# -*- coding: utf-8 -*-
# t08-extract.py — A4 关键页批注区提取
import io, sys
sys.stdout = io.TextIOWrapper(sys.stdout.buffer, encoding='utf-8')

base = '.scratch/opportunity-bugfix/lanhu-pages/'
files = [
    'A4-3-3-1_新增客户信息.md',          # 客户主档字段集
    'A4-1-1_客户公海列表（列表视图）.md',  # 公海机制
    'A4-6-3_客户合并.md',                # 合并深水区
    'A4-7-1_客户交割（发起交接）.md',      # 交割深水区
]
for fn in files:
    lines = open(base + fn, 'r', encoding='utf-8').read().splitlines()
    print(f'\n===== {fn} ({len(lines)} 行) =====')
    idx = [i for i, ln in enumerate(lines) if '、开发说明' in ln or '【页面名称】' in ln]
    seen = set()
    printed = 0
    for i in idx:
        for j in range(i, min(i + 5, len(lines))):
            if j in seen:
                continue
            seen.add(j)
            ln = lines[j]
            if len(ln) > 400:  # 大单元格=批注内容
                print(ln[:6000])
                printed += 1
                break
    if not printed:
        print('（无大单元格批注）')
print('\ndone')