def __init__(self):
# self.queueName = queueName
# 初始化消息队列
self.mqObj = rabbitmqConnection()
self.pushChannel = self.mqObj.getChannel(targetQueue)
# 获取runId
self.run_id = getUUID()
def __init__(self):
# self.queueName = queueName
# 初始化消息队列
self.mqObj = rabbitmqConnection()
# self.myConnection = self.mqObj.getRabbitmqConnection()
self.readChannel = self.mqObj.getChannel(processQueue)
# 获取runId
self.run_id = getUUID()
def __init__(self):
# self.queueName = queueName
# 初始化消息队列
self.mqObj = rabbitmqConnection()
self.channel = self.mqObj.getChannel(queueName)
# 初始化mysql连接
self.mysqlSession = mysqlConnection()
# 获取runId
self.run_id = getUUID()
def parseErrorLog(self, message):
dict = json.loads(message)
run_id = dict.get('run_id')
file_name = dict.get('file_name')
error_message = dict.get('error_message')
create_time = formatDateTime(0)
soup = dict.get('soup')
id = getUUID()
# 异常入库（为本次运行统计做准备）
self.mysqlSession.addOne(
ErrorMessageEntity(id=id,
run_id=run_id,
file_name=file_name,
message=error_message,
create_time=create_time,
soup=soup))
logger.info('已入库，异常id:%s,异常信息:%s' % (id, error_message[0:100]))
def parse(self, message):
dictObj = json.loads(message)
province_id = dictObj.get('province_id')
province_name = dictObj.get('province_name')
city_name = dictObj.get('city_name')
cityUrl = dictObj.get('cityUrl')
cityOnsaleUrl = dictObj.get('cityOnsaleUrl')
# # 目前已知雄安是没有二手在售的
# if city_name == '雄安新区' or city_name =='江阴':
# logger.info('[%s]市跳过，暂不处理')
# return
logger.info('处理[%s]市,在售Url：[%s]' % (city_name, cityOnsaleUrl))
# ，这里比较特殊，因为在bk执行https://diqing.fang.ke.com/loupan/楼盘url的时候，如果不存在会走302跳转，这里不让他跳转，否则后续判断不对
cityOnsaleUrl = 'https://xan.ke.com/ershoufang/'
status_code, text, content = myGetRequest(cityOnsaleUrl,
Config.PROXY_FLAG,
allowRedirects=False)
soup = BeautifulSoup(text, "html.parser")
id = getUUID()
# 插入市到数据库
city_id = regionService.addOrInsert(
RegionEntity(id=id,
pid=province_id,
name=city_name,
level=Config.CITY_LEVEL))
if city_id != id:
logger.warning('城市id：[%s]，name：[%s]，已存在' % (city_id, city_name))
else:
logger.info('城市id：[%s]，name：[%s]，入库成功' % (city_id, city_name))
if status_code == 200:
# 判断当前页面的总套数
totalOnsale = soup.find_all(
'h2',
class_='total fl')[0].find_all('span')[0].getText().strip()
if totalOnsale == '0':
# 有页面但没有在售数据
logger.warning('[%s]市没有在售数据，url：%s' %
(city_name, cityOnsaleUrl))
# 有在售
# 抓区县信息
areaSoup = soup.find_all('a', attrs={'data-click-evtid': '12339'})
for area in areaSoup:
areaUri = area['href']
area_name = area.getText().strip()
areaUrl = cityUrl + areaUri
id = getUUID()
# 插入区域数据库
area_id = regionService.addOrInsert(
RegionEntity(id=id,
pid=province_id,
name=city_name,
level=Config.AREA_LEVEL))
# 组装数据推送到区域队列
dict = {
'province_id': province_id,
'province_name': province_name,
'city_id': city_id,
'city_name': city_name,
'cityUrl': cityUrl,
'area_id': area_id,
'area_name': area_name,
'areaUrl': areaUrl
}
self.pushResult(dict)
else:
logger.info('[%s]市没有在售数据，url：%s' % (city_name, cityOnsaleUrl))
def parse(self, message):
dictObj = json.loads(message)
province_id = dictObj.get('province_id')
province_name = dictObj.get('province_name')
city_id = dictObj.get('city_id')
city_name = dictObj.get('city_name')
cityUrl = dictObj.get('cityUrl')
area_id=dictObj.get('area_id')
area_name = dictObj.get('area_name')
areaUrl = dictObj.get('areaUrl')
logger.info('处理[%s]区,url:[%s] ' % (area_name, areaUrl))
# 打开区县url，抓取商圈信息
status_code, text, content = myGetRequest(areaUrl, Config.PROXY_FLAG)
soup = BeautifulSoup(text,"html.parser")
# 获取到该区下面的全部商圈的url然后推送
townSoupTemp = soup.find_all(attrs={"data-role": "ershoufang"})
# if townSoupTemp == []:
if len(townSoupTemp[0].find_all('div')) == 1:
# 如果是空集合，表示该区县下没有商圈，采用区县名称作为商圈名称
town_name = area_name
id = getUUID()
townSoupTemp = BeautifulSoup(text, "html.parser")
# 插入商圈到数据库
town_id = regionService.addOrInsert(RegionEntity(id=id,
pid=province_id,
name=city_name,
level=Config.TOWN_LEVEL
))
townUrl = areaUrl
logger.info('处理[%s]商圈,url:%s ' % (town_name, townUrl))
# 抓取商圈页面的小区总数-用于计算总页数
totalOnsaleNum = int(
townSoupTemp.find_all('div', class_='total fl')[0].find_all('span')[0].getText().strip())
# 计算总页数
totalPageNum = self.getCurrentPages(totalOnsaleNum)
for x in range(1,totalPageNum+1):
townUrlByPage=townUrl+'/pg'+str(x)
# 组装数据推送到区域队列
dict = {
'province_id': province_id,
'province_name': province_name,
'city_id': city_id,
'city_name': city_name,
'cityUrl': cityUrl,
'area_id': area_id,
'area_name': area_name,
'town_id': town_id,
'town_name':town_name,
'townUrlByPage':townUrlByPage
}
self.pushResult(dict)
else:
for t in townSoupTemp[0].find_all('div')[1].find_all('a'):
town_name = t.getText()
id = getUUID()
townUri = t['href']
townUrl = cityUrl + townUri
townSoupTemp = soup
# 插入商圈到数据库
town_id = regionService.addOrInsert(RegionEntity(id=id,
pid=province_id,
name=city_name,
level=Config.TOWN_LEVEL
))
# 抓取商圈页面的小区总数-用于计算总页数
totalOnsaleNum = int(
townSoupTemp.find_all('div', class_='total fl')[0].find_all('span')[0].getText())
# 计算总页数
totalPageNum = self.getCurrentPages(totalOnsaleNum)
logger.info('处理[%s]商圈,url:%s,获取到[%d]页数据 ' % (town_name, townUrl,totalPageNum))
for x in range(1, totalPageNum + 1):
townUrlByPage = townUrl + 'pg' + str(x)
# 组装数据推送到区域队列
dict = {
'province_id': province_id,
'province_name': province_name,
'city_id': city_id,
'city_name': city_name,
'area_id': area_id,
'area_name': area_name,
'town_id': town_id,
'town_name': town_name,
'townUrlByPage': townUrlByPage
}
self.pushResult(dict)
