import json
d = json.load(open('02_加工数据/参校本对照_药目核对.json', encoding='utf-8'))
det = d['异文明细']
BOILER = {'治', '无毒', '有毒', '小', '微', '寒', '大寒'}
substantive = {}
for name, frags in det.items():
    real = []
    for f in frags:
        s, w = f['孙'], f['维基']
        if s == '' and w in BOILER:
            continue
        if w == '' and s in BOILER:
            continue
        real.append(f)
    if real:
        substantive[name] = real
print('实字异文条目(粗):', len(substantive), '/ 总异文', len(det))
for n in list(substantive)[:25]:
    line = '; '.join("孙[%s]→维[%s]" % (f['孙'], f['维基']) for f in substantive[n][:5])
    print('■', n, '|', line[:140])
json.dump(substantive, open('02_加工数据/_tmp_实字异文.json', 'w', encoding='utf-8'),
          ensure_ascii=False, indent=1)
print('已写 02_加工数据/_tmp_实字异文.json')