def start_getweiboinfo(self,response):
db = MysqlStore()
conn = db.get_connection()
sql1 = "select * from t_user_follow where contentstate = 0"
cursor1 = db.select_operation(conn,sql1)
sql2 = "select count(*) from t_user_follow where contentstate = 0"
cursor2 = db.select_operation(conn,sql2)
count = cursor2.fetchone()
for i in range(10): #(count[0]): #count[0]为需要再爬取的用户数
for result in cursor1.fetchmany(1):
if result[1]:
mainpageurl = 'http://weibo.com/u/'+str(result[1])+'?from=otherprofile&wvr=3.6&loc=tagweibo'
GetWeibopage.data['uid'] = result[1]
getweibopage = GetWeibopage()
for page in range(WeiboSpider.page_num):
GetWeibopage.data['page'] = page+1
firstloadurl = mainpageurl + getweibopage.get_firstloadurl()
yield Request(url=firstloadurl,meta={'cookiejar':response.meta['cookiejar'],'uid':result[1]},callback=self.parse_firstload)
secondloadurl = mainpageurl + getweibopage.get_secondloadurl()
yield Request(url=secondloadurl,meta={'cookiejar':response.meta['cookiejar'],'uid':result[1]},callback=self.parse_secondload)
thirdloadurl = mainpageurl + getweibopage.get_thirdloadurl()
yield Request(url=thirdloadurl,meta={'cookiejar':response.meta['cookiejar'],'uid':result[1]},callback=self.parse_thirdload)
def search_from_keywordDB(self,response):
db = MysqlStore();conn = db.get_connection()
main_url = "http://s.weibo.com/weibo/"
getsearchpage = GetSearchpage()
sql1 = "select keyword from cauc_keyword_test"
cursor = db.select_operation(conn,sql1)
for keyword in cursor.fetchall():
print "this is the keyword:",keyword
keywords = ['机场 炸弹','飞机 炸弹']
for i in range(15):
for keyword in keywords:
sql3 = "select max(publish_time) from cauc_keyword_info where keyword = '%s'" % keyword #检查是否爬取过该关键词
cursor = db.select_operation(conn,sql3)
newest_time = cursor.fetchone()[0]
if newest_time is not None: #已经爬取过该关键词，获取最新时间用于"最新时间-当前时间"时间段内容获取
current_time = get_current_time()
newest_time = format_time(newest_time)
print "爬取过的关键词:%s,搜索时间段%s~%s间的内容" % (keyword,newest_time,current_time)
search_url = main_url + getsearchpage.get_searchurl_time(keyword,newest_time,current_time)
yield Request(url=search_url,meta={'cookiejar':response.meta['cookiejar'],'search_url':search_url,'keyword':keyword},callback=self.parse_total_page)
else: #未爬取过该关键词
print "未爬取过的关键词:%s" % keyword
search_url = main_url + getsearchpage.get_searchurl(keyword)
yield Request(url=search_url,meta={'cookiejar':response.meta['cookiejar'],'search_url':search_url,'keyword':keyword},callback=self.parse_total_page)
time.sleep(100000)
conn.close()
def clear_flag():
conn = MysqlStore().get_connection()
cursor = conn.cursor()
sql = "update cauc_parameters set param_value = 0 where param_key = 'flag'"
n = cursor.execute(sql)
conn.commit()
db.close_connection(conn,cursor)
if n:
logger.info('clear_flag success!')
def update_cookies(username,cookies):
'''更新cookie到数据库'''
db = MysqlStore()
conn = db.get_connection()
timestamp = int(time.time())
cursor = conn.cursor()
cursor.execute("insert into cauc_login_cookie_info(USERNAME,SUHB,SUB,SUBP,SUE,SUS,SUP,timestamp) values (%s,%s,%s,%s,%s,%s,%s,%s) ON DUPLICATE KEY UPDATE SUHB=%s,SUB=%s,SUBP=%s,SUE=%s,SUS=%s,SUP=%s,timestamp=%s",(md5(username),cookies['SUHB'],cookies['SUB'],cookies['SUBP'],cookies['SUE'],cookies['SUS'],cookies['SUP'],timestamp,cookies['SUHB'],cookies['SUB'],cookies['SUBP'],cookies['SUE'],cookies['SUS'],cookies['SUP'],timestamp))
conn.commit()
db.close_connection(conn,cursor)
logger.info('Update cookies into database...')
def allcookie_fetch():
'''获取所有cookie'''
db = MysqlStore()
conn = db.get_connection()
cursor = conn.cursor(MySQLdb.cursors.DictCursor)
nums = cursor.execute('SELECT SUHB,SUB,SUBP,SUE,SUS,SUP,USERNAME FROM cauc_login_cookie_info WHERE (unix_timestamp()-cast(timestamp as signed)) < (%s)',(EXPIRES,))
rows = cursor.fetchallDict()
conn.commit()
db.close_connection(conn,cursor)
return nums,rows
def set_flag():
'''设置flag'''
db = MysqlStore()
conn = db.get_connection()
cursor = conn.cursor()
sql = "update cauc_parameters set param_value = 1 where param_key = 'flag'"
n = cursor.execute(sql)
conn.commit()
db.close_connection(conn,cursor)
if n:
logger.info('Set flag success!')
else:
logger.error('Set flag failed,flag is already 1!')
def user_fetch():
'''获取账号信息'''
db = MysqlStore()
conn = db.get_connection()
cursor = conn.cursor(MySQLdb.cursors.DictCursor)
sql = 'SELECT id,username,password FROM cauc_login_account_info where is_delete = 0'
nums = cursor.execute(sql)
users = cursor.fetchallDict()
conn.commit()
db.close_connection(conn,cursor)
if users:
logger.info('User fetch success!')
return users
else:
logger.error('There is no user in database!')
def search_from_keywordDB(self, response):
db = MysqlStore()
main_url = "http://s.weibo.com/weibo/"
getsearchpage = GetSearchpage()
for round in range(1): #遍历数据库的轮数
conn = db.get_connection()
#选取is_search位为0的关键词
sql1 = "select keyword from cauc_keyword_test_copy where is_search = 0 and is_delete = 0"
cursor1 = db.select_operation(conn, sql1)
#对is_search位为0的关键词进行爬取
for keyword in cursor1.fetchall():
keyword = keyword[0]
logger.info("this is the unsearched keyword:%s", keyword)
search_url = main_url + getsearchpage.get_searchurl(keyword)
yield Request(url=search_url,
cookies=random.choice(COOKIES),
meta={
'search_url': search_url,
'keyword': keyword
},
callback=self.parse_total_page)
#选取is_search位为1的关键词
sql2 = "select keyword from cauc_keyword_test_copy where is_search = 1 and is_delete = 0"
cursor2 = db.select_operation(conn, sql2)
#对is_search位为1的关键词进行爬取
for keyword in cursor2.fetchall():
keyword = keyword[0]
logger.info("this is the searched keyword:%s", keyword)
end_time = get_current_time()
#start_time = get_time_by_interval(int(time.time()),3600) #爬取3600秒，即1小时前的内容
start_time = get_time_by_interval(int(
time.time()), int(self.interval)) #爬取interval秒前的内容
search_url = main_url + getsearchpage.get_searchurl_time(
keyword, start_time, end_time)
yield Request(url=search_url,
cookies=random.choice(COOKIES),
meta={
'search_url': search_url,
'keyword': keyword
},
callback=self.parse_total_page)
#更新is_search标志位为1
sql3 = "update cauc_keyword_test_copy set is_search = 1 where is_search = 0 and is_delete = 0"
db.update_operation(conn, sql3)
db.close_connection(conn)
def start_getweiboinfo(self, response):
db = MysqlStore()
conn = db.get_connection()
sql1 = "select * from t_user_follow where contentstate = 0"
cursor1 = db.select_operation(conn, sql1)
sql2 = "select count(*) from t_user_follow where contentstate = 0"
cursor2 = db.select_operation(conn, sql2)
count = cursor2.fetchone()
for i in (count[0]): #count[0]为需要再爬取的用户数
for result in cursor1.fetchmany(1):
if result[1]:
mainpageurl = 'http://weibo.com/u/' + str(
result[1]) + '?from=otherprofile&wvr=3.6&loc=tagweibo'
GetWeibopage.data['uid'] = result[1]
getweibopage = GetWeibopage()
for page in range(WeiboSpider.page_num):
GetWeibopage.data['page'] = page + 1
firstloadurl = mainpageurl + getweibopage.get_firstloadurl(
)
yield Request(url=firstloadurl,
meta={
'cookiejar':
response.meta['cookiejar'],
'uid': result[1]
},
callback=self.parse_firstload)
secondloadurl = mainpageurl + getweibopage.get_secondloadurl(
)
yield Request(url=secondloadurl,
meta={
'cookiejar':
response.meta['cookiejar'],
'uid': result[1]
},
callback=self.parse_secondload)
thirdloadurl = mainpageurl + getweibopage.get_thirdloadurl(
)
yield Request(url=thirdloadurl,
meta={
'cookiejar':
response.meta['cookiejar'],
'uid': result[1]
},
callback=self.parse_thirdload)
def get_userinfo(self,response):
db = MysqlStore()
conn = db.get_connection()
#sql1 = "select * from t_user_follow where infostate = 0 and contentstate = 0"
sql1 = "select * from t_user_info where imagestate = 1 and imageurl = 1"
cursor1 = db.select_operation(conn,sql1)
sql2 = "select count(*) from t_user_follow where infostate = 0 and contentstate = 0"
cursor2 = db.select_operation(conn,sql2)
count = cursor2.fetchone()
for i in range(1): #count[0]):
for result in cursor1.fetchmany(1):
if result[0]:
mainpageurl = 'http://weibo.com/u/'+str(result[0])+'?from=otherprofile&wvr=3.6&loc=tagweibo'
GetWeibopage.data['uid'] = result[0] #result[1]
getweibopage = GetWeibopage()
GetWeibopage.data['page'] = 1
firstloadurl = mainpageurl + getweibopage.get_firstloadurl()
yield Request(url=firstloadurl,meta={'cookiejar':response.meta['cookiejar'],'uid':result[0]},callback=self.get_userurl)
def search_from_keywordDB(self,response):
db = MysqlStore();conn = db.get_connection()
main_url = "http://s.weibo.com/weibo/"
getsearchpage = GetSearchpage()
sql1 = "select keyword from cauc_keyword_test_copy where is_search = 0"
cursor = db.select_operation(conn,sql1)
for round in range(3):
for keyword in cursor.fetchall():
keyword = keyword[0]
print "this is the unsearched keyword:",keyword
sql2 = "update cauc_keyword_test_copy set is_search = 1 where keyword = '%s'" % keyword
db.update_operation(conn,sql2)
search_url = main_url + getsearchpage.get_searchurl(keyword)
#yield Request(url=search_url,meta={'cookiejar':response.meta['cookiejar'],'search_url':search_url,'keyword':keyword},callback=self.parse_total_page)
print "current timestamp:",int(time.time())
#设置循环爬取间隔
time.sleep(WeiboSpider.settings['KEYWORD_INTERVAL'])
sql3 = "select keyword from cauc_keyword_test_copy where is_search = 1"
cursor = db.select_operation(conn,sql3)
for keyword in cursor.fetchall():
keyword = keyword[0]
print "this is the searched keyword",keyword
#yield Request(url=search_url,meta={'cookiejar':response.meta['cookiejar'],'search_url':search_url,'keyword':keyword},callback=self.parse_total_page)
conn.close()
def start_getweibo_info(self,response):
db = MysqlStore();
#取出没有爬取过的且is_delete=0的重点人员
GetWeibopage.data['page'] = 1; getweibopage = GetWeibopage()
#for round in range(1): #遍历数据库的轮数
conn = db.get_connection()
sql1 = "select user_id from cauc_warning_man a \
where a.is_search = 0 and a.is_delete = 0"
cursor1 = db.select_operation(conn,sql1)
for user_id in cursor1.fetchall():
user_id = user_id[0]
logger.info("this is the unsearched user_id:%s",user_id)
#获取需要爬取的总页面数
start_time = self.start_time;end_time = get_current_time('day')
mainpage_url = "http://weibo.com/" + str(user_id) + "?is_ori=1&is_forward=1&is_text=1&is_pic=1&key_word=&start_time=" + start_time + "&end_time=" + end_time + "&is_search=1&is_searchadv=1&"
GetWeibopage.data['uid'] = user_id;
thirdload_url = mainpage_url + getweibopage.get_thirdloadurl()
yield Request(url=thirdload_url,cookies=random.choice(COOKIES),meta={'mainpage_url':mainpage_url,'uid':user_id,'is_search':0},callback=self.parse_total_page)
#取出已经爬取过is_search=1的且is_delete=0的预警人员
sql2 = "select user_id from cauc_warning_man a \
where a.is_search = 1 and a.is_delete = 0"
cursor2 = db.select_operation(conn,sql2)
for user_id in cursor2.fetchall():
user_id = user_id[0]
logger.info("this is the searched user_id:%s",user_id)
#start_time = get_time_by_interval(int(time.time()),86400,'day');end_time = get_current_time('day') #起始和结束间隔时间为1天(86400s),即过去一天的内容
start_time = get_time_by_interval(int(time.time()),int(self.interval),'day');end_time = get_current_time('day') #起始和结束间隔时间为x天(由interval代表的秒换算而来)
mainpage_url = "http://weibo.com/" + str(user_id) + "?is_ori=1&is_forward=1&is_text=1&is_pic=1&key_word=&start_time=" + start_time + "&end_time=" + end_time + "&is_search=1&is_searchadv=1&"
GetWeibopage.data['uid'] = user_id;
thirdload_url = mainpage_url + getweibopage.get_thirdloadurl()
yield Request(url=thirdload_url,cookies=random.choice(COOKIES),meta={'mainpage_url':mainpage_url,'uid':user_id,'is_search':1},callback=self.parse_total_page)
#更新is_search标志位为1
sql3 = "update cauc_warning_man set is_search = 1 where is_search = 0 and is_delete = 0"
db.update_operation(conn,sql3)
db.close_connection(conn)
def search_from_keywordDB(self, response):
db = MysqlStore()
main_url = "http://s.weibo.com/weibo/"
getsearchpage = GetSearchpage()
for round in range(1): #遍历数据库的轮数
conn = db.get_connection()
#对is_search位为0的关键词进行爬取
sql1 = "select keyword from cauc_keyword_test_copy where is_search = 0"
cursor = db.select_operation(conn, sql1)
for keyword in cursor.fetchall():
keyword = keyword[0]
logger.info("this is the unsearched keyword:%s", keyword)
#更新is_search标志位为1
sql2 = "update cauc_keyword_test_copy set is_search = 1 where keyword = '%s'" % keyword
db.update_operation(conn, sql2)
search_url = main_url + getsearchpage.get_searchurl(keyword)
yield Request(url=search_url,
meta={
'cookiejar': response.meta['cookiejar'],
'search_url': search_url,
'keyword': keyword
},
callback=self.parse_total_page)
logger.info("current timestamp:%d", int(time.time()))
#设置循环爬取间隔
time.sleep(WeiboSpider.settings['KEYWORD_INTERVAL']) #可以采用间隔15min
#对is_search位为1的关键词进行爬取
sql3 = "select keyword from cauc_keyword_test_copy where is_search = 1"
cursor = db.select_operation(conn, sql3)
for keyword in cursor.fetchall():
keyword = keyword[0]
logger.info("this is the searched keyword:%s", keyword)
end_time = get_current_time()
start_time = get_time_by_interval(int(time.time()),
3600) #爬取3600秒，即1小时前的内容
search_url = main_url + getsearchpage.get_searchurl_time(
keyword, start_time, end_time)
yield Request(url=search_url,
meta={
'cookiejar': response.meta['cookiejar'],
'search_url': search_url,
'keyword': keyword
},
callback=self.parse_total_page)
conn.close()
def start_getweibo_info(self,response):
db = MysqlStore();
#取出没有爬取过的且is_delete=0的重点人员
GetWeibopage.data['page'] = 1; getweibopage = GetWeibopage()
for round in range(1): #遍历数据库的轮数
conn = db.get_connection()
sql1 = "select user_id from cauc_warning_man_test a \
where a.is_search = 0 and a.is_delete = 0"
cursor = db.select_operation(conn,sql1)
for user_id in cursor.fetchall():
user_id = user_id[0]
logger.info("this is the unsearched user_id:%s",user_id)
#更新is_search标志位为1
sql2 = "update cauc_warning_man_test set is_search = 1 where user_id = '%s'" % user_id
db.update_operation(conn,sql2)
#获取需要爬取的总页面数
start_time = self.start_time;end_time = get_current_time('hour')
mainpage_url = "http://weibo.com/" + str(user_id) + "?is_ori=1&is_forward=1&is_text=1&is_pic=1&key_word=&start_time=" + start_time + "&end_time=" + end_time + "&is_search=1&is_searchadv=1&"
GetWeibopage.data['uid'] = user_id;
thirdload_url = mainpage_url + getweibopage.get_thirdloadurl()
yield Request(url=thirdload_url,meta={'cookiejar':response.meta['cookiejar'],'mainpage_url':mainpage_url,'uid':user_id,'is_search':0},callback=self.parse_total_page)
logger.info("current timestamp:%d",int(time.time()))
#设置循环爬取间隔
time.sleep(WeiboSpider.settings['WEIBOCONTENT_INTERVAL']) #可以采用间隔15min
#取出已经爬取过is_search=1的且is_delete=0的预警人员
sql3 = "select user_id from cauc_warning_man_test a \
where a.is_search = 1 and a.is_delete = 0"
cursor = db.select_operation(conn,sql3)
for user_id in cursor.fetchall():
user_id = user_id[0]
logger.info("this is the searched user_id:%s",user_id)
start_time = get_time_by_interval(int(time.time()),86400,'hour');end_time = get_current_time('hour') #起始和结束间隔时间为1天(86400s)
mainpage_url = "http://weibo.com/" + str(user_id) + "?is_ori=1&is_forward=1&is_text=1&is_pic=1&key_word=&start_time=" + start_time + "&end_time=" + end_time + "&is_search=1&is_searchadv=1&"
GetWeibopage.data['uid'] = user_id;
thirdload_url = mainpage_url + getweibopage.get_thirdloadurl()
#yield Request(url=thirdload_url,meta={'cookiejar':response.meta['cookiejar'],'mainpage_url':mainpage_url,'uid':user_id,'is_search':1},callback=self.parse_total_page)
conn.close()
def search_from_keywordDB(self,response):
db = MysqlStore();main_url = "http://s.weibo.com/weibo/"
getsearchpage = GetSearchpage()
for round in range(1): #遍历数据库的轮数
conn = db.get_connection()
#选取is_search位为0的关键词
sql1 = "select keyword from cauc_keyword_test_copy where is_search = 0 and is_delete = 0"
cursor1 = db.select_operation(conn,sql1)
#对is_search位为0的关键词进行爬取
for keyword in cursor1.fetchall():
keyword = keyword[0]
logger.info("this is the unsearched keyword:%s",keyword)
search_url = main_url + getsearchpage.get_searchurl(keyword)
yield Request(url=search_url,cookies=random.choice(COOKIES),meta={'search_url':search_url,'keyword':keyword},callback=self.parse_total_page)
#选取is_search位为1的关键词
sql2 = "select keyword from cauc_keyword_test_copy where is_search = 1 and is_delete = 0"
cursor2 = db.select_operation(conn,sql2)
#对is_search位为1的关键词进行爬取
for keyword in cursor2.fetchall():
keyword = keyword[0]
logger.info("this is the searched keyword:%s",keyword)
end_time = get_current_time()
#start_time = get_time_by_interval(int(time.time()),3600) #爬取3600秒，即1小时前的内容
start_time = get_time_by_interval(int(time.time()),int(self.interval)) #爬取interval秒前的内容
search_url = main_url + getsearchpage.get_searchurl_time(keyword,start_time,end_time)
yield Request(url=search_url,cookies=random.choice(COOKIES),meta={'search_url':search_url,'keyword':keyword},callback=self.parse_total_page)
#更新is_search标志位为1
sql3 = "update cauc_keyword_test_copy set is_search = 1 where is_search = 0 and is_delete = 0"
db.update_operation(conn,sql3)
db.close_connection(conn)
def start_getweibo_info(self,response):
db = MysqlStore();
#取出没有爬取过的且is_delete=0的重点人员
GetWeibopage.data['page'] = 1; getweibopage = GetWeibopage()
for round in range(1): #遍历数据库的轮数
conn = db.get_connection()
sql1 = "select user_id from cauc_black_man_test a \
where a.is_search = 0 and a.is_delete = 0"
cursor = db.select_operation(conn,sql1)
for user_id in cursor.fetchall():
user_id = user_id[0]
logger.info("this is the unsearched user_id:%s",user_id)
#更新is_search标志位为1
sql2 = "update cauc_black_man_test set is_search = 1 where user_id = '%s'" % user_id
db.update_operation(conn,sql2)
#获取需要爬取的总页面数
start_time = self.start_time;end_time = get_current_time('hour')
mainpage_url = "http://weibo.com/" + str(user_id) + "?is_ori=1&is_forward=1&is_text=1&is_pic=1&key_word=&start_time=" + start_time + "&end_time=" + end_time + "&is_search=1&is_searchadv=1&"
GetWeibopage.data['uid'] = user_id;
thirdload_url = mainpage_url + getweibopage.get_thirdloadurl()
yield Request(url=thirdload_url,meta={'cookiejar':response.meta['cookiejar'],'mainpage_url':mainpage_url,'uid':user_id,'is_search':0},callback=self.parse_total_page)
logger.info("current timestamp:%d",int(time.time()))
#设置循环爬取间隔
time.sleep(WeiboSpider.settings['FRIENDCIRCAL_INTERVAL']) #可以采用间隔15min
#取出已经爬取过is_search=1的且is_delete=0的重点人员
sql3 = "select user_id from cauc_black_man_test a \
where a.is_search = 1 and a.is_delete = 0"
cursor = db.select_operation(conn,sql3)
for user_id in cursor.fetchall():
user_id = user_id[0]
logger.info("this is the searched user_id:%s",user_id)
start_time = get_time_by_interval(int(time.time()),86400,'hour');end_time = get_current_time('hour') #起始和结束间隔时间为1天(86400s)
mainpage_url = "http://weibo.com/" + str(user_id) + "?is_ori=1&is_forward=1&is_text=1&is_pic=1&key_word=&start_time=" + start_time + "&end_time=" + end_time + "&is_search=1&is_searchadv=1&"
GetWeibopage.data['uid'] = user_id;
thirdload_url = mainpage_url + getweibopage.get_thirdloadurl()
#yield Request(url=thirdload_url,meta={'cookiejar':response.meta['cookiejar'],'mainpage_url':mainpage_url,'uid':user_id,'is_search':1},callback=self.parse_total_page)
conn.close()
def search_from_keywordDB(self,response):
db = MysqlStore();main_url = "http://s.weibo.com/weibo/"
getsearchpage = GetSearchpage()
for round in range(1): #遍历数据库的轮数
conn = db.get_connection()
#对is_search位为0的关键词进行爬取
sql1 = "select keyword from cauc_keyword_test_copy where is_search = 0"
cursor = db.select_operation(conn,sql1)
for keyword in cursor.fetchall():
keyword = keyword[0]
logger.info("this is the unsearched keyword:%s",keyword)
#更新is_search标志位为1
sql2 = "update cauc_keyword_test_copy set is_search = 1 where keyword = '%s'" % keyword
db.update_operation(conn,sql2)
search_url = main_url + getsearchpage.get_searchurl(keyword)
yield Request(url=search_url,meta={'cookiejar':response.meta['cookiejar'],'search_url':search_url,'keyword':keyword},callback=self.parse_total_page)
logger.info("current timestamp:%d",int(time.time()))
#设置循环爬取间隔
time.sleep(WeiboSpider.settings['KEYWORD_INTERVAL']) #可以采用间隔15min
#对is_search位为1的关键词进行爬取
sql3 = "select keyword from cauc_keyword_test_copy where is_search = 1"
cursor = db.select_operation(conn,sql3)
for keyword in cursor.fetchall():
keyword = keyword[0]
logger.info("this is the searched keyword:%s",keyword)
end_time = get_current_time()
start_time = get_time_by_interval(int(time.time()),3600) #爬取3600秒，即1小时前的内容
search_url = main_url + getsearchpage.get_searchurl_time(keyword,start_time,end_time)
yield Request(url=search_url,meta={'cookiejar':response.meta['cookiejar'],'search_url':search_url,'keyword':keyword},callback=self.parse_total_page)
conn.close()
from datamysql import MysqlStore
d = MysqlStore()
conn = d.get_connection()
sql = "select * from t_user_follow"
cursor = d.select_operation(conn,sql)
for i in range(100):
print 'i:',i
for result in cursor.fetchmany(5):
if result[1]:
print 'hhhhhhh'
