def __init__(self):
# 抓取设置
self.crawler = TCCrawler()
self.retrycrawler = RetryCrawler()
self.crawling_time = Common.now() # 当前爬取时间
self.crawling_time_s = Common.time_s(self.crawling_time)
self.crawling_begintime = '' # 本次抓取开始时间
self.crawling_beginDate = '' # 本次爬取日期
self.crawling_beginHour = '' # 本次爬取小时
# 频道信息
self.platform = '同程-pc' # 品牌团所在平台
self.channel_id = '' # 频道id
self.channel_url = '' # 频道链接
self.channel_name = '' # 频道name
self.channel_type = '' # 频道类型
# 原数据信息
self.channel_page = '' # 频道页面html内容
self.channel_pages = {} # 频道页面内请求数据列表
# channel items
self.channel_items = []
# channel list
self.channel_list = []
def __init__(self):
# 抓取设置
self.crawler = TCCrawler()
# dial client
self.dial_client = DialClient()
# local ip
self._ip = Common.local_ip()
# router tag
self._tag = 'ikuai'
# wait time
self.w_time = 1
def __init__(self):
# 商品页面抓取设置
self.crawler = TCCrawler()
self.crawling_time = Common.now() # 当前爬取时间
self.crawling_begintime = '' # 本次抓取开始时间
self.crawling_beginDate = '' # 本次爬取日期
self.crawling_beginHour = '' # 本次爬取小时
# 单品类型商品所属频道
self.channel_id = ''
self.channel_name = ''
self.channel_url = ''
self.channel_type = ''
self.item_position = 0
# 商品信息
self.item_id = '' # 商品Id
self.item_url = '' # 商品链接
self.item_pic_url = '' # 商品展示图片链接
self.item_name = '' # 商品Name
self.item_desc = '' # 商品说明
self.item_book_status = 1 # 商品是否售卖 0:不售,1:在售
self.item_level = '' # 级别
self.item_area = '' # 地址
self.item_service = '' # 服务
self.item_comment = '' # 评论数
self.item_comment_rate = '' # 好评率
self.item_comment_good = '' # 好评数
# 商品交易
self.item_oriprice = '' # 商品原价
self.item_disprice = '' # 商品折扣价
self.item_discount = '' # 商品打折
# 门票
self.item_tickets = []
# 原数据信息
self.item_pageData = '' # 商品所属数据项内容
self.item_page = '' # 商品页面html内容
self.item_pages = {} # 商品页面内请求数据列表
def __init__(self):
# 抓取设置
self.crawler = TCCrawler()
# dial client
self.dial_client = DialClient()
# local ip
self._ip = Common.local_ip()
# router tag
self._tag = 'ikuai'
# wait time
self.w_time = 1
def __init__(self):
# tc spot type
self.worker_type = Config.TC_Spot
# DB
self.tc_type = Config.TC_TYPE # queue type
self.mysqlAccess = MysqlAccess() # mysql access
self.redisQueue = RedisQueue() # redis queue
self.mongofsAccess = MongofsAccess() # mongodb fs access
# 抓取设置
self.crawler = TCCrawler()
# message
self.message = Message()
# 抓取时间设定
self.crawling_time = Common.now() # 当前爬取时间
self.begin_time = Common.now()
self.begin_date = Common.today_s()
self.begin_hour = Common.nowhour_s()
def __init__(self):
# 商品页面抓取设置
self.crawler = TCCrawler()
self.crawling_time = Common.now() # 当前爬取时间
self.crawling_begintime = '' # 本次抓取开始时间
self.crawling_beginDate = '' # 本次爬取日期
self.crawling_beginHour = '' # 本次爬取小时
# 单品类型商品所属频道
self.channel_id = ''
self.channel_name = ''
self.channel_url = ''
self.channel_type = ''
self.item_position = 0
# 商品信息
self.item_id = '' # 商品Id
self.item_url = '' # 商品链接
self.item_pic_url = '' # 商品展示图片链接
self.item_name = '' # 商品Name
self.item_desc = '' # 商品说明
self.item_book_status = 1 # 商品是否售卖 0:不售,1:在售
self.item_level = '' # 级别
self.item_area = '' # 地址
self.item_service = '' # 服务
self.item_comment = '' # 评论数
self.item_comment_rate = '' # 好评率
self.item_comment_good = '' # 好评数
# 商品交易
self.item_oriprice = '' # 商品原价
self.item_disprice = '' # 商品折扣价
self.item_discount = '' # 商品打折
# 门票
self.item_tickets = []
# 原数据信息
self.item_pageData = '' # 商品所属数据项内容
self.item_page = '' # 商品页面html内容
self.item_pages = {} # 商品页面内请求数据列表
class Item():
'''A class of tc Item'''
def __init__(self):
# 商品页面抓取设置
self.crawler = TCCrawler()
self.crawling_time = Common.now() # 当前爬取时间
self.crawling_begintime = '' # 本次抓取开始时间
self.crawling_beginDate = '' # 本次爬取日期
self.crawling_beginHour = '' # 本次爬取小时
# 单品类型商品所属频道
self.channel_id = ''
self.channel_name = ''
self.channel_url = ''
self.channel_type = ''
self.item_position = 0
# 商品信息
self.item_id = '' # 商品Id
self.item_url = '' # 商品链接
self.item_pic_url = '' # 商品展示图片链接
self.item_name = '' # 商品Name
self.item_desc = '' # 商品说明
self.item_book_status = 1 # 商品是否售卖 0:不售,1:在售
self.item_level = '' # 级别
self.item_area = '' # 地址
self.item_service = '' # 服务
self.item_comment = '' # 评论数
self.item_comment_rate = '' # 好评率
self.item_comment_good = '' # 好评数
# 商品交易
self.item_oriprice = '' # 商品原价
self.item_disprice = '' # 商品折扣价
self.item_discount = '' # 商品打折
# 门票
self.item_tickets = []
# 原数据信息
self.item_pageData = '' # 商品所属数据项内容
self.item_page = '' # 商品页面html内容
self.item_pages = {} # 商品页面内请求数据列表
# 商品页信息
def spotConfig(self, _val):
self.item_book_status, self.item_id, self.item_url, self.item_pic_url, self.item_name, self.item_desc, self.item_level, self.item_area, self.item_position, self.crawling_begintime = _val
# 本次抓取开始日期
self.crawling_beginDate = time.strftime(
"%Y-%m-%d", time.localtime(self.crawling_begintime))
# 本次抓取开始小时
self.crawling_beginHour = time.strftime(
"%H", time.localtime(self.crawling_begintime))
if self.item_book_status == 1:
# 商品页信息
self.itemPage()
page = self.item_page
self.item_pages['item-home'] = (self.item_url, self.item_page)
m = re.search(
r'<div class="s-ppp">\s*<div class="s-tp">(.+?)</div>',
page,
flags=re.S)
if m:
i_disprice = re.sub(r'<.+?>', '', m.group(1))
m = re.search(r'(\d+)', i_disprice)
if m:
self.item_disprice = m.group(1)
else:
self.item_disprice = Common.htmlDecode(i_disprice)
m = re.search(
r'<div class="s-ppp">.+?<div class="s-mp"><s>(.+?)</s>\s+<span>(.+?)</span>',
page,
flags=re.S)
if m:
i_oriprice, i_discount = m.group(1), m.group(2)
m = re.search(r'(\d+)', i_oriprice)
if m:
self.item_oriprice = m.group(1)
else:
self.item_oriprice = Common.htmlDecode(i_oriprice)
m = re.search(r'(\d+\.\d+)', i_discount)
if m:
self.item_discount = m.group(1)
else:
m = re.search(r'(\d+)', i_discount)
if m:
self.item_discount = m.group(1)
else:
self.item_discount = i_discount
m = re.search(r'<div class="s-sr">\s+<span.+?>(.+?)</div>',
page,
flags=re.S)
if m:
self.item_service = re.sub(r'<.+?>', '',
';'.join(m.group(1).split()))
self.itemComment()
self.itemTicket()
def itemComment(self):
c_url = 'http://tctj.ly.com/jrec/wlfrec?_dAjax=callback&cid=105&userId=1001&rid=%d&projectId=3&type=1&flag=2&pageSize=10&page=1&callback=tc%s' % (
int(self.item_id), Common.rand_n(11))
#c_url = 'http://tctj.ly.com/jrec/wlfrec?cid=105&userId=1001&rid=%d&projectId=3&type=0&flag=1&pageSize=10&page=1&_dAjax=callback&callback=tc%s' % (int(self.item_id), Common.rand_n(11))
c_page = self.crawler.getData(c_url, self.item_url)
if c_page:
m = re.search(r'"count":.+?"all":"(\d+)"', c_page, flags=re.S)
if m:
self.item_comment = m.group(1)
m = re.search(r'"count":.+?"satisfaction":"(.+?)"',
c_page,
flags=re.S)
if m:
self.item_comment_rate = m.group(1)
m = re.search(r'"count":.+?"good":"(\d+)"', c_page, flags=re.S)
if m:
self.item_comment_good = m.group(1)
def itemTicket(self):
t_url = 'http://www.ly.com/scenery/AjaxHelper/SceneryPriceFrame.aspx?action=GETNEWPRICEFRAMEFORLAST&ids=%d&isSimple=1&isShowAppThree=0&widthtype=1&isGrap=1&nobookid=&isyry=1&YpState=1&lon=null&lat=null' % int(
self.item_id)
result = self.crawler.getData(t_url, self.item_url)
if result:
try:
scenery = json.loads(result)
if scenery.has_key('SceneryPrices'):
scenery_list = scenery['SceneryPrices']
for destination in scenery_list:
if destination.has_key('DestinationId') and int(
destination['DestinationId']) == int(
self.item_id):
if destination.has_key(
'ChannelPriceModelEntityList'):
for pricemodel in destination[
'ChannelPriceModelEntityList']:
if pricemodel.has_key(
'ConsumersTypeId'
) and pricemodel.has_key(
'ConsumersTypeName'
) and pricemodel.has_key(
'ChannelPriceEntityList'):
consumer_type = pricemodel[
'ConsumersTypeId']
consumer_type_name = pricemodel[
'ConsumersTypeName']
t_i = 1
for t_data in pricemodel[
'ChannelPriceEntityList']:
val = (self.item_id,
self.item_name,
self.channel_type,
consumer_type,
consumer_type_name, t_i,
t_data,
self.crawling_begintime)
t = Ticket()
t.antPage(val)
self.item_tickets.append(
t.outSql())
t_i += 1
except Exception as e:
Common.log('# itemTicket,exception err in load json: %s' % e)
Common.traceback_log()
# 商品详情页html
def itemPage(self):
if self.item_url != '':
refer_url = self.channel_url
page = self.crawler.getData(self.item_url, refer_url)
if type(self.crawler.history) is list and len(
self.crawler.history) != 0 and re.search(
r'302', str(self.crawler.history[0])):
if not self.itempage_judge(page):
Common.log('#crawler history:')
Common.log(self.crawler.history)
raise Common.NoPageException(
"# itemPage: not find item page, redirecting to other page,id:%s,item_url:%s"
% (str(self.item_id), self.item_url))
if not page or page == '':
Common.log('#crawler history:')
Common.log(self.crawler.history)
raise Common.InvalidPageException(
"# itemPage: find item page empty,id:%s,item_url:%s" %
(str(self.item_id), self.item_url))
self.item_page = page
else:
raise Common.NoPageException(
"# itemPage: not find item page, url is null,id:%s,item_url:%s"
% (str(self.item_id), self.item_url))
# 执行
def antPage(self, val):
self.channel_id, self.channel_name, self.channel_url, self.channel_type, i_val = val
if self.channel_type == 1:
self.spotConfig(i_val)
def outTuple(self):
return (self.channel_id, self.channel_name, self.channel_url,
self.channel_type, self.item_position, self.item_book_status,
self.item_id, self.item_url, self.item_pic_url, self.item_name,
self.item_desc, self.item_level, self.item_area,
self.item_service, self.item_comment, self.item_comment_good,
self.item_comment_rate, self.item_oriprice, self.item_disprice,
self.item_discount, self.crawling_beginDate,
self.crawling_beginHour)
def outSql(self):
return (Common.time_s(float(self.crawling_time)), str(self.item_id),
self.item_name, self.item_desc, self.item_url,
self.item_pic_url, str(self.item_book_status), self.item_level,
self.item_area, self.item_service, str(self.item_comment),
str(self.item_comment_good), self.item_comment_rate,
str(self.item_oriprice), str(self.item_disprice),
str(self.item_discount), str(self.channel_id),
str(self.item_position), self.crawling_beginDate,
self.crawling_beginHour)
class Item():
'''A class of tc Item'''
def __init__(self):
# 商品页面抓取设置
self.crawler = TCCrawler()
self.crawling_time = Common.now() # 当前爬取时间
self.crawling_begintime = '' # 本次抓取开始时间
self.crawling_beginDate = '' # 本次爬取日期
self.crawling_beginHour = '' # 本次爬取小时
# 单品类型商品所属频道
self.channel_id = ''
self.channel_name = ''
self.channel_url = ''
self.channel_type = ''
self.item_position = 0
# 商品信息
self.item_id = '' # 商品Id
self.item_url = '' # 商品链接
self.item_pic_url = '' # 商品展示图片链接
self.item_name = '' # 商品Name
self.item_desc = '' # 商品说明
self.item_book_status = 1 # 商品是否售卖 0:不售,1:在售
self.item_level = '' # 级别
self.item_area = '' # 地址
self.item_service = '' # 服务
self.item_comment = '' # 评论数
self.item_comment_rate = '' # 好评率
self.item_comment_good = '' # 好评数
# 商品交易
self.item_oriprice = '' # 商品原价
self.item_disprice = '' # 商品折扣价
self.item_discount = '' # 商品打折
# 门票
self.item_tickets = []
# 原数据信息
self.item_pageData = '' # 商品所属数据项内容
self.item_page = '' # 商品页面html内容
self.item_pages = {} # 商品页面内请求数据列表
# 商品页信息
def spotConfig(self, _val):
self.item_book_status, self.item_id, self.item_url, self.item_pic_url, self.item_name, self.item_desc, self.item_level, self.item_area, self.item_position, self.crawling_begintime = _val
# 本次抓取开始日期
self.crawling_beginDate = time.strftime("%Y-%m-%d", time.localtime(self.crawling_begintime))
# 本次抓取开始小时
self.crawling_beginHour = time.strftime("%H", time.localtime(self.crawling_begintime))
if self.item_book_status == 1:
# 商品页信息
self.itemPage()
page = self.item_page
self.item_pages['item-home'] = (self.item_url, self.item_page)
m = re.search(r'<div class="s-ppp">\s*<div class="s-tp">(.+?)</div>', page, flags=re.S)
if m:
i_disprice = re.sub(r'<.+?>', '', m.group(1))
m = re.search(r'(\d+)', i_disprice)
if m:
self.item_disprice = m.group(1)
else:
self.item_disprice = Common.htmlDecode(i_disprice)
m = re.search(r'<div class="s-ppp">.+?<div class="s-mp"><s>(.+?)</s>\s+<span>(.+?)</span>', page, flags=re.S)
if m:
i_oriprice, i_discount = m.group(1), m.group(2)
m = re.search(r'(\d+)', i_oriprice)
if m:
self.item_oriprice = m.group(1)
else:
self.item_oriprice = Common.htmlDecode(i_oriprice)
m = re.search(r'(\d+\.\d+)', i_discount)
if m:
self.item_discount = m.group(1)
else:
m = re.search(r'(\d+)', i_discount)
if m:
self.item_discount = m.group(1)
else:
self.item_discount = i_discount
m = re.search(r'<div class="s-sr">\s+<span.+?>(.+?)</div>', page, flags=re.S)
if m:
self.item_service = re.sub(r'<.+?>', '', ';'.join(m.group(1).split()))
self.itemComment()
self.itemTicket()
def itemComment(self):
c_url = 'http://tctj.ly.com/jrec/wlfrec?_dAjax=callback&cid=105&userId=1001&rid=%d&projectId=3&type=1&flag=2&pageSize=10&page=1&callback=tc%s' % (int(self.item_id), Common.rand_n(11))
#c_url = 'http://tctj.ly.com/jrec/wlfrec?cid=105&userId=1001&rid=%d&projectId=3&type=0&flag=1&pageSize=10&page=1&_dAjax=callback&callback=tc%s' % (int(self.item_id), Common.rand_n(11))
c_page = self.crawler.getData(c_url, self.item_url)
if c_page:
m = re.search(r'"count":.+?"all":"(\d+)"', c_page, flags=re.S)
if m:
self.item_comment = m.group(1)
m = re.search(r'"count":.+?"satisfaction":"(.+?)"', c_page, flags=re.S)
if m:
self.item_comment_rate = m.group(1)
m = re.search(r'"count":.+?"good":"(\d+)"', c_page, flags=re.S)
if m:
self.item_comment_good = m.group(1)
def itemTicket(self):
t_url = 'http://www.ly.com/scenery/AjaxHelper/SceneryPriceFrame.aspx?action=GETNEWPRICEFRAMEFORLAST&ids=%d&isSimple=1&isShowAppThree=0&widthtype=1&isGrap=1&nobookid=&isyry=1&YpState=1&lon=null&lat=null' % int(self.item_id)
result = self.crawler.getData(t_url, self.item_url)
if result:
try:
scenery = json.loads(result)
if scenery.has_key('SceneryPrices'):
scenery_list = scenery['SceneryPrices']
for destination in scenery_list:
if destination.has_key('DestinationId') and int(destination['DestinationId']) == int(self.item_id):
if destination.has_key('ChannelPriceModelEntityList'):
for pricemodel in destination['ChannelPriceModelEntityList']:
if pricemodel.has_key('ConsumersTypeId') and pricemodel.has_key('ConsumersTypeName') and pricemodel.has_key('ChannelPriceEntityList'):
consumer_type = pricemodel['ConsumersTypeId']
consumer_type_name = pricemodel['ConsumersTypeName']
t_i = 1
for t_data in pricemodel['ChannelPriceEntityList']:
val = (self.item_id, self.item_name, self.channel_type, consumer_type, consumer_type_name, t_i, t_data, self.crawling_begintime)
t = Ticket()
t.antPage(val)
self.item_tickets.append(t.outSql())
t_i += 1
except Exception as e:
Common.log('# itemTicket,exception err in load json: %s' % e)
Common.traceback_log()
# 商品详情页html
def itemPage(self):
if self.item_url != '':
refer_url = self.channel_url
page = self.crawler.getData(self.item_url, refer_url)
if type(self.crawler.history) is list and len(self.crawler.history) != 0 and re.search(r'302',str(self.crawler.history[0])):
if not self.itempage_judge(page):
Common.log('#crawler history:')
Common.log(self.crawler.history)
raise Common.NoPageException("# itemPage: not find item page, redirecting to other page,id:%s,item_url:%s"%(str(self.item_id), self.item_url))
if not page or page == '':
Common.log('#crawler history:')
Common.log(self.crawler.history)
raise Common.InvalidPageException("# itemPage: find item page empty,id:%s,item_url:%s"%(str(self.item_id), self.item_url))
self.item_page = page
else:
raise Common.NoPageException("# itemPage: not find item page, url is null,id:%s,item_url:%s"%(str(self.item_id), self.item_url))
# 执行
def antPage(self, val):
self.channel_id, self.channel_name, self.channel_url, self.channel_type, i_val = val
if self.channel_type == 1:
self.spotConfig(i_val)
def outTuple(self):
return (self.channel_id, self.channel_name, self.channel_url, self.channel_type, self.item_position, self.item_book_status, self.item_id, self.item_url, self.item_pic_url, self.item_name, self.item_desc, self.item_level, self.item_area, self.item_service, self.item_comment, self.item_comment_good, self.item_comment_rate, self.item_oriprice, self.item_disprice, self.item_discount, self.crawling_beginDate, self.crawling_beginHour)
def outSql(self):
return (Common.time_s(float(self.crawling_time)), str(self.item_id), self.item_name, self.item_desc, self.item_url, self.item_pic_url, str(self.item_book_status), self.item_level, self.item_area, self.item_service, str(self.item_comment), str(self.item_comment_good), self.item_comment_rate, str(self.item_oriprice), str(self.item_disprice), str(self.item_discount), str(self.channel_id), str(self.item_position), self.crawling_beginDate, self.crawling_beginHour)
class RetryCrawler():
'''A class of retry crawl data'''
def __init__(self):
# 抓取设置
self.crawler = TCCrawler()
# dial client
self.dial_client = DialClient()
# local ip
self._ip = Common.local_ip()
# router tag
self._tag = 'ikuai'
# wait time
self.w_time = 1
# To dial router
def dialRouter(self, _type, _obj):
try:
_module = '%s_%s' % (_type, _obj)
self.dial_client.send((_module, self._ip, self._tag))
except Exception as e:
Common.log('# To dial router exception: %s' % e)
def getData(self, url, refers='', max_retry=20):
page = ''
retry = 1
while True:
try:
page = self.crawler.getData(url, refers)
break
except Common.InvalidPageException as e:
if retry >= max_retry:
break
retry += 1
Common.log('# Invalid page exception: %s' % e)
time.sleep(self.w_time * retry)
except Common.DenypageException as e:
if retry >= max_retry:
break
retry += 1
Common.log('# Deny page exception: %s' % e)
# 重新拨号
try:
self.dialRouter(4, 'chn')
except Exception as e:
Common.log('# DailClient Exception err: %s' % e)
time.sleep(random.uniform(10, 30))
except Common.SystemBusyException as e:
if retry >= max_retry:
break
retry += 1
Common.log('# System busy exception: %s' % e)
time.sleep(self.w_time * retry)
except Exception as e:
Common.log('# exception err in retry crawler: %s' % e)
if str(e).find('Read timed out') != -1:
if retry >= max_retry:
break
retry += 1
time.sleep(random.uniform(1, 3))
elif str(e).find('Name or service not known') != -1 or str(
e).find('Temporary failure in name resolution'):
if retry >= max_retry:
break
retry += 1
# 重新拨号
try:
self.dialRouter(4, 'chn')
except Exception as e:
Common.log('# DailClient Exception err: %s' % e)
time.sleep(random.uniform(10, 30))
else:
break
return page
class Channel():
'''A class of TC channel'''
def __init__(self):
# 抓取设置
self.crawler = TCCrawler()
self.retrycrawler = RetryCrawler()
self.crawling_time = Common.now() # 当前爬取时间
self.crawling_time_s = Common.time_s(self.crawling_time)
self.crawling_begintime = '' # 本次抓取开始时间
self.crawling_beginDate = '' # 本次爬取日期
self.crawling_beginHour = '' # 本次爬取小时
# 频道信息
self.platform = '同程-pc' # 品牌团所在平台
self.channel_id = '' # 频道id
self.channel_url = '' # 频道链接
self.channel_name = '' # 频道name
self.channel_type = '' # 频道类型
# 原数据信息
self.channel_page = '' # 频道页面html内容
self.channel_pages = {} # 频道页面内请求数据列表
# channel items
self.channel_items = []
# channel list
self.channel_list = []
# 频道页初始化
def init(self, channel_id, channel_url, channel_type, begin_time):
self.channel_id = channel_id
self.channel_url = channel_url
self.channel_type = channel_type
self.crawling_begintime = begin_time
self.crawling_beginDate = time.strftime("%Y-%m-%d", time.localtime(self.crawling_begintime))
self.crawling_beginHour = time.strftime("%H", time.localtime(self.crawling_begintime))
def config(self):
self.channelPage()
if self.channel_type == 1:
self.spot()
#elif self.channel_type == 2:
else:
Common.log('# not find this channel type...')
def spot(self):
if self.channel_page:
m = re.search(r'<title>(.+?)</title>', self.channel_page, flags=re.S)
if m:
self.channel_name = m.group(1)
keyword, pid, cid, cyid = '', 0, 0, 0
m = re.search(r'<span id="hdKeyWord">(.*?)</span>', self.channel_page, flags=re.S)
if m:
keyword = m.group(1)
m = re.search(r'<span id="hdPid">(.*?)</span>', self.channel_page, flags=re.S)
if m:
pid = int(m.group(1))
m = re.search(r'<span id="hdCid">(.*?)</span>', self.channel_page, flags=re.S)
if m:
cid = int(m.group(1))
m = re.search(r'<span id="hdCyid">(.*?)</span>', self.channel_page, flags=re.S)
if m:
cyid = int(m.group(1))
i_p = 1
i_page = 1
m_page = 1
page_main = ''
m = re.search(r'<div class="scenery_main" id="sceneryListInfo">(.+?)<div id="pageNum_box" class="s_pager none">', self.channel_page, flags=re.S)
if m:
page_main = m.group(1)
else:
page_main = self.channel_page
Common.log(i_page)
i_p = self.get_items(page_main, i_p)
m = re.search(r'<input type="hidden" id="txt_AllpageNumber" value="(.+?)">', page_main, flags=re.S)
if m:
m_page = int(m.group(1))
page_url = 'http://www.ly.com/scenery/SearchList.aspx?&action=getlist&page=%d&kw=&pid=%d&cid=%d&cyid=%d&theme=0&grade=0&money=0&sort=0&paytype=0&ismem=0&istuan=0&isnow=0&spType=&isyiyuan=0&lbtypes=&IsNJL=0&classify=0'
while i_page < m_page:
i_page += 1
p_url = page_url % (i_page, pid, cid, cyid)
Common.log(i_page)
page = self.retrycrawler.getData(p_url, self.channel_url)
i_p = self.get_items(page, i_p)
def get_items(self, page_main, i_p):
if page_main:
p = re.compile(r'<div class="scenery_list(.+?)">\s*<div class="s_info"><div class="img_con"><a class="a_img".+?href="(.+?)"><img.+?src="(.+?)".*?></a></div><div class="info_con"><dl class="info_top"><dt><a class="fir_name".+?>(.+?)</a>.+?<span class="s_level">(.*?)</span>.+?<dd class="scenery_area"><span>(.+?)</span>.+?</dl></div></div>', flags=re.S)
for info in p.finditer(page_main):
all_info, i_info, i_url, i_img, i_name, i_level, i_area = info.group(), info.group(1), (Config.tc_home + info.group(2)), info.group(3), info.group(4), re.sub(r'<.+?>', '', info.group(5)), ' '.join(info.group(6).split())
i_book = 1
i_desc = ''
m = re.search(r'<dd class="scenery_desc"><p>(.+?)</p>', all_info, flags=re.S)
if m:
i_desc = m.group(1)
if i_info.find('nobook') != -1:
i_book = 0
if i_desc == '':
m = re.search(r'<dd class="scenery_state">(.+?)<a', all_info, flags=re.S)
if m:
i_desc = m.group(1)
i_id = 0
if i_url != '':
m = re.search(r'BookSceneryTicket_(\d+).html', i_url)
if m:
i_id = m.group(1)
val = (self.channel_id, self.channel_name, self.channel_url, self.channel_type, (i_book, i_id, i_url, i_img, i_name, i_desc, i_level, i_area, i_p, self.crawling_begintime))
self.channel_items.append(val)
#if i_p == 1: Common.log(val)
i_p += 1
return i_p
def channelList(self):
self.channelPage()
if self.channel_page:
city_list = self.moreCity(self.channel_page, 'city')
for city in city_list:
city_url, city_name, province_id, city_id, dcity_id = city
if city_url:
province_name, city_name = '', ''
city_page = self.crawler.getData(city_url, self.channel_url)
if city_page:
m = re.search(r'<div class="search_screen_dl"><dl action="province">.+?<div class="right"><a href=".+?" class="current" tvalue="(\d+)" title="(.+?)">', city_page, flags=re.S)
if m:
province_id, province_name = m.group(1), m.group(2)
m = re.search(r'<div class="search_screen_dl">.+?<dl action="city">.+?<div class="right"><a href=".+?" class="current" tvalue="(\d+)" title="(.+?)">', city_page, flags=re.S)
if m:
city_id, city_name = m.group(1), m.group(2)
channel_list = self.moreCity(city_page, 'district')
if channel_list and len(channel_list) > 0:
for channel in channel_list:
channel_url, channel_name, p_id, c_id, cy_id = channel
self.channel_list.append((str(cy_id), channel_name, channel_url, str(self.channel_type), city_id, city_name, province_id, province_name))
def moreCity(self, page, action_key):
city_list = []
if page:
p_id, c_id, cy_id = 0, 0, 0
m = re.search(r'<span id="hdPid">(.*?)</span>', page, flags=re.S)
if m:
p_id = int(m.group(1))
m = re.search(r'<span id="hdCid">(.*?)</span>', page, flags=re.S)
if m:
c_id = int(m.group(1))
m = re.search(r'<span id="hdCyid">(.*?)</span>', page, flags=re.S)
if m:
cy_id = int(m.group(1))
p_url = 'http://www.ly.com/scenery/scenerysearchlist_%d_%d__0_0_0_%d_0_0_0.html'
m = re.search(r'<div class="search_screen_box" id="searchScreenBox">.+?<dl action="%s".+?>.+?<dd>.+?<div class="right">(.+?)</div></dd>' % action_key, page, flags=re.S)
if m:
city_infos = m.group(1)
p = re.compile(r'<a.+?tvalue="(\d+)" title="(.+?)">', flags=re.S)
for city in p.finditer(city_infos):
city_id, city_name = int(city.group(1)), city.group(2)
if action_key == 'city':
c_id = city_id
elif action_key == 'district':
cy_id = city_id
city_url, city_name = p_url % (p_id, c_id, cy_id), city.group(2)
city_list.append((city_url, city_name, p_id, c_id, cy_id))
return city_list
def channelPage(self):
if self.channel_url:
data = self.crawler.getData(self.channel_url, Config.tc_home)
if not data and data == '': raise Common.InvalidPageException("# channelPage:not find channel page,channel_id:%s,channel_url:%s"%(str(self.channel_id), self.channel_url))
if data and data != '':
self.channel_page = data
self.channel_pages['channel-home'] = (self.channel_url, data)
def antPage(self, val):
channel_id, channel_url, channel_type, begin_time = val
self.init(channel_id, channel_url, channel_type, begin_time)
self.config()
def antChannelList(self, val):
self.channel_url, self.channel_type = val
self.channelList()
class RetryCrawler():
'''A class of retry crawl data'''
def __init__(self):
# 抓取设置
self.crawler = TCCrawler()
# dial client
self.dial_client = DialClient()
# local ip
self._ip = Common.local_ip()
# router tag
self._tag = 'ikuai'
# wait time
self.w_time = 1
# To dial router
def dialRouter(self, _type, _obj):
try:
_module = '%s_%s' %(_type, _obj)
self.dial_client.send((_module, self._ip, self._tag))
except Exception as e:
Common.log('# To dial router exception: %s' % e)
def getData(self, url, refers='', max_retry=20):
page = ''
retry = 1
while True:
try:
page = self.crawler.getData(url, refers)
break
except Common.InvalidPageException as e:
if retry >= max_retry:
break
retry += 1
Common.log('# Invalid page exception: %s' % e)
time.sleep(self.w_time*retry)
except Common.DenypageException as e:
if retry >= max_retry:
break
retry += 1
Common.log('# Deny page exception: %s' % e)
# 重新拨号
try:
self.dialRouter(4, 'chn')
except Exception as e:
Common.log('# DailClient Exception err: %s' % e)
time.sleep(random.uniform(10,30))
except Common.SystemBusyException as e:
if retry >= max_retry:
break
retry += 1
Common.log('# System busy exception: %s' % e)
time.sleep(self.w_time*retry)
except Exception as e:
Common.log('# exception err in retry crawler: %s' % e)
if str(e).find('Read timed out') != -1:
if retry >= max_retry:
break
retry += 1
time.sleep(random.uniform(1,3))
elif str(e).find('Name or service not known') != -1 or str(e).find('Temporary failure in name resolution'):
if retry >= max_retry:
break
retry += 1
# 重新拨号
try:
self.dialRouter(4, 'chn')
except Exception as e:
Common.log('# DailClient Exception err: %s' % e)
time.sleep(random.uniform(10,30))
else:
break
return page
