import json, re

d = json.load(open('/data/news/json/yes_data/0926data.json'))
if isinstance(d, list):
    d = [{'name': s['name'], 'list': s['list']} for s in d]

def ncontent(s):
    return len(re.sub(r'[^\w]', '', str(s), flags=re.UNICODE))

print('=== 标题 ===')
for sec in d:
    for i, e in enumerate(sec['list']):
        t = str(e.get('title', '') or '')
        n = ncontent(t)
        flags = []
        if n > 35: flags.append(f'⚠️{n}字超限')
        if '…' in t or '...' in t: flags.append('⚠️含省略号')
        print(f"  [{sec['name']}][{i}] {n:>2}字 {' '.join(flags):<14} {t}")

print()
print('=== 摘要 ===')
for sec in d:
    for i, e in enumerate(sec['list']):
        s = str(e.get('summary', '') or '')
        n = ncontent(s)
        flags = []
        if n > 45: flags.append(f'⚠️{n}字>45')
        if n > 100: flags.append('⚠️疑似全文')
        if n <= 21 and not re.search(r'\d', s): flags.append('⚠️口号化?')
        print(f"  [{sec['name']}][{i}] {n:>3}字 {' '.join(flags):<16} {s}")