329 lines
22 KiB
Python
329 lines
22 KiB
Python
#!/usr/bin/env python3
|
||
# -*- coding: utf-8 -*-
|
||
"""
|
||
辅行诀五脏用药法要-药性探真 资料库 ETL 第 2 步
|
||
01_来源数据/辅行诀脏腑用药法要_古本原文_20260915.md -> 02_加工数据/
|
||
|
||
用途：把用户提供的《辅行诀脏腑用药法要》古本原文导入资料库，作为全书原文依据参考。
|
||
本阶段仍只做数据清理，不做释义抽取 / 跨库关联 / 主题分析。
|
||
|
||
清理项：
|
||
A 版式残留：全角缩进（U+3000）、行尾空格、词间半角空格
|
||
B 校勘符号规范化：\\[X\\] -> 〔X〕（Markdown 转义残留 -> 传统校补号）
|
||
C 讹字/衍字修正：每条附核校依据（参校本＝维基文库本；内证＝本书条文自证）
|
||
D 标点规范化：重复标点、缺失括号、
|
||
E 存疑登记：原文不改，逐条附疑正与参校本异文
|
||
F 版本差异/缺文登记：与参校本结构不同处，不径改
|
||
输出：
|
||
02_加工数据/辅行诀脏腑用药法要_清洗版.md
|
||
02_加工数据/原文清洗报告.md （核对数据按篇/条列明）
|
||
02_加工数据/原文人工核对清单.md （按篇列明存疑 + 缺文与版本差异登记）
|
||
02_加工数据/原文结构_上下卷.json （卷 / 篇 / 方 三级 + 行号锚定）
|
||
幂等：可重复运行，输出覆盖生成。
|
||
"""
|
||
import os, re, json, hashlib, sys
|
||
|
||
BASE = os.path.expanduser("~/Documents/ai_agent_scraper_study/data/辅行诀五脏用药法要-药性探真")
|
||
SRC = os.path.join(BASE, "01_来源数据/辅行诀脏腑用药法要_古本原文_20260915.md")
|
||
REF_DIR = os.path.join(BASE, "01_来源数据/校核参考")
|
||
OUT_MD = os.path.join(BASE, "02_加工数据/辅行诀脏腑用药法要_清洗版.md")
|
||
OUT_REPORT = os.path.join(BASE, "02_加工数据/原文清洗报告.md")
|
||
OUT_CHECK = os.path.join(BASE, "02_加工数据/原文人工核对清单.md")
|
||
OUT_STRUCT = os.path.join(BASE, "02_加工数据/原文结构_上下卷.json")
|
||
|
||
REF_NAME = "维基文库本《辅行诀脏腑用药法要》"
|
||
|
||
# C 类：讹字 / 衍字修正（依据＝参校本 + 内证；逐条留痕）
|
||
FIXES = [
|
||
("谨钭五脏", "谨将五脏", f"{REF_NAME}作「谨将」；钭/将形近致误"),
|
||
("耳有所闻", "耳无所闻", f"{REF_NAME}作「耳无所闻」；《素问·脏气法时论》「虚则目䀮䀮无所见，耳无所闻」"),
|
||
("大枣（十二个", "大枣（十二枚", f"{REF_NAME}作「十二枚」；大枣量词作枚"),
|
||
("薯蓣，；当从", "薯蓣，当从", "衍分号"),
|
||
("病心中前", "病心中痛", f"{REF_NAME}作「病心中痛」；《灵枢·五邪》「邪在心，则病心痛」"),
|
||
("崦调之", "而调之", f"{REF_NAME}作「而调之」；崦/而形近致误"),
|
||
("以白浆一斗煮取四或", "以白浆一斗煮取四升", f"{REF_NAME}作「煮取四升」；或/升形近致误（白浆/白酨浆之异文见存疑）"),
|
||
("每服二或", "每服二升", f"{REF_NAME}作「每服二升」；或/升形近致误"),
|
||
("眩中澹澹大动", "心中澹澹大动", f"{REF_NAME}作「心中澹澹大动」；眩/心形近致误"),
|
||
("以不八升", "以水八升", f"{REF_NAME}作「以水八升」；不/水形近致误"),
|
||
("怔惊不发者", "怔惊不安者", f"{REF_NAME}作「不安」；与下条「懊憹不安」同例"),
|
||
("懊憹不发", "懊憹不安", f"{REF_NAME}作「懊憹不安」；发/安形近致误"),
|
||
("殓泄", "飧泄", f"{REF_NAME}作「飧泄」；《素问》「脾实则腹满，飧泄」"),
|
||
("煮取二勤务员，温升再服", "煮取二升，温分再服", f"{REF_NAME}作「煮取二升，温分再服，日二」；OCR混入「勤务员」衍文，升/分互讹"),
|
||
("治烦热汗邮", "治烦热汗出", f"{REF_NAME}作「汗出」；邮/出形近致误"),
|
||
("取之勇泉", "取之涌泉", f"{REF_NAME}作「涌泉」；《灵枢·五邪》「取之涌泉、昆仑」"),
|
||
("肾甘燥", "肾苦燥", f"{REF_NAME}作「肾苦燥」；与「肝苦急、心苦缓、脾苦湿、肺苦气上逆」同例"),
|
||
("时足胫有者方", "时足胫肿者方", f"{REF_NAME}作「时足胫肿」；有/肿形近致误"),
|
||
("加伏龙肝如鸡子在", "加伏龙肝如鸡子大", f"{REF_NAME}作「如鸡子大」；在/大形近致误"),
|
||
("渐分四服", "温分四服", f"{REF_NAME}作「温分四服」；煎服法常语作「温分」"),
|
||
("止方五味", "上方五味", f"{REF_NAME}作「上方五味」；止/上形近致误"),
|
||
("以水五升，称煮椒", "以水五升，先煮椒", f"{REF_NAME}作「先煮椒」；称/先形近致误"),
|
||
("韭叶、区药", "韭叶、芍药", f"{REF_NAME}作「韭叶、芍药」；区/芍形近致误"),
|
||
("温分三他", "温分三服", f"{REF_NAME}作「温分三服」；他/服形近致误"),
|
||
("渐分三他", "温分三服", f"{REF_NAME}作「温分三服」；渐/温、他/服并误"),
|
||
("羸唐如柴", "羸瘦如柴", f"{REF_NAME}作「羸瘦如柴」；唐/瘦形近致误"),
|
||
("竹味（三把）", "竹叶（三把）", f"{REF_NAME}作「竹叶」；味/叶形近致误"),
|
||
("竹味为水", "竹叶为水", f"{REF_NAME}作「竹叶为水」；味/叶形近致误"),
|
||
("五茶为充", "五菜为充", f"{REF_NAME}作「五菜为充」；《素问·脏气法时论》「五菜为充」"),
|
||
("五畜为益，，尔乃", "五畜为益，尔乃", "衍顿号"),
|
||
("补有汤内加羊肝", "补肝汤内加羊肝", f"{REF_NAME}作「补肝汤内加羊肝」；有/肝形近致误"),
|
||
("亦百十十首", "亦百二十首", f"{REF_NAME}作「亦百二十首」；十/二形近致误"),
|
||
("可黩契经方", "可默契经方", f"{REF_NAME}作「可默契经方」；黩/默形近致误"),
|
||
("区药为土", "芍药为土", f"{REF_NAME}作「芍药为土」；区/芍形近致误"),
|
||
("身热之自愈也", "身热去自愈也", f"{REF_NAME}作「身热去自愈」；之/去形近致误"),
|
||
("黄芩（五两）人参桂枝生姜", "黄耆（五两）人参桂枝生姜", f"{REF_NAME}作「黄耆五两」；本篇末「阳旦者，升阳之方，以黄芪为主」自证"),
|
||
("煮取四或，去滓", "煮取四升，去滓", f"{REF_NAME}作「煮取四升」；或/升形近致误"),
|
||
("一斗二升，，煮取六升", "一斗二升，煮取六升", "衍顿号"),
|
||
("治天行热痛", "治天行热病", f"{REF_NAME}作「治天行热病」；本书他篇亦作「天行热病」"),
|
||
("皂角刮去皮絃", "皂角刮去皮弦", f"{REF_NAME}作「皮弦」；絃/弦异体，取通行字"),
|
||
("鸡子黄二枚）", "鸡子黄（二枚）", "夺前括号，据下文「阿胶（三锭）」例补"),
|
||
("内胶，列上火", "内胶，更上火", f"{REF_NAME}作「更上火」；列/更形近致误"),
|
||
("蠃瘦如柴", "羸瘦如柴", f"{REF_NAME}作「羸瘦如柴」；蠃/羸形近致误"),
|
||
("更上炎", "更上火", f"{REF_NAME}作「更上火」；炎/火形近致误"),
|
||
("附了（一枚", "附子（一枚", f"{REF_NAME}作「附子」；了/子形近致误"),
|
||
("少腹中冷嘲热讽", "少腹中冷", f"{REF_NAME}作「少腹中冷」；「嘲热讽」为成语混入之衍文"),
|
||
("日三夜一他", "日三夜一服", f"{REF_NAME}作「日三、夜一服」；他/服形近致误"),
|
||
("以鸡了黄为主", "以鸡子黄为主", f"{REF_NAME}作「鸡子黄」；了/子形近致误"),
|
||
("手足遂冷", "手足逆冷", f"{REF_NAME}作「手足逆冷」；遂/逆形近致误"),
|
||
("扑克其人", "看其人", f"{REF_NAME}作「看其人」；扑克/看形近致误"),
|
||
("得大吐鲁番即愈", "得大吐即愈", f"{REF_NAME}作「得大吐即愈」；「鲁番」为衍文"),
|
||
("少时小便通，愈）", "少时小便通，立愈）", f"{REF_NAME}有「立」字；夺文"),
|
||
(r"枚，炮\]干姜", "枚，炮）干姜", f"原文「\\]」为「）」之误；{REF_NAME}作「附子（一枚，炮）干姜」"),
|
||
("《《金匮要略》", "《金匮要略》", "书名号重复"),
|
||
]
|
||
|
||
# E/F 类：存疑与版本差异（不改原文）
|
||
SUSPECT_RULES = [
|
||
(r'白浆', "参校本作「白酨浆」，本书及集成网本作「白浆」，本书校注亦称「白浆」——异文待定"),
|
||
(r'（）（）为水', "二十五味药精「味辛」条缺字，参校本作「附子（一本此二字不明」；本书缺字未标"),
|
||
(r'视有余者尽取之', "参校本作「视有馀血者尽取之」，疑夺「血」字（《灵枢·五邪》「视有血者尽取之」）"),
|
||
(r'上七味，以水一斗，温服一升', "参校本亦同，疑夺「煮取」句（如「煮取四升」）"),
|
||
(r'绞去滓，顿。', "疑夺「服」字（参校本作「顿服」）"),
|
||
(r'日数传，生死', "参校本作「日数传变」，疑夺「变」字"),
|
||
(r'耎意为祛弱', "本书校注；疑当作「耎意为软弱」（《说文》「耎，弱也」）"),
|
||
(r'气结\n', "泻肺汤方证末夺「方：」（参校本作「气结方。」）"),
|
||
(r'汗出不止\n', "泻肾汤方证末夺「方：」（参校本作「汗出不止方。」）"),
|
||
(r'鸡子黄二枚）阿胶|黄连（四两）芍药（各二两）', "小朱鸟汤、大朱鸟汤药味疑夺「黄芩」（本方煎法作「先煮连、芩、芍三物」；参校本有黄芩）"),
|
||
(r'^弘景曰：“外感天行', "此段起首引号未收（参校本此段不用引号）；原文不改，待核影印本"),
|
||
]
|
||
|
||
# F 类：结构缺题/版本差异（登记，不改）
|
||
VERSION_NOTES = [
|
||
("篇题缺失", "「辨心包络病证文并方」篇题（参校本有；本书与集成网本皆无，心胞条文直接接于心脏病证篇后）"),
|
||
("篇题缺失", "「五脏泻方」篇题（本书以「陶曰：又有泻方五首」起；参校本有篇题）"),
|
||
("篇题缺失", "「二旦六神大小汤」篇题（本书直接以「弘景曰：外感天行……」起；参校本有篇题）"),
|
||
("篇题缺失", "「救五脏中恶卒死方」篇题（本书以「陶经隐居云：中恶卒死者……」起；参校本有篇题）"),
|
||
("编排差异", "心胞补方编排：本书「小补心汤」＝治胸痹心中痞满（栝蒌薤白半夏厚朴桂枝）；参校本作「大补心汤」，另立「小补心汤」＝治胸痹不得卧（栝蒌薤白半夏）"),
|
||
("内容节略", "大泻心汤煎服法节略：参校本有「上六味，以酢六升，先煮上五味，得三升许，去滓。内戎盐…」全法，本书仅存「上六味，服一升，当大吐……」"),
|
||
("图版缺失", "《汤液经法图》及其图注（「左阳进为补，其数七，火数也；右阴退为泻，其数六，水数也」）本书无图无注；参校本存图（SVG）与图注"),
|
||
("异体不校", "着/著（着舌、着散、着汤）、纳/内、郄/郤、憹/𢙐、芪/耆 等异体字，两本互见，本阶段不校"),
|
||
]
|
||
|
||
|
||
def clean(text):
|
||
lines = text.split("\n")
|
||
events, out = [], []
|
||
for idx, raw in enumerate(lines, start=1):
|
||
line = raw.replace("\u3000", "").rstrip() # A 全角缩进
|
||
if line != "" and not re.match(r'^梁[·・]?华阳隐居', line):
|
||
prev = None
|
||
while prev != line:
|
||
prev = line
|
||
line = re.sub(r'(?<=[\u4e00-\u9fff])\s+(?=[\u4e00-\u9fff])', '', line)
|
||
line = re.sub(r'(?<=[\u4e00-\u9fff])\s+(?=[，。、；：？！“”（）《》])', '', line)
|
||
line = re.sub(r'(?<=[，。、；：？！“”（）《》])\s+(?=[\u4e00-\u9fff])', '', line)
|
||
# B 校勘符号
|
||
b = re.sub(r'\\\[([^\]\\]*)\\\]', r'〔\1〕', line)
|
||
if b != line:
|
||
events.append(dict(行=idx, 类型="B校勘符号", 原="\\[…\\]", 改="〔…〕",
|
||
依据="Markdown 转义残留 → 传统校补号"))
|
||
line = b
|
||
# C 讹字
|
||
for bad, good, why in FIXES:
|
||
if bad != good and bad in line:
|
||
events.append(dict(行=idx, 类型="C讹字修正", 原=bad, 改=good, 依据=why))
|
||
line = line.replace(bad, good)
|
||
# D 标点
|
||
d = re.sub(r'(，，|；；|。。)', lambda m: m.group(1)[0], line)
|
||
if d != line:
|
||
events.append(dict(行=idx, 类型="D标点", 原="重复标点", 改="单一标点", 依据="重复标点衍文"))
|
||
line = d
|
||
out.append(line)
|
||
cleaned = re.sub(r'\n{3,}', '\n\n', "\n".join(out)).strip() + "\n"
|
||
# E 存疑登记（在清洗后文本上定位）
|
||
suspects = []
|
||
clines = cleaned.split("\n")
|
||
for i, l in enumerate(clines, start=1):
|
||
for pat, why in SUSPECT_RULES:
|
||
if re.search(pat, l):
|
||
suspects.append({"行": i, "片段": l[:60], "说明": why})
|
||
seen, uniq = set(), []
|
||
for s in suspects:
|
||
k = (s["行"], s["说明"])
|
||
if k not in seen:
|
||
seen.add(k); uniq.append(s)
|
||
return cleaned, events, uniq
|
||
|
||
|
||
SECTIONS = [
|
||
(r'^辅行诀脏腑用药法要$', "上卷·总题"),
|
||
(r'^梁·华阳隐居', "上卷·撰者"),
|
||
(r'^辨肝脏病证文并方', "上卷·辨肝脏病证文并方"),
|
||
(r'^辨心脏病证文并方', "上卷·辨心脏病证文并方"),
|
||
(r'^心胞气实者', "上卷·心胞病证文（篇题缺）"),
|
||
(r'^辨脾脏病证文并方', "上卷·辨脾脏病证文并方"),
|
||
(r'^辨肺脏病证候文并方', "上卷·辨肺脏病证候文并方"),
|
||
(r'^辨肾脏病证文并方', "上卷·辨肾脏病证文并方"),
|
||
(r'^陶曰：又有泻方五首', "上卷·五脏泻方（篇题缺）"),
|
||
(r'^陶云：经方有救诸劳损病方', "下卷·救五脏诸劳损病方（引）"),
|
||
(r'^辅行诀脏腑用药法要（下）', "下卷·总题"),
|
||
(r'^养生补肝汤', "下卷·救五脏诸劳损病方"),
|
||
(r'^陶云：经云：毒药攻邪', "下卷·汤液经法（引）"),
|
||
(r'^味辛皆属木', "下卷·汤液经法（二十五味药精）"),
|
||
(r'^弘景曰：“外感天行|^弘景曰：外感天行', "下卷·二旦六神大小汤（篇题缺）"),
|
||
(r'^陶经隐居云：中恶卒死者', "下卷·救五脏中恶卒死方（篇题缺）"),
|
||
(r'^校注：$', "下卷·校注"),
|
||
]
|
||
|
||
|
||
def parse_struct(text):
|
||
lines = text.split("\n")
|
||
nodes = []
|
||
for i, l in enumerate(lines, start=1):
|
||
s = l.strip()
|
||
for pat, name in SECTIONS:
|
||
if re.match(pat, s):
|
||
nodes.append({"篇": name, "起始行": i})
|
||
break
|
||
for n, node in enumerate(nodes):
|
||
node["结束行"] = nodes[n + 1]["起始行"] - 1 if n + 1 < len(nodes) else len(lines)
|
||
node["条文数"] = sum(1 for i in range(node["起始行"], node["结束行"] + 1)
|
||
if lines[i - 1].strip())
|
||
return nodes
|
||
|
||
|
||
def main():
|
||
raw = open(SRC, encoding="utf-8").read()
|
||
cleaned, events, suspects = clean(raw)
|
||
nodes = parse_struct(cleaned)
|
||
|
||
def wf(path, content):
|
||
os.makedirs(os.path.dirname(path), exist_ok=True)
|
||
with open(path, "w", encoding="utf-8") as f:
|
||
f.write(content); f.flush(); os.fsync(f.fileno())
|
||
|
||
wf(OUT_MD, cleaned)
|
||
struct = {
|
||
"来源文件": os.path.basename(SRC), "来源文件md5": hashlib.md5(raw.encode()).hexdigest(),
|
||
"清洗版文件": os.path.basename(OUT_MD), "清洗版md5": hashlib.md5(cleaned.encode()).hexdigest(),
|
||
"参校本": REF_NAME,
|
||
"说明": "行号均指清洗版；本文件仅作清理配套定位索引，未做释义抽取与跨库关联",
|
||
"篇": nodes,
|
||
}
|
||
wf(OUT_STRUCT, json.dumps(struct, ensure_ascii=False, indent=2) + "\n")
|
||
|
||
def owner(line):
|
||
for n in nodes:
|
||
if n["起始行"] <= line <= n["结束行"]:
|
||
return n["篇"]
|
||
return "（篇外）"
|
||
per = {}
|
||
for e in events:
|
||
per.setdefault(owner(e["行"]), []).append(e)
|
||
per_sus = {}
|
||
for s in suspects:
|
||
per_sus.setdefault(owner(s["行"]), []).append(s)
|
||
|
||
cnt = lambda t: sum(1 for e in events if e["类型"].startswith(t))
|
||
rep = ["# 《辅行诀脏腑用药法要》古本原文 — 清洗报告", "",
|
||
"> 本库现阶段只做数据清理。本文本作为全书**原文依据参考**入库，未做释义抽取、"
|
||
"跨库关联与主题分析（用户 2026-09-15 指令）。", "",
|
||
"## 0 概览", "",
|
||
f"- 来源：`01_来源数据/{os.path.basename(SRC)}`（md5 `{struct['来源文件md5']}`）",
|
||
f"- 清洗版：`02_加工数据/{os.path.basename(OUT_MD)}`（md5 `{struct['清洗版md5']}`）",
|
||
f"- 参校本：{REF_NAME}（另存 `01_来源数据/校核参考/`，仅作清理核校用）",
|
||
f"- 字符数：{len(raw)} → {len(cleaned)}；行数：{raw.count(chr(10))} → {cleaned.count(chr(10))}",
|
||
f"- 结构：上卷 + 下卷，共 {len(nodes)} 个可识别篇段", "",
|
||
"| 清理类别 | 处数 | 说明 |", "|---------|------|------|",
|
||
f"| A 版式残留 | 全量 | 全角缩进（{sum(1 for l in raw.split(chr(10)) if l.startswith(chr(0x3000)))} 行）、行尾空格、词间半角空格 |",
|
||
f"| B 校勘符号规范化 | {cnt('B')} | `\\[X\\]` → `〔X〕`（补字保留语义，只去 Markdown 转义）|",
|
||
f"| C 讹字/衍字修正 | {cnt('C')} | 每条附核校依据（参校本 + 内证），见下方逐篇明细 |",
|
||
f"| D 标点规范化 | {cnt('D')} | 重复标点、缺失括号 |",
|
||
f"| E 存疑（未改原文） | {len(suspects)} | 逐条附疑正与参校本异文 |",
|
||
f"| F 版本差异/缺文（未改） | {len(VERSION_NOTES)} | 结构、编排、图版差异，不作校改 |", "",
|
||
"### 各篇清理量一览", "",
|
||
"| 篇 | 行范围 | 条文行 | B | C | D | E 存疑 |", "|----|--------|--------|---|---|---|--------|"]
|
||
for n in nodes:
|
||
evs = per.get(n["篇"], [])
|
||
c = lambda t: sum(1 for e in evs if e["类型"].startswith(t))
|
||
rep.append(f"| {n['篇']} | {n['起始行']}–{n['结束行']} | {n['条文数']} | {c('B')} | {c('C')} | {c('D')} | "
|
||
f"{len(per_sus.get(n['篇'], []))} |")
|
||
|
||
rep += ["", "---", "", "## 1 逐篇核对明细（按卷 / 篇列明）", ""]
|
||
for n in nodes:
|
||
evs = per.get(n["篇"], [])
|
||
sus = per_sus.get(n["篇"], [])
|
||
rep += [f"### {n['篇']}（清洗版行 {n['起始行']}–{n['结束行']}）", "",
|
||
f"- 清理量：B {sum(1 for e in evs if e['类型'].startswith('B'))} / "
|
||
f"C {sum(1 for e in evs if e['类型'].startswith('C'))} / "
|
||
f"D {sum(1 for e in evs if e['类型'].startswith('D'))} / 存疑 {len(sus)}", ""]
|
||
if evs:
|
||
rep += ["| 行 | 类型 | 原文 | 处理后 | 核校依据 |", "|----|------|------|--------|---------|"]
|
||
for e in evs:
|
||
rep.append(f"| {e['行']} | {e['类型']} | {e['原'].replace('|','｜')} | "
|
||
f"{e['改'].replace('|','｜')} | {e['依据']} |")
|
||
rep.append("")
|
||
if sus:
|
||
rep += ["存疑项：", "", "| 行 | 片段 | 说明 |", "|----|------|------|"]
|
||
for s in sus:
|
||
rep.append(f"| {s['行']} | {s['片段'].replace('|','｜')} | {s['说明']} |")
|
||
rep.append("")
|
||
if not evs and not sus:
|
||
rep += ["- 本篇无需修正条目。", ""]
|
||
|
||
rep += ["---", "", "## 2 版本差异 / 缺文登记（不校改）", "",
|
||
"| 类别 | 说明 |", "|------|------|"]
|
||
for k, v in VERSION_NOTES:
|
||
rep.append(f"| {k} | {v} |")
|
||
rep += ["", "## 3 篇段结构（清洗版行号）", "",
|
||
"| 卷·篇 | 起始行 | 结束行 | 条文行 |", "|-------|--------|--------|--------|"]
|
||
for n in nodes:
|
||
rep.append(f"| {n['篇']} | {n['起始行']} | {n['结束行']} | {n['条文数']} |")
|
||
wf(OUT_REPORT, "\n".join(rep) + "\n")
|
||
|
||
chk = ["# 原文人工核对清单 — 《辅行诀脏腑用药法要》古本原文", "",
|
||
"> 用途：需对照纸质原书／敦煌本影印本确认的条目。**按卷/篇列明**，本阶段不改动清洗版正文。",
|
||
"> 参校本：维基文库本《辅行诀脏腑用药法要》（存 `01_来源数据/校核参考/`）。", ""]
|
||
for n in nodes:
|
||
sus = per_sus.get(n["篇"], [])
|
||
chk += [f"## {n['篇']}（行 {n['起始行']}–{n['结束行']}）", ""]
|
||
if sus:
|
||
chk += ["| 行 | 片段 | 疑正 / 参校本异文 |", "|----|------|------------------|"]
|
||
for s in sus:
|
||
chk.append(f"| {s['行']} | {s['片段'].replace('|','｜')} | {s['说明']} |")
|
||
else:
|
||
chk.append("- 无存疑项。")
|
||
chk.append("")
|
||
chk += ["---", "", "## 版本差异 / 缺文登记（不校改，待核影印本）", "",
|
||
"| 类别 | 说明 |", "|------|------|"]
|
||
for k, v in VERSION_NOTES:
|
||
chk.append(f"| {k} | {v} |")
|
||
chk += ["", "## 书目与来源", "",
|
||
"- 文本：《辅行诀脏腑用药法要》（梁·华阳隐居 陶弘景撰；敦煌遗书系传本）",
|
||
"- 本次来源：`01_来源数据/辅行诀脏腑用药法要_古本原文_20260915.md`（用户 2026-09-15 提供）",
|
||
"- 参校本：维基文库本（另有集成网本，与本来源同系，讹字位置一致）",
|
||
"- 参校本快照：`01_来源数据/校核参考/`"]
|
||
wf(OUT_CHECK, "\n".join(chk) + "\n")
|
||
|
||
print("清洗版:", OUT_MD, os.path.getsize(OUT_MD), "bytes")
|
||
print("结构 :", OUT_STRUCT, os.path.getsize(OUT_STRUCT), "bytes")
|
||
print("报告 :", OUT_REPORT, os.path.getsize(OUT_REPORT), "bytes")
|
||
print("核对单:", OUT_CHECK, os.path.getsize(OUT_CHECK), "bytes")
|
||
print(f"B {cnt('B')} | C {cnt('C')} | D {cnt('D')} | 存疑 {len(suspects)} | 版本差异 {len(VERSION_NOTES)}")
|
||
for n in nodes:
|
||
evs = per.get(n["篇"], [])
|
||
print(f" {n['篇']} 行{n['起始行']}–{n['结束行']} 条文{n['条文数']} 清理项 {len(evs)}")
|
||
|
||
|
||
if __name__ == "__main__":
|
||
main()
|