",
|
|
"desc": old.get("desc") or "",
|
|
}
|
|
|
|
|
|
def main() -> None:
|
|
dart_text = restore_dart(DART.read_text(encoding="utf-8"))
|
|
DART.write_text(dart_text, encoding="utf-8")
|
|
|
|
dart_by_word: dict[str, int] = {}
|
|
dart_pos: dict[str, str] = {}
|
|
dart_cat: dict[str, str] = {}
|
|
for m in re.finditer(
|
|
r'WordItem\(\s*id:\s*(\d+),\s*word:\s*"([^"]+)",\s*'
|
|
r'partOfSpeech:\s*"([^"]+)",\s*frequency:\s*[^\n]+\n\s*category:\s*"([^"]+)"',
|
|
dart_text,
|
|
):
|
|
dart_by_word[m.group(2)] = int(m.group(1))
|
|
dart_pos[m.group(2)] = m.group(3)
|
|
dart_cat[m.group(2)] = m.group(4)
|
|
|
|
lex_list = json.loads(LEX.read_text(encoding="utf-8"))
|
|
lex_by_word = {e["word"]: e for e in lex_list}
|
|
canon = json.loads(CANON.read_text(encoding="utf-8"))
|
|
runes = canon["runes"]
|
|
word_to_tags: dict[str, list[str]] = defaultdict(list)
|
|
|
|
for st in runes["stations"]:
|
|
sid = st["id"]
|
|
tag = tag_for(st)
|
|
prim = st.get("primary") or {}
|
|
prim_w = prim.get("word")
|
|
if prim_w in WORD_REMAP:
|
|
prim_w = WORD_REMAP[prim_w]
|
|
if prim_w not in dart_by_word:
|
|
raise SystemExit(f"missing primary {sid}/{prim_w} in dart")
|
|
st["primary"] = payload_from(prim_w, dart_by_word[prim_w], lex_by_word, prim)
|
|
# keep the existing primary desc (rune lore) if present
|
|
if prim.get("desc"):
|
|
st["primary"]["desc"] = prim["desc"]
|
|
|
|
old_cluster = {c.get("word"): c for c in st.get("cluster") or []}
|
|
if sid in CLUSTER_OVERRIDE:
|
|
extra_names = CLUSTER_OVERRIDE[sid]
|
|
else:
|
|
extra_names = [c.get("word") for c in st.get("cluster") or [] if c.get("word")]
|
|
extra: list[str] = []
|
|
for w in extra_names:
|
|
w = WORD_REMAP.get(w, w)
|
|
if w == prim_w or w not in dart_by_word or w in extra:
|
|
continue
|
|
extra.append(w)
|
|
st["cluster"] = [
|
|
payload_from(w, dart_by_word[w], lex_by_word, old_cluster.get(w))
|
|
for w in extra
|
|
]
|
|
|
|
word_to_tags[prim_w].append(tag)
|
|
for w in extra:
|
|
word_to_tags[w].append(tag)
|
|
|
|
text = json.dumps(canon, ensure_ascii=False, indent=2) + "\n"
|
|
CANON.write_text(text, encoding="utf-8")
|
|
ASSET.write_text(text, encoding="utf-8")
|
|
|
|
dart_lines = dart_text.splitlines(keepends=True)
|
|
nstamp = 0
|
|
i = 0
|
|
while i < len(dart_lines):
|
|
wm = re.match(r'(\s*word: ")([^"]+)(",\s*)$', dart_lines[i])
|
|
if wm:
|
|
word = wm.group(2)
|
|
for j in range(i + 1, min(i + 12, len(dart_lines))):
|
|
raw = dart_lines[j].rstrip("\n")
|
|
dm = re.match(r'^(\s*description: ")(.*)(",)\s*$', raw)
|
|
if dm:
|
|
body = RUNE_TAG_RE.sub("", dm.group(2)).strip()
|
|
tags = word_to_tags.get(word)
|
|
if tags:
|
|
dart_lines[j] = f"{dm.group(1)}{' '.join(tags)} {body}{dm.group(3)}\n"
|
|
nstamp += 1
|
|
else:
|
|
dart_lines[j] = f"{dm.group(1)}{body}{dm.group(3)}\n"
|
|
break
|
|
if dart_lines[j].strip().startswith("WordItem("):
|
|
break
|
|
i += 1
|
|
dart2 = "".join(dart_lines)
|
|
DART.write_text(dart2, encoding="utf-8")
|
|
|
|
html = HTML.read_text(encoding="utf-8")
|
|
start = html.find(" const TRADITIONS_CANON = ")
|
|
end = html.find(" const WORDS_DATA = ")
|
|
if start > 0 and end > start:
|
|
html = (
|
|
html[:start]
|
|
+ " const TRADITIONS_CANON = "
|
|
+ json.dumps(canon, ensure_ascii=False, indent=2)
|
|
+ ";\n\n"
|
|
+ html[end:]
|
|
)
|
|
m = re.search(r"const WORDS_DATA = (\[.*?\]);", html, flags=re.DOTALL)
|
|
if not m:
|
|
raise SystemExit("WORDS_DATA not found")
|
|
words = json.loads(m.group(1))
|
|
dart_descs = dict(re.findall(r'word: "([^"]+)"[\s\S]*?description: "([^"]*)"', dart2))
|
|
for w in words:
|
|
wd = w.get("word")
|
|
if wd in dart_descs:
|
|
w["description"] = dart_descs[wd]
|
|
html = html[: m.start(1)] + json.dumps(words, ensure_ascii=False) + html[m.end(1) :]
|
|
HTML.write_text(html, encoding="utf-8")
|
|
|
|
if MD.exists():
|
|
md = MD.read_text(encoding="utf-8")
|
|
|
|
def repl_block(mm: re.Match) -> str:
|
|
word = mm.group(1)
|
|
body = mm.group(3)
|
|
if word not in dart_descs:
|
|
return mm.group(0)
|
|
body2 = re.sub(
|
|
r"\* \*\*Толкование:\*\* .+",
|
|
f"* **Толкование:** {dart_descs[word]}",
|
|
body,
|
|
count=1,
|
|
)
|
|
return f"### {word} `{mm.group(2)}`\n{body2}"
|
|
|
|
md2 = re.sub(
|
|
r"### ([^\s`]+) `(#[^`]+)`\n([\s\S]*?)(?=\n### |\n## |\Z)",
|
|
repl_block,
|
|
md,
|
|
)
|
|
MD.write_text(md2, encoding="utf-8")
|
|
|
|
print(f"stamped {nstamp} dart words, tagged {len(word_to_tags)}")
|
|
for st in runes["stations"]:
|
|
prim = (st.get("primary") or {}).get("word")
|
|
cl = [c["word"] for c in st.get("cluster") or []]
|
|
print(f" {st['id']:12} prim={prim:12} n={len(cl)} {cl}")
|
|
|
|
|
|
if __name__ == "__main__":
|
|
main()
|