Skip to content

Z
zzsn_spider
提交 7bada3ac 作者: LiuLiYuan
操作

Merge remote-tracking branch 'origin/master'

上级 a237c960 f0246e72
正在显示 包含 175 行增加 和 149 行删除
base/BaseCore.py
......@@ -17,14 +17,15 @@ import langid
# 注意 程序退出前 调用BaseCore.close() 关闭相关资源
class BaseCore:
# 序列号
__seq = 0
# 代理池 数据库连接
__cnx_proxy =None
__cursor_proxy = None
cnx = None
cursor = None
r = None
# agent 池
__USER_AGENT_LIST = [
'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_14_5) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/75.0.3770.90 Safari/537.36',
......@@ -392,7 +393,7 @@ class BaseCore:
# 从Redis的List中获取并移除一个元素
def redicPullData(self,key):
item = self.r.rpop(key)
item = self.r.lpop(key)
return item.decode() if item else None
# 获得脚本进程PID
......@@ -480,7 +481,7 @@ class BaseCore:
def writerToExcel(self,detailList,filename):
# filename='baidu搜索.xlsx'
# 读取已存在的xlsx文件
existing_data = pd.read_excel(filename,engine='openpyxl')
existing_data = pd.read_excel(filename,engine='openpyxl',dtype=str)
# 创建新的数据
new_data = pd.DataFrame(data=detailList)
# 将新数据添加到现有数据的末尾
......
base/RedisPPData.py
import time
import pymysql
from base import BaseCore
from apscheduler.schedulers.blocking import BlockingScheduler
basecore = BaseCore.BaseCore()
log = basecore.getLogger()
#144数据库
cnx = basecore.cnx
cursor = basecore.cursor
r = basecore.r
#11数据库
cnx_ = pymysql.connect(host='114.116.44.11', user='root', password='f7s0&7qqtK', db='clb_project', charset='utf8mb4')
cursor_ = cnx_.cursor()
# # 连接到Redis
# r = redis.Redis(host="114.115.236.206", port=6379, password='clbzzsn', db=6)
#
......@@ -32,9 +39,9 @@ r = basecore.r
#企业动态
def NewsEnterprise():
# #获取国内企业
# gn_query = "select SocialCode from EnterpriseInfo where Place = '1'"
# cursor.execute(gn_query)
# gn_result = cursor.fetchall()
gn_query = "select SocialCode from EnterpriseInfo where Place = '1'"
cursor.execute(gn_query)
gn_result = cursor.fetchall()
#获取国外企业
gw_query = "select SocialCode from EnterpriseInfo where Place = '2'"
cursor.execute(gw_query)
......@@ -42,20 +49,20 @@ def NewsEnterprise():
gw_social_list = [item[0] for item in gw_result]
#todo:打印长度
print(len(gw_social_list))
# gn_social_list = [item[0] for item in gn_result]
# print(len(gw_social_list))
gn_social_list = [item[0] for item in gn_result]
print('=======')
#将数据插入到redis中
# for item in gn_social_list:
# r.rpush('NewsEnterprise:gnqy_socialCode', item)
count = 0
# count = 0
for item in gw_social_list:
r.rpush('NewsEnterprise:gwqy_socialCode', item)
count+=1
print(item)
print(count)
# count+=1
# print(item)
# print(count)
#企业动态定时任务
def NewsEnterprise_task():
# 实例化一个调度器
......@@ -140,6 +147,29 @@ def BaseInfoEnterprise_task():
print('定时采集异常', e)
pass
#东方财富网财务数据
def FinanceFromEast():
#从上市企业库中读取数据
sql_sel = '''select social_credit_code from sys_base_enterprise_ipo where category = '1' limit 10 '''
cursor_.execute(sql_sel)
finance = cursor_.fetchall()
finance_list = [item[0] for item in finance]
print('=======')
for item in finance_list:
r.rpush('FinanceFromEast:finance_socialCode', item)
def FinanceFromEase_task():
# 实例化一个调度器
scheduler = BlockingScheduler()
# 每个季度执行一次
scheduler.add_job(FinanceFromEast, 'cron', month='1-12/3', day='1',hour=0, minute=0)
try:
# redisPushData # 定时开始前执行一次
scheduler.start()
except Exception as e:
print('定时采集异常', e)
pass
#微信公众号
def WeiXingetFromSql():
selectSql = "SELECT info_source_code from info_source where site_uri like '%mp.weixin.qq.com%'"
......@@ -207,9 +237,6 @@ def FBS():
r.rpush('NewsEnterpriseFbs:gnqy_socialCode', item)
if __name__ == "__main__":
start = time.time()
# NoticeEnterprise()
......@@ -222,8 +249,9 @@ if __name__ == "__main__":
# FBS()
# NoticeEnterprise_task()
# AnnualEnterprise_task()
NoticeEnterprise()
# NoticeEnterprise()
FinanceFromEast()
log.info(f'====={basecore.getNowTime(1)}=====添加数据成功======耗时：{basecore.getTimeCost(start,time.time())}===')
# cnx.close()
# cursor.close()
cnx_.close()
cursor_.close()
# basecore.close()
comData/weixin_solo/oneWeixin.py
......@@ -41,30 +41,6 @@ def add_url(sid, article_url):
else:
return False
def get_proxy():
cnx = pymysql.connect(host="114.115.159.144", user="root", password="zzsn9988", db="clb_project", charset="utf8mb4")
with cnx.cursor() as cursor:
sql = "select proxy from clb_proxy"
cursor.execute(sql)
proxy_lists = cursor.fetchall()
ip_list = []
for proxy_ in proxy_lists:
ip_list.append(str(proxy_).replace("('", '').replace("',)", ''))
proxy_list = []
for str_ip in ip_list:
str_ip_list = str_ip.split('-')
proxyMeta = "http://%(host)s:%(port)s" % {
"host": str_ip_list[0],
"port": str_ip_list[1],
}
proxy = {
"HTTP": proxyMeta,
"HTTPS": proxyMeta
}
proxy_list.append(proxy)
return proxy_list
#定时
def getFromSql():
selectSql = "SELECT info_source_code from info_source where site_uri like '%mp.weixin.qq.com%'"
cursor.execute(selectSql)
......@@ -93,7 +69,7 @@ def flushAndGetToken(list_b):
def rePutIntoR(item):
r.rpush('WeiXinGZH:infoSourceCode', item)
def get_info(sid,json_search,origin,info_source_code):
def get_info(sid,json_search,origin,url_,info_source_code,page):
list_all_info = []
num_caiji = 0
kaishi_time = time.strftime("%Y-%m-%d %H:%M:%S", time.localtime())
......@@ -103,7 +79,7 @@ def get_info(sid,json_search,origin,info_source_code):
server='https://obs.cn-north-1.myhuaweicloud.com' # 你的桶的地址
)
list_all_news = json_search['app_msg_list']
#采集第几篇文章
for one_news in list_all_news:
news_title = one_news['title']
timestamp = one_news['create_time']
......@@ -114,10 +90,12 @@ def get_info(sid,json_search,origin,info_source_code):
url_ft = check_url(sid, url_news)
if url_ft:
log.info(f'已采过该篇文章----文章链接-----{url_news}')
log.info(f'-----{origin}--第{page}页--已采过该篇文章--文章链接--{url_news}-----')
return list_all_info,num_caiji
try:
res_news = requests.get(url_news, timeout=20)
ip = baseCore.get_proxy()
res_news = requests.get(url_news, timeout=20,proxies=ip)
time.sleep(2)
except:
continue
soup_news = BeautifulSoup(res_news.content, 'html.parser')
......@@ -132,16 +110,17 @@ def get_info(sid,json_search,origin,info_source_code):
try:
news_content = news_html.text
except:
log.info(f'--------内容为空--------{url_news}--------')
log.info(f'----{origin}--第{page}页--该篇文章内容为空--{url_news}-----')
time_now = time.strftime("%Y-%m-%d %H:%M:%S", time.localtime())
false = [
news_title,
url_news,
news_html,
url_,
info_source_code,
'文章内容为空',
time_now
time_now,
url_news
]
insertSql = f"insert into WeixinGZH (site_name,site_url,json_error_info,error_type,create_time) values (%s,%s,%s,%s,%s)"
insertSql = f"insert into WeixinGZH (site_name,site_url,info_source_code,error_type,create_time,news_url) values (%s,%s,%s,%s,%s,%s)"
cursor_.execute(insertSql, tuple(false))
cnx_.commit()
continue
......@@ -182,6 +161,8 @@ def get_info(sid,json_search,origin,info_source_code):
section.name = 'div'
time_now = time.strftime("%Y-%m-%d %H:%M:%S", time.localtime())
#将信息传输到kafka中
dic_info = {
'sid': sid,
'title': news_title,
......@@ -196,8 +177,8 @@ def get_info(sid,json_search,origin,info_source_code):
'createDate': time_now
}
for nnn in range(0, 3):
producer = KafkaProducer(bootstrap_servers=['114.115.159.144:9092'])
try:
producer = KafkaProducer(bootstrap_servers=['114.115.159.144:9092'])
kafka_result = producer.send("crawlerInfo", json.dumps(dic_info, ensure_ascii=False).encode('utf8'))
kafka_time_out = kafka_result.get(timeout=10)
add_url(sid, url_news)
......@@ -205,8 +186,11 @@ def get_info(sid,json_search,origin,info_source_code):
except:
time.sleep(5)
continue
finally:
producer.close()
num_caiji = num_caiji + 1
list_all_info.append(dic_info)
time.sleep(5)
time_now = time.strftime("%Y-%m-%d %H:%M:%S", time.localtime())
dic_info2 = {
'infoSourceId': sid,
......@@ -235,18 +219,18 @@ def job(count,key):
log.info('===========获取公众号============')
start_ = time.time()
# #todo:redis中数据 pop一条
#todo:redis中数据 pop一条
infoSourceCode = baseCore.redicPullData('WeiXinGZH:infoSourceCode')
if infoSourceCode == 'None':
if infoSourceCode == 'None' or infoSourceCode == None:
#当一次采集完之后，重新插入数据并等待插入完成
getFromSql()
time.sleep(20)
log.info(f'========本次公众号已采集完毕，共采集{count}个公众号=========总耗时：{baseCore.getTimeCost(start_,time.time())}')
return
return count
sql = f"SELECT site_uri,id,site_name,info_source_code from info_source where info_source_code = '{infoSourceCode}' "
# '一带一路百人论坛'
# sql = f"SELECT site_uri,id,site_name,info_source_code from info_source where site_name = '一带一路百人论坛' "
# sql = f"-- SELECT site_uri,id,site_name,info_source_code from info_source where info_source_code = 'IN-20220609-57436' "
cursor.execute(sql)
row = cursor.fetchone()
......@@ -282,28 +266,28 @@ def job(count,key):
insertSql = f"insert into WeixinGZH (site_name,site_url,info_source_code,json_error_info,error_type,create_time) values (%s,%s,%s,%s,%s,%s)"
cursor_.execute(insertSql, tuple(error))
cnx_.commit()
return
fakeid = biz + '=='
return count
fakeid = biz + '=='
url_search = f'https://mp.weixin.qq.com/cgi-bin/appmsg?action=list_ex&begin=0&count=5&fakeid={fakeid}&type=9&query=&token={token}&lang=zh_CN&f=json&ajax=1'
# https://mp.weixin.qq.com/cgi-bin/appmsg?action=list_ex&begin=0&count=5&fakeid=MzAwNDA5Njc1Mg==&type=9&query=&token=550883192&lang=zh_CN&f=json&ajax=1
#获取页数
try:
ip = get_proxy()[random.randint(0, 3)]
json_search = s.get(url_search, headers=headers, proxies=ip,
verify=False).json() # , proxies=ip, verify=False
# ip = baseCore.get_proxy()
json_search = s.get(url_search, headers=headers,
verify=False).json() # , proxies=ip, verify=False
str_t = json.dumps(json_search)
time.sleep(2)
time.sleep(1)
except Exception as e:
log.error(f'===公众号{origin}请求失败！当前时间：{baseCore.getNowTime(1)}======={e}===')
rePutIntoR(info_source_code)
return
#{"base_resp": {"ret": 200003, "err_msg": "invalid session"}}
time.sleep(20)
return count
ret = json_search['base_resp']['ret']
# {"base_resp": {"ret": 200003, "err_msg": "invalid session"}}
# TODO:需要判断返回值，根据返回值判断是封号还是biz错误
# {'base_resp': {'err_msg': 'freq control', 'ret': 200013}}========= 封号
# {'base_resp': {'err_msg': 'invalid args', 'ret': 200002}} 公众号biz错误 链接
# 'base_resp': {'err_msg': 'ok', 'ret': 0} 正常
ret = json_search['base_resp']['ret']
if ret == 0:
pass
elif ret == 200013:
......@@ -318,10 +302,9 @@ def job(count,key):
# log.info('=======等待时间600秒=====刷新浏览器=====')
# browser_run = list_b[0]
# browser_run.refresh()
r.set(key, 50)
r.expire(key, 3600)
return
r.expire(key, 5400)
return count
elif ret == 200002:
# 公众号链接错误 保存库里 记录错误信息及错误类型
time_now = time.strftime("%Y-%m-%d %H:%M:%S", time.localtime())
......@@ -336,8 +319,8 @@ def job(count,key):
insertSql = f"insert into WeixinGZH (site_name,site_url,info_source_code,json_error_info,error_type,create_time) values (%s,%s,%s,%s,%s,%s)"
cursor_.execute(insertSql, tuple(error))
cnx_.commit()
log.info(f'公众号----{origin}----耗时{baseCore.getTimeCost(start_,time.time())}')
return
log.info(f'公众号----{origin}----耗时{baseCore.getTimeCost(start_, time.time())}')
return count
elif ret == 200003:
# 无效的session
time_now = time.strftime("%Y-%m-%d %H:%M:%S", time.localtime())
......@@ -353,7 +336,7 @@ def job(count,key):
cursor_.execute(insertSql, tuple(error))
cnx_.commit()
log.info(f'公众号----{origin}----耗时{baseCore.getTimeCost(start_, time.time())}')
return
return count
else:
log.info(f'----其他情况-----{json_search}---公众号{origin}------')
time_now = time.strftime("%Y-%m-%d %H:%M:%S", time.localtime())
......@@ -368,55 +351,84 @@ def job(count,key):
insertSql = f"insert into WeixinGZH (site_name,site_url,info_source_code,json_error_info,error_type,create_time) values (%s,%s,%s,%s,%s,%s)"
cursor_.execute(insertSql, tuple(error))
cnx_.commit()
return
list_all = json_search['app_msg_list']
return count
try:
list_all_info,num_caiji= get_info(sid,json_search,origin,info_source_code)
print(f'----------{len(list_all_info)}------{num_caiji}-------')
time.sleep(2)
if len(list_all_info) != 0:
count += 1
Max_data = int(json_search['app_msg_cnt'])
Max_page = int(int(json_search['app_msg_cnt']) / 5)
if int(json_search['app_msg_cnt']) % 5 != 0:
Max_page = Max_page + 1
else:
Max_page = Max_page
except:
Max_page = 1
Max_data = 5
log.info(f'开始采集{origin}-----共{Max_page}页---{Max_data}条数据-----')
for i in range(0, Max_data, 5):
url_search = f'https://mp.weixin.qq.com/cgi-bin/appmsg?action=list_ex&begin={i}&count=5&fakeid={fakeid}&type=9&query=&token={token}&lang=zh_CN&f=json&ajax=1'
# url_search = f'https://mp.weixin.qq.com/cgi-bin/appmsg?action=list_ex&begin=0&count=5&fakeid={fakeid}&type=9&query=&token={token}&lang=zh_CN&f=json&ajax=1'
# https://mp.weixin.qq.com/cgi-bin/appmsg?action=list_ex&begin=0&count=5&fakeid=MzAwNDA5Njc1Mg==&type=9&query=&token=550883192&lang=zh_CN&f=json&ajax=1
try:
# ip = get_proxy()[random.randint(0, 3)]
json_search = s.get(url_search, headers=headers,
verify=False).json() # , proxies=ip, verify=False
str_t = json.dumps(json_search)
time.sleep(2)
except Exception as e:
log.error(f'===公众号{origin}请求失败！当前时间：{baseCore.getNowTime(1)}======={e}===')
rePutIntoR(info_source_code)
return count
list_all = json_search['app_msg_list']
try:
#开始采集每一页文章信息
page = int(i/5+1)
log.info(f'---{origin}---------开始采集第{page}个分页-----------')
list_all_info,num_caiji= get_info(sid,json_search,origin,url_,info_source_code,page)
print(f'----第{page}页采集到文章个数-----{len(list_all_info)}------{num_caiji}-------')
time.sleep(2)
if len(list_all_info) != 0:
count += 1
time_now = time.strftime("%Y-%m-%d %H:%M:%S", time.localtime())
success = [
origin,
url_,
info_source_code,
'采集成功',
num_caiji,
time_now,
]
#成功信息保存
insertSql = f"insert into WeixinGZH (site_name,site_url,info_source_code,success_info,success_num,create_time) values (%s,%s,%s,%s,%s,%s)"
cursor_.execute(insertSql, tuple(success))
cnx_.commit()
# 该公众号的所有文章采集完成
log.info(f'---第{page}页采集到文章个数---{len(list_all_info)}---{num_caiji}---耗时{baseCore.getTimeCost(start_,time.time())}')
else:
log.info(f'----第{page}页采集到文章个数{num_caiji}--网址已存在！-----耗时{baseCore.getTimeCost(start_,time.time())}')
return count
except Exception as e:
# json解析该公众号成功但采集数据失败
time_now = time.strftime("%Y-%m-%d %H:%M:%S", time.localtime())
success = [
false = [
origin,
url_,
info_source_code,
'采集成功',
num_caiji,
e,
'采集失败',
time_now
]
#成功信息保存
insertSql = f"insert into WeixinGZH (site_name,site_url,info_source_code,success_info,success_num,create_time) values (%s,%s,%s,%s,%s,%s)"
cursor_.execute(insertSql, tuple(success))
# 失败信息保存
insertSql = f"insert into WeixinGZH (site_name,site_url,info_source_code,json_error_info,error_type,create_time) values (%s,%s,%s,%s,%s,%s)"
cursor_.execute(insertSql, tuple(false))
cnx_.commit()
# 该公众号的所有文章采集完成
log.info(f'{fakeid}、公众号{origin}:采集成功！、已采集{count}个公众号、耗时{baseCore.getTimeCost(start_,time.time())}')
else:
log.info(f'{fakeid}、公众号{origin}、网址已存在！耗时{baseCore.getTimeCost(start_,time.time())}')
except Exception as e:
# json解析该公众号成功但采集数据失败
count += 1
time_now = time.strftime("%Y-%m-%d %H:%M:%S", time.localtime())
false = [
origin,
url_,
info_source_code,
e,
'采集失败',
time_now
]
# 失败信息保存
insertSql = f"insert into WeixinGZH (site_name,site_url,info_source_code,json_error_info,error_type,create_time) values (%s,%s,%s,%s,%s,%s)"
cursor_.execute(insertSql, tuple(false))
cnx_.commit()
log.info(f'{fakeid}、公众号：{origin}采集失败！！！！！！耗时{baseCore.getTimeCost(start_, time.time())}')
log.info(f'{fakeid}、公众号：{origin}采集失败！！！！！！耗时{baseCore.getTimeCost(start_, time.time())}')
count += 1
log.info(f'{fakeid}、公众号{origin}:采集成功！、已采集{count}个公众号、耗时{baseCore.getTimeCost(start_, time.time())}')
time.sleep(2)
return count
if __name__=="__main__":
requests.DEFAULT_RETRIES = 5
time_start = time.strftime("%Y-%m-%d %H:%M:%S", time.localtime())
......@@ -456,15 +468,19 @@ if __name__=="__main__":
# 设置计数器的初始值为0
key = baseCore.getNextSeq()
r.set(key, 0)
# 设置key的过期时间为10秒
r.expire(key, 3600)
# 设置key的过期时间为一个半小时
r.expire(key, 5400)
while True:
time.sleep(2)
new_value = baseCore.incrSet(key)
baseCore.getttl(key)
if new_value < 50:
aa = job(count,key)
count = aa
try:
aa = job(count,key)
count = aa
time.sleep(20)
except:
time.sleep(10)
else:
#刷新浏览器
browser_run = list_b[0]
......
百度采集/baidu_comm/baiduSpider.py
#coding=utf-8
#coding=utf-8
......@@ -266,13 +266,13 @@ class BaiduSpider(object):
break
for detail in lists:
publishTag=detail['publishTag']
if publishTag:
pubtime = datetime.datetime.strptime(publishTag, "%Y-%m-%d %H:%M:%S")
needDate='2022-01-01 00:00:00'
needTime = datetime.datetime.strptime(needDate, "%Y-%m-%d %H:%M:%S")
if pubtime < needTime:
timeFlag = True
break
# if publishTag:
# pubtime = datetime.datetime.strptime(publishTag, "%Y-%m-%d %H:%M:%S")
# needDate='2022-01-01 00:00:00'
# needTime = datetime.datetime.strptime(needDate, "%Y-%m-%d %H:%M:%S")
# if pubtime < needTime:
# timeFlag = True
# break
is_member = self.r.sismember('pybaidu_baidu_'+self.wordsCode, durl)
if is_member:
continue
......@@ -398,7 +398,7 @@ class BaiduSpider(object):
processitem=self.getProcessitem(bdetail)
try:
self.sendkafka(processitem)
self.r.sadd('pybaidu_test_'+self.wordsCode, processitem['sourceAddress'])
self.r.sadd('pybaidu_baidu_'+self.wordsCode, processitem['sourceAddress'])
except Exception as e:
self.logger.info("放入kafka失败！")
#插入数据库
......
百度采集/baidu_comm/baidutaskJob_loc.py
# -*- coding: utf-8 -*-
# -*- coding: utf-8 -*-
......@@ -190,29 +190,7 @@ if __name__ == '__main__':
while True:
try:
codeList=[]
codeList.append('KW-20221114-0007')
codeList.append('KW-20221114-0006')
codeList.append('KW-20221114-0005')
codeList.append('KW-20221114-0009')
codeList.append('KW-20221114-0011')
codeList.append('KW-20221114-0012')
codeList.append('KW-20221114-0013')
codeList.append('KW-20221114-0014')
codeList.append('KW-20221114-0018')
codeList.append('KW-20221213-0006')
codeList.append('KW-20221114-0008')
codeList.append('KW-20221114-0015')
codeList.append('KW-20221114-0016')
codeList.append('KW-20221114-0017')
codeList.append('KW-20221114-0019')
codeList.append('KW-20221114-0022')
codeList.append('KW-20221114-0023')
codeList.append('KW-20221114-0024')
codeList.append('KW-20221114-0025')
codeList.append('KW-20221114-0026')
codeList.append('KW-20221114-0027')
codeList.append('KW-20221114-0020')
codeList.append('KW-20221114-0021')
codeList.append('KW-20230818-0003')
for codeid in codeList:
try:
# keymsg=baiduTaskJob.getkafka()
......
百度采集/baidu_comm/requirements.txt
......@@ -12,6 +12,9 @@ pip install tqdm -i https://pypi.douban.com/simple
pip install goose3 -i https://mirrors.aliyun.com/pypi/simple
pip install Beautifulsoup4 -i https://mirrors.aliyun.com/pypi/simple
pip install langid -i https://mirrors.aliyun.com/pypi/simple/
pip install jieba -i https://mirrors.aliyun.com/pypi/simple
selenium==3.141.0
selenium-wire==5.1.0
......
Markdown 格式
0% 或
您添加了 0 人 到此讨论。请谨慎行事。
请先完成此评论的编辑！
请 注册 或者 后发表评论