def crawler_twitter_init():
config = read_config()
print('<-----初始化程序----->')
twitter_crawler_queue = RedisQueue(name='twitter_users',
redis_config=config['redis_config'])
if twitter_crawler_queue.qsize() > 0:
print('<-----有%s个任务还未完成---->' % twitter_crawler_queue.qsize())
if twitter_crawler_queue.empty():
with open(os.path.abspath('twitter_user_ids.json'), 'r') as f:
user_ids = json.load(f)
for id in user_ids['ids']:
twitter_crawler_queue.put(id)
print('<-----有%s个任务需要完成----->' % twitter_crawler_queue.qsize())
print('<-----twitter初始化完成----->')
def crawler_init(name='twitter'):
print('<-----初始化程序----->')
if name == 'twitter':
twitter_crawler_queue = RedisQueue(name='twitter')
if twitter_crawler_queue.qsize() == 0:
with open(os.path.abspath('twitter_user_ids.json'), 'r') as f:
user_ids = json.load(f)
for id in user_ids['ids']:
twitter_crawler_queue.put(id)
else:
facebook_crawler_queue = RedisQueue(name='facebook')
if facebook_crawler_queue.qsize() == 0:
with open(os.path.abspath('facebook_user_ids.json'), 'r') as f:
user_ids = json.load(f)
for id in user_ids['ids']:
facebook_crawler_queue.put(id)
print('<-----初始化成功------>')
def update_facebook_users_count(self, crawler): #facebook 用户账户的like_num字段
print('【爬取twitter usersInfos】')
db = self.create_mongo_conn(db='FaceBook', collection='facebook')
weipa_count = 1
while True:
try:
facebook_users_queue = RedisQueue(
name='facebook_users',
redis_config=self.app_config['redis_config'])
print('[===未抓取的的id个数为:%s===]' % facebook_users_queue.qsize())
if facebook_users_queue.empty():
if weipa_count >= 3:
print('<-----用户抓取完成----->')
break
else:
weipa_count += 1
print('[==Retry:%s==]' % (weipa_count - 1))
time.sleep(5)
continue
id = facebook_users_queue.get()
tweet = db.find_one({"id": id})
print(tweet)
if not tweet:
continue
if tweet['link'].endswith('/'):
url = tweet['link'] + 'community/'
else:
url = tweet['link'] + '/' + 'community/'
content = crawler.crawler_user_likes(url)
for item in content:
# print(item)
if "isLoginStatus" in item:
update_doc = db.find_one_and_update(
{"id": id}, {
'$set': {
'likes_num': item['like_count'],
'fan_count': item['fan_count'],
"update_time": datetime.now(),
"isLoginStatus": item["isLoginStatus"]
}
},
return_document=ReturnDocument.AFTER)
print('更新了%s文档成功' % update_doc['id'])
else:
update_doc = db.find_one_and_update(
{"id": id}, {
'$set': {
'likes_num': item['like_count'],
'fan_count': item['fan_count'],
"update_time": datetime.now(),
}
},
return_document=ReturnDocument.AFTER)
print('更新了%s文档成功' % update_doc['id'])
weipa_count = 1
except Exception as e:
raise e
def check_queue_isEmpty():
try:
queue = RedisQueue(name='twitter_replay',
redis_config=app_config['redis_config'])
if queue.qsize() > 0:
twitter_every_day_update_count_job()
else:
print('[没有任务！！]')
return
except Exception as e:
print(e)
return
def crawler_init(name='twitter'):
print('<-----初始化程序----->')
config = read_config()
if name == 'twitter':
twitter_crawler_queue = RedisQueue(name='twitter',
redis_config=config['redis_config'])
if twitter_crawler_queue.qsize() > 0:
print('<-----有%s个任务还未完成---->' % twitter_crawler_queue.qsize())
if twitter_crawler_queue.empty():
with open(os.path.abspath('twitter_user_ids.json'), 'r') as f:
user_ids = json.load(f)
for id in user_ids['ids']:
twitter_crawler_queue.put(id)
# print(id)
print('<-----有%s个任务需要完成----->' % twitter_crawler_queue.qsize())
print('<-----twitter初始化完成----->')
else:
facebook_crawler_queue = RedisQueue(
name='facebook', redis_config=config['redis_config'])
if facebook_crawler_queue.qsize() > 0:
print('<-----有%s个任务还未完成---->' % facebook_crawler_queue.qsize())
if facebook_crawler_queue.empty():
with open(os.path.abspath('facebook_user_ids.json'), 'r') as f:
user_ids = json.load(f)
for id in user_ids['ids']:
facebook_crawler_queue.put(id)
print('<-----有%s个任务需要完成----->' % facebook_crawler_queue.qsize())
print('<-----facebook初始化完成----->')
def completion_twitter_text(conn):
# with open('twitter_user_ids.json','r') as f:
# ids = json.load(f)
current = 0;
# for id in ids['ids']:
with open(os.path.abspath('config.json'), 'r') as f:
app_config = json.load(f)
twitter_crawler_queue = RedisQueue(name='twitter_test_ids', redis_config=app_config['redis_config2'])
while True:
try:
# print(id)
if twitter_crawler_queue.empty():
print('所有的用户全部跑完')
break
id = twitter_crawler_queue.get()
print('[===取出的ID为%s,目前还有%s条需要抓取====]' % (id,twitter_crawler_queue.qsize()))
es_url = 'http://narnia.idatage.com/stq/api/v1/rowlet/findEsTextByUserIdOrKeywords?startDate=2016-12-30&endDate=2018-02-12&category=tw&ids=%s' % (id,) # if i in [1,2,3,4,5,6,7,8] else 'http://narnia.idatage.com/stq/api/v1/rowlet/findEsTextByUserIdOrKeywords?startDate=2017-%s-01&endDate=2017-%s-01&category=%s&ids=%s' % (i,i+1,category,id)
es_body = requests.get(es_url)
# print('取出的内容为：')
# print(es_body.text)
es_body_tw = json.loads(es_body.text)['tw']
print(len(es_body_tw))
def handele(x):
# print(x)
x['_source']['index_name'] = x['_index']
x['_source']['type_name'] = x['_type']
x['_source']['id']=x['_id']
x['_source']['url'] = 'https://twitter.com/%s/status/%s' % (x['_source']['user']['screen_name'], x['_source']['id_str'])
return x['_source'];
es_body_tw_urls = list(map(handele,filter(lambda x:not x['_source']['truncated'],es_body_tw)))
# print(es_body_tw_urls)
if len(es_body_tw_urls)>200:
pool = mp.Pool()
res = pool.map(asynchronous_request,(es_body_tw_urls[i:i+200] for i in range(0,len(es_body_tw_urls),200)))
# current += 1;
print('更新%s用户' % id)
elif 0<len(es_body_tw_urls)<200:
asynchronous_request(ops=es_body_tw_urls)
# current += 1;
print('更新%s用户' % id)
# print('第几%s个' % current)
else:
current += 1;
print('该用户%s无需更新' % id)
print('第几%s个' % current)
conn.send(id)
except Exception as e:
current = 0
# print(e)
raise e
def update_twitter_users_count(self, crawler): #twitter 用户的list字段
# with open(os.path.abspath('twitter_user_ids.json'), 'r') as f:
# user_ids = json.load(f)
# count=1;
print('【爬取twitter usersInfos】')
db = self.create_mongo_conn(db='Twitter', collection='twitter')
weipa_count = 1
while True:
try:
twitter_users_queue = RedisQueue(
name='twitter_users',
redis_config=self.app_config['redis_config'])
print('[===未抓取的的id个数为:%s===]' % twitter_users_queue.qsize())
if twitter_users_queue.empty():
if weipa_count >= 3:
print('<-----用户抓取完成----->')
break
else:
weipa_count += 1
print('[==Retry:%s==]' % (weipa_count - 1))
time.sleep(5)
continue
id = twitter_users_queue.get()
tweet = db.find_one({"id_str": id})
tweet_count, flowing_count, followers_count, favorites_count, list_count, moment_count = crawler.crawler_list_count(
tweet["screen_name"], user_id=id)
print(tweet_count, flowing_count, followers_count,
favorites_count, list_count, moment_count)
after_doc = db.find_one_and_update(
{"id_str": tweet['id_str']}, {
"$set": {
"list_num": list_count,
"moment_num": moment_count,
"followers_count": followers_count,
"friends_count": flowing_count,
"statuses_count": tweet_count,
"favourites_count": favorites_count,
"update_time": datetime.now(),
}
},
return_document=ReturnDocument.AFTER)
print('更新%s文档成功' % after_doc['_id'])
# count +=1;
weipa_count = 1
except Exception as e:
raise e
import sys,os,json
sys.path.append(".")
from src.redis_helper import RedisQueue
if __name__ == '__main__':
twitter_crawler_queue = RedisQueue(name='twittter')
while True:
print(twitter_crawler_queue.qsize())
print(twitter_crawler_queue.get())
def crawler_tweets_replay_count(self,
crawler,
history=False): #推文表中的replay_count字段
print('<=====启动tweet_replay抓取====>')
db = self.create_mongo_conn()
crawler_tweet_replay_queue = RedisQueue(
name='twitter_replay',
redis_config=self.app_config['redis_config'])
weipa_count = 1
err_count = 1
if not history:
es = Espusher()
while True:
try:
print('[===未抓取的个数为:%s===]' %
crawler_tweet_replay_queue.qsize())
# if crawler_tweet_replay_queue.empty():
# if weipa_count >=3:
# print('<-----twitter_replay抓取完成----->')
# break
# else:
# weipa_count+=1
# print('[==Retry:%s==]' % (weipa_count-1))
# time.sleep(10)
# continue
# print(weipa_count)
urls = crawler_tweet_replay_queue.get()
print("取出%s个" % len(urls))
content = crawler.crawler_replay_num(urls)
for item in content:
print(item)
# print(item['url'].split('/')[-1])
if history:
update_doc = db.update_many(
{
"id_str": item['url'].split('/')[-1],
'site': 'twitter'
}, {
'$set': {
'replay_count': item['reply_count'],
'retweet_count': item['retweet_count'],
'favorite_count': item['favorite_count'],
"update_status": True
}
})
print('更新了%s个' % (update_doc.modified_count))
else:
# print('push item to es')
# print(item)
data = db.find_one_and_delete({
"id_str":
item['url'].split('/')[-1],
'site':
'twitter'
})
data['replay_count'] = item['reply_count']
data['favorite_count'] = item['favorite_count']
data['retweet_count'] = item['retweet_count']
es.twitter_pusher(data)
weipa_count = 1
err_count = 1
except Exception as e:
print(e)
continue
def crawler_reactions(self, crawler, history=False): #facebook posts表中的各种量
print('<=====启动facebook_reactions抓取====>')
db = self.create_mongo_conn()
crawler_reactions_queue = RedisQueue(
name='facebook_reactions',
redis_config=self.app_config['redis_config'])
weipa_count = 1
err_count = 1
if not history:
es = Espusher()
while True:
try:
# print(db.count({"site": "facebook","update_status":False}))
# tweets = list(db.find({"site": "facebook","update_status":False}).limit(20))
# if (len(tweets) == 0):
# print('全部爬取完成')
# break;
print('[===未抓取的个数为:%s===]' % crawler_reactions_queue.qsize())
# if crawler_reactions_queue.empty():
# if weipa_count >=3:
# print('<-----facebook_reactions抓取完成----->')
# break
# else:
# weipa_count+=1
# print('[==Retry:%s==]' % (weipa_count-1))
# time.sleep(10)
# continue
urls = crawler_reactions_queue.get(
) #map(lambda x:{"url":'https://facebook.com%s' % x['permalink_url'],'id':x['_id']},tweets)
content = crawler.crawler_reactions_nums(urls)
# print(content)
if not content: continue
for item in content:
# print(item)
print(item['reactions'])
# print(item['url'])
# print(url)
if not item['reactions']:
print(item)
else:
if history:
print(objectid.ObjectId(item['url']['id']))
update_doc = db.find_one_and_update(
{"_id": objectid.ObjectId(item['url']['id'])},
{
'$set': {
'comment_num':
item['reactions']['comment_count'],
'likes_num':
item['reactions']['likes_count'],
'share_count':
item['reactions']["share_count"],
"update_status":
True
}
},
return_document=ReturnDocument.AFTER)
if update_doc != None:
print('更新了%s个' % update_doc['_id'])
else:
data = db.find_one_and_delete(
{'_id': objectid.ObjectId(item['url']['id'])})
data['comment_num'] = item['reactions'][
'comment_count']
data['likes_num'] = item['reactions'][
'likes_count']
data['share_count'] = item['reactions'][
"share_count"]
es.facebook_pusher(data)
weipa_count = 1
err_count = 1
except Exception as e:
print(e)
continue
def crawler_tweets(self, crawler, site='facebook', deadtime='2017-1-1'):
print('<-----启动文章抓取----->')
weipa_count = 1
if site == 'twitter':
twitter_crawler_queue = RedisQueue(
name='twitter', redis_config=self.app_config['redis_config'])
twitter_crawler_error_queue = RedisQueue(
name='twitter_error',
redis_config=self.app_config['redis_config'])
while True:
if twitter_crawler_error_queue.qsize() > 0:
err_item = twitter_crawler_error_queue.get()
print('取出出错条目:%s', err_item)
current_max_id = err_item['current_max_id']
id = err_item['user_id']
crawler.fetch_user_tweets(user_id=id,
current_max_id=current_max_id,
deadline=deadtime)
else:
print('\n')
print('[===Info:%s 未抓取的的id个数为:%s===]' %
(datetime.now(), twitter_crawler_queue.qsize()))
if twitter_crawler_queue.empty():
if weipa_count >= 3:
print('<-----文章抓取完成----->')
break
else:
weipa_count += 1
print('[==Retry:%s==]' % (weipa_count - 1))
time.sleep(5)
continue
id = twitter_crawler_queue.get()
crawler.fetch_user_tweets(user_id=id, deadline=deadtime)
weipa_count = 1
else:
facebook_crawler_queue = RedisQueue(
name='facebook', redis_config=self.app_config['redis_config'])
facebook_crawler_error_queue = RedisQueue(
name='facebook_error',
redis_config=self.app_config['redis_config'])
db = self.create_mongo_conn(db='FaceBook', collection='facebook')
while True:
if facebook_crawler_error_queue.qsize() > 0:
err_item = facebook_crawler_error_queue.get()
print('取出出错条目:%s', err_item)
id = err_item['id']
url = err_item['url']
if url:
crawler.fetch_user_tweets(id=id,
urls=url,
deadline=deadtime)
else:
print('\n')
print('[===Info:%s 未抓取的的id个数为:%s===]' %
(datetime.now(), facebook_crawler_queue.qsize()))
if facebook_crawler_queue.empty():
if weipa_count >= 3:
print('<-----文章抓取完成----->')
break
else:
weipa_count += 1
print('[==Retry:%s==]' % (weipa_count - 1))
time.sleep(5)
continue
id = facebook_crawler_queue.get()
print(id)
doc = db.find_one({"id": str(id)})
# print(doc)
crawler.fetch_user_tweets(id=id,
urls=doc['link'] + 'posts/',
deadline=deadtime)
# print('完成全部抓取')
weipa_count = 1