# -*- coding: utf-8 -*-
# t08-fields.py — A4-3-3-1 新增客户表单字段标签提取
import io, sys, re
sys.stdout = io.TextIOWrapper(sys.stdout.buffer, encoding='utf-8')

fn = '.scratch/opportunity-bugfix/lanhu-pages/A4-3-3-1_新增客户信息.md'
lines = open(fn, 'r', encoding='utf-8').read().splitlines()
# 抓「字段名：」式标签与输入类控件的短文本行
pats = re.compile(r'(\*\s*[\u4e00-\u9fa5A-Za-z0-9/]{2,12}|[\u4e00-\u9fa5A-Za-z0-9/]{2,12}：)')
seen = []
for ln in lines:
    m = re.match(r'\| u\d+ \| (矩形|单元格|按钮|文本框|下拉框|单选按钮|复选框) \| Unnamed \| `?(.+?)`? \|$', ln.strip())
    if not m:
        continue
    txt = m.group(2).strip()
    if not txt or len(txt) > 25 or txt.startswith('itc') or txt in ('Unnamed',):
        continue
    if any(k in txt for k in ('管理','首页','菜单','工作台','收起','商机','线索','看板','系统设置','进入后台','项目','公海','我的','联系人20','联系人10','联系人8')):
        continue
    seen.append((m.group(1), txt))
# 去重保序
out, done = [], set()
for t, s in seen:
    if s not in done:
        done.add(s)
        out.append((t, s))
for t, s in out:
    print(f'{t}\t{s}')
print('TOTAL', len(out))