def get_during_keywords(topic, start_ts, end_ts): #关键词云,unit=MinInterval
keywords = []
# if (end_ts-start_ts)>unit:
# begin_ts = end_ts-unit
# else:
# begin_ts = start_ts
# print begin_ts,end_ts
query_body = {
'query': {
'filtered': {
'filter': {
'range': {
'timestamp': {
'gte': start_ts,
'lt': end_ts
}
}
}
}
},
'size': MAX_LANGUAGE_WEIBO
}
keywords_dict = {}
weibo_text = []
keyword_weibo = weibo_es.search(index=topic,
doc_type=weibo_index_type,
body=query_body)['hits']['hits']
for key_weibo in keyword_weibo:
weibo_text.append(key_weibo['_source']['text'].encode('utf-8'))
keywords_dict = get_weibo(weibo_text, n_gram=2, n_count=100)
'''
print keyword_weibo
for key_weibo in keyword_weibo:
keywords_dict_list = json.loads(key_weibo['_source']['keywords_dict']) #
#print keywords_dict_list,type(keywords_dict_list)
for k,v in keywords_dict_list.iteritems():
try:
keywords_dict[k] += v
except:
keywords_dict[k] = v
'''
word_results = sorted(keywords_dict.iteritems(),
key=lambda x: x[1],
reverse=True)[:MAX_FREQUENT_WORDS]
return json.dumps(word_results)
def get_during_keywords(topic,start_ts,end_ts): #关键词云,unit=MinInterval
keywords = []
# if (end_ts-start_ts)>unit:
# begin_ts = end_ts-unit
# else:
# begin_ts = start_ts
# print begin_ts,end_ts
query_body = {
'query':{
'filtered':{
'filter':{
'range':{
'timestamp':{'gte': start_ts, 'lt':end_ts}
}
}
}
},
'size':MAX_LANGUAGE_WEIBO
}
keywords_dict = {}
weibo_text = []
keyword_weibo = weibo_es.search(index=topic,doc_type=weibo_index_type,body=query_body)['hits']['hits']
for key_weibo in keyword_weibo:
weibo_text.append(key_weibo['_source']['text'].encode('utf-8'))
keywords_dict = get_weibo(weibo_text,n_gram=2,n_count=100)
'''
print keyword_weibo
for key_weibo in keyword_weibo:
keywords_dict_list = json.loads(key_weibo['_source']['keywords_dict']) #
#print keywords_dict_list,type(keywords_dict_list)
for k,v in keywords_dict_list.iteritems():
try:
keywords_dict[k] += v
except:
keywords_dict[k] = v
'''
word_results = sorted(keywords_dict.iteritems(),key=lambda x:x[1],reverse=True)[:MAX_FREQUENT_WORDS]
return json.dumps(word_results)