Simon
112 lines
3.2 KiB
Python
112 lines
3.2 KiB
Python
# -*- coding: utf-8 -*-
|
|
"""
|
|
阶段5：将 docs 内 .md 对旧结构路径的引用改写为新结构路径。
|
|
映射来源（全部已验证）：
|
|
- A类 content_dup: old -> new
|
|
- B1类 name_diff(排除b2): old -> new
|
|
- b2: docs_repair_b2.py 的 MOVES（改名迁入目标）
|
|
- C类: docs_repair_c.py 的 M（逐文件归类目标）
|
|
旧路径统一用 '/' 匹配，替换也用 '/'。
|
|
"""
|
|
import os, json, re, sys
|
|
|
|
ROOT = "docs"
|
|
DRY = "--go" not in sys.argv
|
|
|
|
# ---- 复用已验证的映射 ----
|
|
def load_mod(path, stop_marker):
|
|
src = open(path, encoding='utf-8').read()
|
|
seg = src.split(stop_marker)[0]
|
|
ns = {}
|
|
exec(compile(seg, path, 'exec'), ns)
|
|
return ns
|
|
|
|
b2ns = load_mod('docs_repair_b2.py', 'ok, skip = [], []')
|
|
MOVES = {k: v for k, v in b2ns['MOVES']}
|
|
cns = load_mod('docs_repair_c.py', 'print("映射条目')
|
|
M_C = cns['M']
|
|
|
|
d = json.load(open('docs_cmp_result2.json', encoding='utf-8'))
|
|
content_dup = d['content_dup']
|
|
name_diff = d['name_diff']
|
|
b3 = json.load(open('docs_cmp_result3.json', encoding='utf-8'))
|
|
b2_old = set(row[0] for row in b3.get('b2', []))
|
|
|
|
full = {}
|
|
for old, new in content_dup:
|
|
full[old] = new
|
|
for row in name_diff:
|
|
old, new = row[0], row[3]
|
|
if old in b2_old:
|
|
continue
|
|
full[old] = new
|
|
for old, new in MOVES.items():
|
|
full[old] = new
|
|
for old, new in M_C.items():
|
|
full[old] = new
|
|
|
|
# 统一成 '/' 分隔的键
|
|
norm_map = {}
|
|
for old, new in full.items():
|
|
ko = old.replace('\\', '/')
|
|
kn = new.replace('\\', '/')
|
|
norm_map[ko] = kn
|
|
|
|
# 按键长降序，避免短键先匹配导致部分替换
|
|
keys = sorted(norm_map.keys(), key=len, reverse=True)
|
|
|
|
def rewrite_text(text):
|
|
total = 0
|
|
for ko in keys:
|
|
kn = norm_map[ko]
|
|
# 同时匹配 / 与 \ 两种分隔
|
|
for variant in (ko, ko.replace('/', '\\')):
|
|
if variant in text:
|
|
cnt = text.count(variant)
|
|
text = text.replace(variant, kn)
|
|
total += cnt
|
|
return text, total
|
|
|
|
# 扫描所有 .md
|
|
md_files = []
|
|
for root, _, fs in os.walk(ROOT):
|
|
for f in fs:
|
|
if f.endswith('.md'):
|
|
md_files.append(os.path.join(root, f))
|
|
|
|
changed_files = []
|
|
total_repl = 0
|
|
samples = []
|
|
for fp in md_files:
|
|
with open(fp, encoding='utf-8') as fh:
|
|
text = fh.read()
|
|
new_text, cnt = rewrite_text(text)
|
|
if cnt:
|
|
changed_files.append(fp)
|
|
total_repl += cnt
|
|
if len(samples) < 8:
|
|
# 找一处样例
|
|
for ko in keys:
|
|
if ko.replace('\\','/') in text or ko.replace('/','\\') in text:
|
|
samples.append((fp, ko, norm_map[ko]))
|
|
break
|
|
|
|
print("映射条目:", len(norm_map))
|
|
print("扫描 .md 文件:", len(md_files))
|
|
print("将修改文件数:", len(changed_files))
|
|
print("替换引用处数:", total_repl)
|
|
print("\n样例:")
|
|
for fp, ko, kn in samples:
|
|
print(" %s\n %s -> %s" % (fp, ko, kn))
|
|
|
|
if DRY:
|
|
print("\n[DRY] 未写入。")
|
|
else:
|
|
for fp in changed_files:
|
|
with open(fp, encoding='utf-8') as fh:
|
|
text = fh.read()
|
|
new_text, _ = rewrite_text(text)
|
|
with open(fp, 'w', encoding='utf-8') as fh:
|
|
fh.write(new_text)
|
|
print("\n已写入", len(changed_files), "个文件。")
|