result = es_cluster.search(index=index_name, doc_type="bci", body=query_body)['hits']['hits']
sensitive_uid = []
for item in result:
sensitive_uid.append(item['_source']['uid'])
return sensitive_uid
if __name__ == '__main__':
'''
f = open('sensitive_uid_list.txt', 'wb')
uid_list = search_sensitive_weibo('20130904')
for uid in uid_list:
f.write(str(uid) + '\n')
f.close()
'''
f = open('sensitive_uid_list.txt', 'rb')
for line in f:
uid = line.strip()
try:
result = es_cluster.get(index='sensitive_user_portrait', doc_type='user', id=uid)['_source']
except:
print uid
continue
if result['sensitive_words_string']:
es.update(index='sensitive_user_portrait', doc_type='user', id=uid, body={"doc":{"type":1}})
else:
es.update(index='sensitive_user_portrait', doc_type='user', id=uid, body={"doc":{"type":0}})
result = dict()
results = dict()
try:
for i in range(1,13):
item = line[i].split('*')
number = item[0].split('E')
result[item[1]] = int(number[1][1:])
except:
print number, count, uid
continue
sort_list = sorted(result.items(), key=lambda x:x[1], reverse=False)
results['domain'] = domain_dict[sort_list[0][0]]
action = {"update": {'_id': uid}}
bulk_action.extend([action, {'doc': results}])
if count % 100 == 0:
es.bulk(bulk_action, index='sensitive_user_portrait', doc_type='user', timeout=60)
bulk_action = []
print count
if bulk_action:
es.bulk(bulk_action, index='sensitive_user_portrait', doc_type='user', timeout=60)
'''
'''
es.delete(index= 'custom_attribute', doc_type='attribute', id='AVBa32QMGk0Kt7GIxCQW')
es.delete(index= 'custom_attribute', doc_type='attribute', id='AVBa32rwGk0Kt7GIxCQX')
es.delete(index= 'custom_attribute', doc_type='attribute', id='AVBa4Qx7y3nG3t1ED6gr')
'''
#es.update(index="sensitive_user_portrait", doc_type="user", id=1408848023, body={"doc": {"domain": "公知分子"}})
es.update(index="sensitive_user_portrait", doc_type="user", id=1892680725, body={"doc":{"domain": "公职人员"}})
