print 'error5', link
#抓新聞按讚數
res = requests.get(
'http://api.facebook.com/restserver.php?method=links.getstats&format=json&urls=%s'
% link)
data = json.loads(res.text)[0]
like_count = data['like_count']
share_count = data['share_count']
#抓新聞的 comments
try:
comments = function.get_fb_comments(data['comments_fbid'])
except Exception as e:
comments = []
print '抓取fb_comments錯誤', e
#提取文本關鍵字
keywords = function.keyword_extract(content)
#存進 mongodb
doc = {
'author': '壹電視新聞',
'date': date,
'title': title,
'content': content,
'href': href,
'share_count': share_count,
'like_count': like_count,
'comments': comments,
'keywords': keywords
}
collect.insert_one(doc)
#存進 pgdb
function.keywords_insert_pgdb(keywords)
date = post['created_time'].replace('T',' ')
date = date.split('+')[0]
post_one_data['date'] = date #創建時間 (UTC+00)
print post_one_data['date']
try:
post_one_data['share_count'] = post['shares']['count'] #分享數
except Exception as e:
post_one_data['share_count'] = 0
print '沒有分享數',post_id
post_one_data['like_count'] = get_like_count(post_id)
#post_one_data['likes'] = get_like_list(post_id)
#post_one_data['shared'] = get_shared_list(post_id) #分享名單
post_one_data['comments'] = get_comment_list(post_id)
#提取文本關鍵字
keywords = function.keyword_extract(post_one_data['message'])
post_one_data['keywords'] = keywords
#存進 mongodb
collect.insert_one(post_one_data)
#存進 pgdb
function.keywords_insert_pgdb(keywords)
function.kw_relation_insert_pgdb(keywords)
function.doc_insert_pgdb(post_one_data,18,3) #doc,source,big_source
function.doc_join_kw_insert_pgdb(keywords,post_one_data['href'])
function.daily_kw_insert_pgdb(keywords,post_one_data['date'],18) #keywords,date,source_fk
function.fb_doc_relation_keyword(post_one_data['href'],page['id']) #某粉絲團貼文與該粉絲團關聯一起
#更新貼文
for post_id in crawled_post_ids:
res = requests.get('https://graph.facebook.com/v2.3/%s?access_token=%s'%(post_id,token))
post = json.loads(res.text)
print resu[0][0], resu[0][1]
print '-------------------------------------------------'
'''
for o in combi(a):
print o
print len(combi(a))
print type(combi(a))
'''
# In[ ]:
import requests
import json
import function
for k in function.keyword_extract('蔡英文明訪日 日本李登輝之友會協辦晚宴'):
print k
res = requests.get(
'https://ajax.googleapis.com/ajax/services/search/web?v=1.0&q=的')
a = json.loads(res.text)
print a['responseData']['cursor']['estimatedResultCount']
# In[1]:
#兩文本相似度
#!/usr/bin/env python
# -*- coding: utf-8 -*-
from copy import deepcopy
# 作業系統
import os
