169 lines
8.5 KiB
Python
169 lines
8.5 KiB
Python
#!/usr/bin/env python3
|
||
# -*- coding: utf-8 -*-
|
||
"""
|
||
04_ref_index.py —— 从《辅行诀五脏用药法要》基础文本（印刷本清洗版）抽取
|
||
「核校参考索引」，供《药性探真》后续章节清洗时卡药名 / 方名 / 五行互含名位。
|
||
|
||
产出：
|
||
02_加工数据/参考索引_辅行诀药名方名.json 机器可读索引
|
||
02_加工数据/参考索引_辅行诀药名方名.md 人工速查表
|
||
|
||
底本：02_加工数据/辅行诀五脏用药法要_基础文本_清洗版.md（baseline，md5 见 README）
|
||
性质：只读底本、不修改任何原文；幂等可重跑。
|
||
用法：
|
||
python3 05_脚本工具/04_ref_index.py # 生成索引
|
||
python3 05_脚本工具/04_ref_index.py 药名 # 查某个词是否在参考表内
|
||
"""
|
||
import json
|
||
import re
|
||
import sys
|
||
import hashlib
|
||
from pathlib import Path
|
||
|
||
ROOT = Path(__file__).resolve().parent.parent
|
||
SRC = ROOT / "02_加工数据" / "辅行诀五脏用药法要_基础文本_清洗版.md"
|
||
OUT_JSON = ROOT / "02_加工数据" / "参考索引_辅行诀药名方名.json"
|
||
OUT_MD = ROOT / "02_加工数据" / "参考索引_辅行诀药名方名.md"
|
||
|
||
TASTE_TO_XING = {"辛": "木", "咸": "火", "甘": "土", "酸": "金", "苦": "水"}
|
||
|
||
# 底本中同物异写的形态（同一底本内两形并见者，仅登记不改动底本）
|
||
ALIAS_NOTES = [
|
||
{"规范形": "栝楼", "异写": ["栝蒌"], "依据": "底本第429行「栝楼为土中土」，方剂篇作「栝蒌」；《药性探真》第一章作「栝蒌」"},
|
||
{"规范形": "山茱萸", "异写": ["萸肉"], "依据": "药精表作「萸肉」，方剂篇加减法作「山茱萸」"},
|
||
{"规范形": "干姜", "异写": ["甘草炙→炙甘草"], "依据": "药精表「甘草炙」「甘草」分列二味（土中金／土中火），方剂篇作「甘草，炙」"},
|
||
]
|
||
|
||
|
||
def parse(lines):
|
||
yaojing, hunhan, fangming = [], [], []
|
||
# ── 1. 二十五味药精 ───────────────────────────────
|
||
for i, line in enumerate(lines, 1):
|
||
m = re.match(r"^味([辛咸甘酸苦])皆属([木火土金水])[，,]", line)
|
||
if not m:
|
||
continue
|
||
taste, benxing = m.group(1), m.group(2)
|
||
assert TASTE_TO_XING[taste] == benxing, f"第{i}行 味-行不符：{taste}≠{benxing}"
|
||
for seg in re.split(r"[；;。]", line[m.end():]):
|
||
seg = seg.strip()
|
||
if not seg:
|
||
continue
|
||
mm = re.match(r"^(\S+?)\s+(\S+?)(?:为之主|为([木火土金水]))$", seg)
|
||
if not mm:
|
||
raise SystemExit(f"第{i}行 药精条无法解析：{seg}")
|
||
drug, mineral, hanxing = mm.group(1), mm.group(2), mm.group(3)
|
||
is_main = hanxing is None
|
||
hanxing = benxing if is_main else hanxing
|
||
yaojing.append({
|
||
"药名": drug, "味": taste, "本行": benxing,
|
||
"互含名位": f"{benxing}中{hanxing}",
|
||
"配对金石药": mineral, "药精位次": "主" if is_main else f"含{hanxing}",
|
||
"行号": i, "底本依据": line.strip(),
|
||
})
|
||
# ── 2. 十三种药（五行互含，备心病方之用） ──────────
|
||
start = None
|
||
for i, line in enumerate(lines, 1):
|
||
if "又有药十三种" in line:
|
||
start = i
|
||
break
|
||
if start is None:
|
||
raise SystemExit("未找到「又有药十三种」条")
|
||
body = lines[start] # 该条正文紧随其后
|
||
for i in range(start, min(start + 4, len(lines))):
|
||
if re.search(r"为[木火土金水]中[木火土金水]", lines[i]):
|
||
body = lines[i]
|
||
body_line = i + 1
|
||
break
|
||
for seg in re.split(r"[；;。，,]", body):
|
||
seg = seg.strip()
|
||
if not seg:
|
||
continue
|
||
if seg.startswith("又为"):
|
||
if not hunhan:
|
||
raise SystemExit(f"第{body_line}行「又为」无前项：{seg}")
|
||
hunhan[-1]["互含名位"].append(seg[2:])
|
||
continue
|
||
mm = re.match(r"^(\S+?)为([木火土金水]中[木火土金水])$", seg)
|
||
if not mm:
|
||
raise SystemExit(f"第{body_line}行 十三种药条无法解析：{seg}")
|
||
hunhan.append({"药名": mm.group(1), "互含名位": [mm.group(2)], "行号": body_line,
|
||
"底本依据": body.strip()})
|
||
# ── 3. 方名 ───────────────────────────────────────
|
||
seen = set()
|
||
for i, line in enumerate(lines, 1):
|
||
m = re.match(r"^([^：:\s]{2,16}?(?:汤|散|丸|煎|膏|酒|方))[：:]", line)
|
||
if not m:
|
||
continue
|
||
name = m.group(1)
|
||
if name in seen:
|
||
continue
|
||
seen.add(name)
|
||
fangming.append({"方名": name, "行号": i, "主治": line.split("：", 1)[-1][:40]})
|
||
return yaojing, hunhan, fangming, body_line
|
||
|
||
|
||
def main():
|
||
text = SRC.read_text(encoding="utf-8")
|
||
lines = text.split("\n")
|
||
yaojing, hunhan, fangming, hunhan_line = parse(lines)
|
||
|
||
drugs = [d["药名"] for d in yaojing] + [d["药名"] for d in hunhan]
|
||
index = {
|
||
"来源": SRC.name,
|
||
"来源md5": hashlib.md5(text.encode("utf-8")).hexdigest(),
|
||
"生成脚本": "05_脚本工具/04_ref_index.py",
|
||
"说明": "《药性探真》后续章节清洗的核校参考表：药名、五行互含名位、配对金石药、方名。",
|
||
"药精二十五味": yaojing,
|
||
"五行互含十三种药": hunhan,
|
||
"方名清单": fangming,
|
||
"药名一览": sorted(set(drugs)),
|
||
"同物异写登记": ALIAS_NOTES,
|
||
"计数": {"药精": len(yaojing), "十三种药": len(hunhan),
|
||
"药名合计": len(set(drugs)), "方名": len(fangming)},
|
||
}
|
||
OUT_JSON.write_text(json.dumps(index, ensure_ascii=False, indent=2), encoding="utf-8")
|
||
|
||
# ── 人工速查 md ───────────────────────────────────
|
||
L = ["# 参考索引｜《辅行诀五脏用药法要》药名·方名·五行互含名位", "",
|
||
f"> 生成：`05_脚本工具/04_ref_index.py`（幂等）｜底本：`{SRC.name}`（md5 `{index['来源md5']}`）",
|
||
"> 用途：《药性探真》后续章节清洗时的**核校参考**（药名规范形、五行互含名位、配对金石药、方名）。",
|
||
"> 原则：只作旁证，不改动《药性探真》原文；发现不一致→登记人工核对清单。", "",
|
||
"## 一、二十五味药精（《汤液经法》五行互含）", "",
|
||
"| 味 | 本行 | 互含名位 | 药名 | 配对金石药 | 底本行号 |", "|---|---|---|---|---|---|"]
|
||
for d in yaojing:
|
||
L.append(f"| {d['味']} | {d['本行']} | {d['互含名位']} | {d['药名']} | {d['配对金石药']} | {d['行号']} |")
|
||
L += ["", "## 二、五行互含十三种药（备心病方之用）", "", "| 药名 | 互含名位 | 底本行号 |", "|---|---|---|"]
|
||
for d in hunhan:
|
||
L.append(f"| {d['药名']} | {'、'.join(d['互含名位'])} | {d['行号']} |")
|
||
L += ["", "## 三、方名清单", "", "| 方名 | 底本行号 | 主治（截断） |", "|---|---|---|"]
|
||
for f in fangming:
|
||
L.append(f"| {f['方名']} | {f['行号']} | {f['主治']} |")
|
||
L += ["", "## 四、同物异写登记（只登记，不改底本）", ""]
|
||
for a in ALIAS_NOTES:
|
||
L.append(f"- **{a['规范形']}** ／异写：{'、'.join(a['异写'])} —— {a['依据']}")
|
||
L += ["", "## 五、药名一览（去重）", "", "、".join(index["药名一览"]), ""]
|
||
OUT_MD.write_text("\n".join(L), encoding="utf-8")
|
||
|
||
print(f"药精 {len(yaojing)} 味 | 十三种药 {len(hunhan)} 味 | 药名去重 {len(set(drugs))} | 方名 {len(fangming)}")
|
||
print(f"索引：{OUT_JSON.relative_to(ROOT)}")
|
||
print(f"速查：{OUT_MD.relative_to(ROOT)}")
|
||
|
||
|
||
def lookup(word):
|
||
"""查词是否在参考表内（供后续章节清洗时快速卡药名）。"""
|
||
idx = json.loads(OUT_JSON.read_text(encoding="utf-8"))
|
||
hits = [d for d in idx["药精二十五味"] + idx["五行互含十三种药"] if word in d["药名"]]
|
||
fangs = [f for f in idx["方名清单"] if word in f["方名"]]
|
||
print(f"「{word}」 药名命中 {len(hits)} / 方名命中 {len(fangs)}")
|
||
for h in hits:
|
||
print(" 药:", h["药名"], h.get("互含名位"), h.get("配对金石药", ""))
|
||
for f in fangs:
|
||
print(" 方:", f["方名"], "→", f["主治"])
|
||
|
||
|
||
if __name__ == "__main__":
|
||
if len(sys.argv) > 1:
|
||
lookup(sys.argv[1])
|
||
else:
|
||
main()
|