def __init__(self):
# 抓取设置
self.crawler = XCCrawler()
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 = '携程-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
# 频道所属地理位置信息
self.province_id = 0 # 省,州id
self.province_name = '' # 省,州名称
# 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):
# mysql
self.mysqlAccess = MysqlAccess()
# 抓取设置
#self.crawler = TBCrawler()
self.crawler = RetryCrawler()
# 页面模板解析
self.brand_temp = JHSBrandTEMP()
# 获取Json数据
self.jsonpage = Jsonpage()
# 首页的品牌团列表
self.home_brands = {}
# 品牌团页面的最上面推广位
self.top_brands = {}
# 页面信息
self.ju_home_page = '' # 聚划算首页
self.ju_brand_page = '' # 聚划算品牌团页面
# 抓取开始时间
self.begin_time = Common.now()
def __init__(self):
# 抓取设置
self.crawler = XCCrawler()
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 = '携程-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
# 频道所属地理位置信息
self.province_id = 0 # 省,州id
self.province_name = '' # 省,州名称
# 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, m_type):
# 抓取设置
self.crawler = RetryCrawler()
# DB
self.mysqlAccess = MysqlAccess() # mysql access
# cat queue
self.cat_queue = JHSQ('cat','main')
# act queue
self.act_queue = JHSQ('act','main')
self.work = JHSWorker()
# 默认类别
#self.category_list = [("http://ju.taobao.com/jusp/nvzhuangpindao/tp.htm#J_FixedNav","女装","1000")]
self.category_list = [
("http://ju.taobao.com/jusp/nvzhuangpindao/tp.htm#J_FixedNav","女装","1000"),
("http://ju.taobao.com/jusp/nanzhuangpindao/tp.htm#J_FixedNav","男装","7000"),
("http://ju.taobao.com/jusp/xiebaopindao/tp.htm#J_FixedNav","鞋包","3000"),
("http://ju.taobao.com/jusp/neiyipindao/tp.htm#J_FixedNav","内衣","4000"),
("http://ju.taobao.com/jusp/zhubaoshipin/tp.htm#J_FixedNav","饰品","42000"),
("http://ju.taobao.com/jusp/yundongpindao/tp.htm#J_FixedNav","运动","38000"),
("http://ju.taobao.com/jusp/meizhuangpindao/tp.htm#J_FixedNav","美妆","2000"),
("http://ju.taobao.com/jusp/tongzhuangpindao/tp.htm#J_FixedNav","童装","23000"),
("http://ju.taobao.com/jusp/shipinpindao/tp.htm#J_FixedNav","零食","5000"),
("http://ju.taobao.com/jusp/muyingpindao/tp.htm#J_FixedNav","母婴","6000"),
("http://ju.taobao.com/jusp/baihuopindao/tp.htm#J_FixedNav","百货","37000"),
("http://ju.taobao.com/jusp/chepinpindao/tp.htm#J_FixedNav","汽车","36000"),
("http://ju.taobao.com/jusp/jiadianpindao/tp.htm#J_FixedNav","家电","34000"),
("http://ju.taobao.com/jusp/shumapindao/tp.htm#J_FixedNav","数码","43000"),
("http://ju.taobao.com/jusp/jiajunewpindao/tp.htm#J_FixedNav","家装","225000"),
("http://ju.taobao.com/jusp/jiajupindao/tp.htm#J_FixedNav","家纺","35000")
]
# 页面
self.site_page = None
# 抓取开始时间
self.begin_time = Common.now()
# 分布式主机标志
self.m_type = m_type
class JHSBrand():
'''A class of JHS category channel'''
def __init__(self, m_type):
# 抓取设置
self.crawler = RetryCrawler()
# DB
self.mysqlAccess = MysqlAccess() # mysql access
# cat queue
self.cat_queue = JHSQ('cat','main')
# act queue
self.act_queue = JHSQ('act','main')
self.work = JHSWorker()
# 默认类别
#self.category_list = [("http://ju.taobao.com/jusp/nvzhuangpindao/tp.htm#J_FixedNav","女装","1000")]
self.category_list = [
("http://ju.taobao.com/jusp/nvzhuangpindao/tp.htm#J_FixedNav","女装","1000"),
("http://ju.taobao.com/jusp/nanzhuangpindao/tp.htm#J_FixedNav","男装","7000"),
("http://ju.taobao.com/jusp/xiebaopindao/tp.htm#J_FixedNav","鞋包","3000"),
("http://ju.taobao.com/jusp/neiyipindao/tp.htm#J_FixedNav","内衣","4000"),
("http://ju.taobao.com/jusp/zhubaoshipin/tp.htm#J_FixedNav","饰品","42000"),
("http://ju.taobao.com/jusp/yundongpindao/tp.htm#J_FixedNav","运动","38000"),
("http://ju.taobao.com/jusp/meizhuangpindao/tp.htm#J_FixedNav","美妆","2000"),
("http://ju.taobao.com/jusp/tongzhuangpindao/tp.htm#J_FixedNav","童装","23000"),
("http://ju.taobao.com/jusp/shipinpindao/tp.htm#J_FixedNav","零食","5000"),
("http://ju.taobao.com/jusp/muyingpindao/tp.htm#J_FixedNav","母婴","6000"),
("http://ju.taobao.com/jusp/baihuopindao/tp.htm#J_FixedNav","百货","37000"),
("http://ju.taobao.com/jusp/chepinpindao/tp.htm#J_FixedNav","汽车","36000"),
("http://ju.taobao.com/jusp/jiadianpindao/tp.htm#J_FixedNav","家电","34000"),
("http://ju.taobao.com/jusp/shumapindao/tp.htm#J_FixedNav","数码","43000"),
("http://ju.taobao.com/jusp/jiajunewpindao/tp.htm#J_FixedNav","家装","225000"),
("http://ju.taobao.com/jusp/jiajupindao/tp.htm#J_FixedNav","家纺","35000")
]
# 页面
self.site_page = None
# 抓取开始时间
self.begin_time = Common.now()
# 分布式主机标志
self.m_type = m_type
def antPage(self):
try:
# 主机器需要配置redis队列
if self.m_type == 'm':
category_list = self.mysqlAccess.selectJhsGroupItemCategory()
if not category_list or len(category_list) == 0:
category_list = self.category_list
if category_list and len(category_list) > 0:
cate_val_list = []
for cate in category_list:
cate_val_list.append((cate[0],cate[2],cate[1],Config.ju_home_today,Config.JHS_GroupItem))
# 清空category redis队列
self.cat_queue.clearQ()
# 保存category redis队列
self.cat_queue.putlistQ(cate_val_list)
# 清空act redis队列
self.act_queue.clearQ()
print '# category queue end:',time.strftime('%Y-%m-%d %H:%M:%S', time.localtime(time.time()))
else:
print '# not find category...'
# 类目的活动Json
obj = 'cat'
crawl_type = 'main'
# 获取还没有开团的活动id
val = (Common.time_s(Common.now()),)
acts = self.mysqlAccess.selectJhsActNotStart(val)
brandact_id_list = []
if acts:
for act in acts:
brandact_id_list.append(str(act[1]))
_val = (self.begin_time, brandact_id_list)
self.work.process(obj,crawl_type,_val)
# 活动数据
act_val_list = self.work.items
print '# act nums:', len(act_val_list)
# 保存到redis队列
self.act_queue.putlistQ(act_val_list)
print '# act queue end:',time.strftime('%Y-%m-%d %H:%M:%S', time.localtime(time.time()))
if self.m_type == 'm':
val = (Common.add_hours(self.begin_time, -2),Common.add_hours(self.begin_time, -2),Common.add_hours(self.begin_time, -1))
# 删除Redis中上个小时结束的活动
_acts = self.mysqlAccess.selectJhsActEndLastOneHour(val)
print '# end acts num:',len(_acts)
self.work.delAct(_acts)
# 删除Redis中上个小时结束的商品
_items = self.mysqlAccess.selectJhsItemEndLastOneHour(val)
print '# end items num:',len(_items)
self.work.delItem(_items)
except Exception as e:
print '# antpage error :',e
Common.traceback_log()
# 商品团频道
def categoryListTEMP(self):
page = self.crawler.getData(Config.ju_home_today, Config.ju_home)
if not page or page == '': print '# not get today page'
category_list = []
m = re.search(r'<div class="J_CatLeft layout-left">.+?<table>(.+?)</table>.+?</div>',page,flags=re.S)
if m:
category_list = self.categoryListType1(m.group(1))
else:
m = re.search(r'<div class="catbg">\s+<div class="ju-wrapper">\s+<div class="cat-menu-h".+?>.+?<ul class="clearfix">(.+?)</ul>',page,flags=re.S)
if m:
category_list = self.categoryListType2(m.group(1))
return category_list
def categoryListType1(self,page):
category_list = []
m = re.search(r'<tr class="h2">.+?</tr>(.+?)<tr class="h2">',page,flags=re.S)
if m:
cate_list = m.group(1)
p = re.compile(r'<a.+?href="(.+?)".+?>(.+?)</a>',flags=re.S)
for cate in p.finditer(cate_list):
category_list.append((cate.group(1),cate.group(2).strip()))
return category_list
def categoryListType2(self,page):
category_list = []
p = re.compile(r'<a.+?href="(.+?)".+?>(.+?)</a>',flags=re.S)
for cate in p.finditer(page):
category_list.append((cate.group(1),cate.group(2).strip()))
return category_list
class JHSActPosition():
'''A class of brand position'''
def __init__(self):
# mysql
self.mysqlAccess = MysqlAccess()
# 抓取设置
#self.crawler = TBCrawler()
self.crawler = RetryCrawler()
# 页面模板解析
self.brand_temp = JHSBrandTEMP()
# 获取Json数据
self.jsonpage = Jsonpage()
# 首页的品牌团列表
self.home_brands = {}
# 品牌团页面的最上面推广位
self.top_brands = {}
# 页面信息
self.ju_home_page = '' # 聚划算首页
self.ju_brand_page = '' # 聚划算品牌团页面
# 抓取开始时间
self.begin_time = Common.now()
def antPage(self):
try:
# 获取首页的品牌团
page = self.crawler.getData(Config.ju_home, Config.tmall_home)
hb = JHSHomeBrand()
hb.antPage(page)
if hb.home_brands == {} or not hb.home_brands:
page = self.crawler.getData(Config.ju_home_today, Config.ju_home)
hb.antPage(page)
self.home_brands = hb.home_brands
page_datepath = 'act/position/' + time.strftime("%Y/%m/%d/%H/", time.localtime(self.begin_time))
Config.writefile(page_datepath,'home.htm',page)
#print '# home activities:', self.home_brands
# 获取品牌团列表页数据
page = self.crawler.getData(Config.ju_brand_home, Config.ju_home)
self.activityList(page)
except Exception as e:
print '# exception err in antPage info:',e
Common.traceback_log()
# 品牌团列表
def activityList(self, page):
if not page or page == '': raise Common.InvalidPageException("# brand activityList: not get JHS brand home.")
self.ju_brand_page = page
# 保存html文件
page_datepath = 'act/marketing/' + time.strftime("%Y/%m/%d/%H/", time.localtime(self.begin_time))
Config.writefile(page_datepath,'brand.htm',self.ju_brand_page)
# 数据接口URL list
self.top_brands = self.brand_temp.activityTopbrandTemp(page)
b_url_valList = self.brand_temp.activityListTemp(page)
if b_url_valList != []:
# 从接口中获取的数据列表
bResult_list = []
json_valList = []
for b_url_val in b_url_valList:
b_url, f_name, f_catid = b_url_val
json_valList.append((b_url,Config.ju_brand_home,(f_catid,f_name)))
bResult_list = self.jsonpage.get_json(json_valList)
act_valList = []
if bResult_list and bResult_list != []:
a_val = (Config.JHS_Brand,'',self.begin_time,)
act_valList = self.jsonpage.parser_brandjson(bResult_list,a_val)
if act_valList != []:
print '# get brand act num:',len(act_valList)
self.run_brandAct(act_valList)
else:
print '# err: not get brandjson parser val list.'
else:
print '# err: not find activity json data URL list.'
def run_brandAct(self, act_valList):
repeatact_num = 0
# 活动数量
act_num = 0
# 需要保存活动sql列表
act_sql_list = []
# 用于活动去重id dict
brandact_id_dict = {}
print '# brand activities start:',time.strftime('%Y-%m-%d %H:%M:%S', time.localtime(time.time()))
# 多线程 控制并发的线程数
if len(act_valList) > Config.act_max_th:
m_Obj = JHSActM(5, Config.act_max_th)
else:
m_Obj = JHSActM(5, len(act_valList))
m_Obj.putItems(act_valList)
m_Obj.createthread()
m_Obj.run()
item_list = m_Obj.items
for b in item_list:
act_num += 1
brandact_id,brandact_name,brandact_url,brandact_sign,val = b
if int(brandact_sign) == 3:
continue
# 去重
if brandact_id_dict.has_key(str(brandact_id)):
repeatact_num += 1
print '# repeat brand act. activity id:%s name:%s'%(brandact_id, brandact_name)
else:
brandact_id_dict[str(brandact_id)] = brandact_name
if self.home_brands.has_key(str(brandact_id)):
val = val + (self.home_brands[str(brandact_id)]["position"],self.home_brands[str(brandact_id)]["datatype"],self.home_brands[str(brandact_id)]["typename"])
elif self.home_brands.has_key(brandact_url):
val = val + (self.home_brands[brandact_url]["position"],self.home_brands[brandact_url]["datatype"],self.home_brands[brandact_url]["typename"])
else:
val = val + (None,None,None)
if self.top_brands.has_key(str(brandact_id)):
val = val + (self.top_brands[str(brandact_id)]["position"],self.top_brands[str(brandact_id)]["datatype"])
elif self.top_brands.has_key(brandact_url):
val = val + (self.top_brands[brandact_url]["position"],self.top_brands[brandact_url]["datatype"])
else:
val = val + (None,None)
act_sql_list.append(val)
print '# brand activities end:',time.strftime('%Y-%m-%d %H:%M:%S', time.localtime(time.time()))
# 品牌团活动位置信息入库
# 保存
actsql_list = []
for sql in act_sql_list:
actsql_list.append(sql)
if len(actsql_list) >= Config.act_max_arg:
self.mysqlAccess.insertJhsActPosition(actsql_list)
actsql_list = []
if len(actsql_list) > 0:
self.mysqlAccess.insertJhsActPosition(actsql_list)
print '# Find act num:', act_num
print '# Repeat brand activity num:', repeatact_num
class Channel():
'''A class of XC channel'''
def __init__(self):
# 抓取设置
self.crawler = XCCrawler()
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 = '携程-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
# 频道所属地理位置信息
self.province_id = 0 # 省,州id
self.province_name = '' # 省,州名称
# 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'<div class="cate_select">(.+?)</div>', self.channel_page, flags=re.S)
if m:
cate_select = m.group(1)
c_list = []
p = re.compile(r'<a.+?class="select">(.+?)</a>', flags=re.S)
for c in p.finditer(cate_select):
c_list.append(re.sub(r'<.+?>', '', c.group(1)).strip())
self.channel_name = '-'.join(c_list)
i_p = 1
i_page = 1
m_page = 1
page_main = ''
m = re.search(r'<div id="base_bd">.+?<div class="bg_miancolor">.+?<div class="vacation_bd">(.+?)<div class="vacation_bd bottom_seo">', 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'<span class="c_page2_numtop">(.+?)</span>', self.channel_page, flags=re.S)
if m:
m_page_info = m.group(1)
m = re.search(r'\d+/(\d+)', m_page_info, flags=re.S)
if m:
m_page = int(m.group(1))
page_url = self.channel_url[0:-1] + 'P%s/'
while i_page < m_page:
i_page += 1
p_url = page_url % str(i_page)
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="searchresult_product04">\s+<div class="search_ticket_caption basefix">\s+<a href="(.+?)".+?>\s+<img src="(.+?)".+?/>.+?<div class="search_ticket_title">\s+<h2>\s+<a.+?>(.+?)</a>.+?</h2>\s+<div class="adress">(.+?)</div>\s+<div class="exercise">(.*?)</div>', flags=re.S)
for info in p.finditer(page_main):
if int(self.channel_type) == 1:
i_url = Config.xc_piao_home + info.group(1)
else:
i_url = Config.xc_home + info.group(1)
i_img, i_name, i_area, i_desc = info.group(2), info.group(3).strip(), info.group(4).strip(), info.group(5).strip()
i_book = 1
i_id = 0
if i_url != '':
m = re.search(r't(\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_area, i_p, self.crawling_begintime))
self.channel_items.append(val)
i_p += 1
return i_p
def channelList(self):
self.channelPage()
if self.channel_page:
m = re.search(r'<ul class="search_cate">\s+<li class="cate_content.+?">\s+<span class="b">.+?<span class="area_box">(.+?)</span>', self.channel_page, flags=re.S)
if m:
area_infos = m.group(1)
p = re.compile(r'<a href="(.+?)".+?>(.+?)</a>', flags=re.S)
for area in p.finditer(area_infos):
channel_url, c_name = Config.xc_piao_home + area.group(1), area.group(2)
channel_id = 0
if channel_url:
m = re.search(r'D(\d+)', channel_url)
if m:
channel_id = m.group(1)
if c_name:
m = re.search(r'(.+?)\(', c_name, flags=re.S)
if m:
channel_name = m.group(1).strip()
else:
channel_name = c_name.strip()
if int(channel_id) != 0 and channel_url:
self.channel_list.append((channel_id, channel_name, channel_url, str(self.channel_type), str(self.province_id), self.province_name))
def channelPage(self):
if self.channel_url:
refers = Config.xc_home
if int(self.channel_type) == 1:
refers = Config.xc_piao_home
data = self.crawler.getData(self.channel_url, Config.xc_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, self.province_id, self.province_name = val
self.channelList()
class Channel():
'''A class of XC channel'''
def __init__(self):
# 抓取设置
self.crawler = XCCrawler()
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 = '携程-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
# 频道所属地理位置信息
self.province_id = 0 # 省,州id
self.province_name = '' # 省,州名称
# 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'<div class="cate_select">(.+?)</div>',
self.channel_page,
flags=re.S)
if m:
cate_select = m.group(1)
c_list = []
p = re.compile(r'<a.+?class="select">(.+?)</a>', flags=re.S)
for c in p.finditer(cate_select):
c_list.append(re.sub(r'<.+?>', '', c.group(1)).strip())
self.channel_name = '-'.join(c_list)
i_p = 1
i_page = 1
m_page = 1
page_main = ''
m = re.search(
r'<div id="base_bd">.+?<div class="bg_miancolor">.+?<div class="vacation_bd">(.+?)<div class="vacation_bd bottom_seo">',
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'<span class="c_page2_numtop">(.+?)</span>',
self.channel_page,
flags=re.S)
if m:
m_page_info = m.group(1)
m = re.search(r'\d+/(\d+)', m_page_info, flags=re.S)
if m:
m_page = int(m.group(1))
page_url = self.channel_url[0:-1] + 'P%s/'
while i_page < m_page:
i_page += 1
p_url = page_url % str(i_page)
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="searchresult_product04">\s+<div class="search_ticket_caption basefix">\s+<a href="(.+?)".+?>\s+<img src="(.+?)".+?/>.+?<div class="search_ticket_title">\s+<h2>\s+<a.+?>(.+?)</a>.+?</h2>\s+<div class="adress">(.+?)</div>\s+<div class="exercise">(.*?)</div>',
flags=re.S)
for info in p.finditer(page_main):
if int(self.channel_type) == 1:
i_url = Config.xc_piao_home + info.group(1)
else:
i_url = Config.xc_home + info.group(1)
i_img, i_name, i_area, i_desc = info.group(2), info.group(
3).strip(), info.group(4).strip(), info.group(5).strip()
i_book = 1
i_id = 0
if i_url != '':
m = re.search(r't(\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_area,
i_p, self.crawling_begintime))
self.channel_items.append(val)
i_p += 1
return i_p
def channelList(self):
self.channelPage()
if self.channel_page:
m = re.search(
r'<ul class="search_cate">\s+<li class="cate_content.+?">\s+<span class="b">.+?<span class="area_box">(.+?)</span>',
self.channel_page,
flags=re.S)
if m:
area_infos = m.group(1)
p = re.compile(r'<a href="(.+?)".+?>(.+?)</a>', flags=re.S)
for area in p.finditer(area_infos):
channel_url, c_name = Config.xc_piao_home + area.group(
1), area.group(2)
channel_id = 0
if channel_url:
m = re.search(r'D(\d+)', channel_url)
if m:
channel_id = m.group(1)
if c_name:
m = re.search(r'(.+?)\(', c_name, flags=re.S)
if m:
channel_name = m.group(1).strip()
else:
channel_name = c_name.strip()
if int(channel_id) != 0 and channel_url:
self.channel_list.append(
(channel_id, channel_name, channel_url,
str(self.channel_type), str(self.province_id),
self.province_name))
def channelPage(self):
if self.channel_url:
refers = Config.xc_home
if int(self.channel_type) == 1:
refers = Config.xc_piao_home
data = self.crawler.getData(self.channel_url, Config.xc_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, self.province_id, self.province_name = val
self.channelList()
class JHSGroupItem():
'''A class of JHS group item channel'''
def __init__(self, m_type):
# 分布式主机标志
self.m_type = m_type
# 抓取设置
self.crawler = RetryCrawler()
# cat queue
self.cat_queue = JHSQ('groupitemcat', 'main')
self.worker = JHSGroupItemWorker()
# 默认类别
#self.category_list = [("http://ju.taobao.com/jusp/nvzhuangpindao/tp.htm#J_FixedNav","女装","1000")]
self.category_list = [
("http://ju.taobao.com/jusp/nvzhuangpindao/tp.htm#J_FixedNav","女装","1000"),
("http://ju.taobao.com/jusp/nanzhuangpindao/tp.htm#J_FixedNav","男装","7000"),
("http://ju.taobao.com/jusp/xiebaopindao/tp.htm#J_FixedNav","鞋包","3000"),
("http://ju.taobao.com/jusp/neiyipindao/tp.htm#J_FixedNav","内衣","4000"),
("http://ju.taobao.com/jusp/zhubaoshipin/tp.htm#J_FixedNav","饰品","42000"),
("http://ju.taobao.com/jusp/yundongpindao/tp.htm#J_FixedNav","运动","38000"),
("http://ju.taobao.com/jusp/meizhuangpindao/tp.htm#J_FixedNav","美妆","2000"),
("http://ju.taobao.com/jusp/tongzhuangpindao/tp.htm#J_FixedNav","童装","23000"),
("http://ju.taobao.com/jusp/shipinpindao/tp.htm#J_FixedNav","零食","5000"),
("http://ju.taobao.com/jusp/muyingpindao/tp.htm#J_FixedNav","母婴","6000"),
("http://ju.taobao.com/jusp/baihuopindao/tp.htm#J_FixedNav","百货","37000"),
("http://ju.taobao.com/jusp/chepinpindao/tp.htm#J_FixedNav","汽车","36000"),
("http://ju.taobao.com/jusp/jiadianpindao/tp.htm#J_FixedNav","家电","34000"),
("http://ju.taobao.com/jusp/shumapindao/tp.htm#J_FixedNav","数码","43000"),
("http://ju.taobao.com/jusp/jiajunewpindao/tp.htm#J_FixedNav","家装","225000"),
("http://ju.taobao.com/jusp/jiajupindao/tp.htm#J_FixedNav","家纺","35000")
]
# 页面
self.site_page = None
# 抓取开始时间
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 antPage(self):
try:
# 主机器需要配置redis队列
if self.m_type == 'm':
category_list = self.worker.scanCategories()
if not category_list or len(category_list) == 0:
category_list = self.category_list
if category_list and len(category_list) > 0:
# 清空category redis队列
self.cat_queue.clearQ()
# 保存category redis队列
self.cat_queue.putlistQ(category_list)
print '# groupitem category queue end:',time.strftime('%Y-%m-%d %H:%M:%S', time.localtime(time.time()))
else:
print '# groupitem not find category...'
obj = 'groupitemcat'
crawl_type = 'main'
self.worker.process(obj, crawl_type, Config.ju_home_today)
items = self.worker.items
print '# all parser items num:',len(items)
# 查找新上商品
self.get_newitems(items)
if self.m_type == 'm':
# 删除Redis中结束商品
#self.worker.scanEndItems()
self.worker.scanEndItemsLasthour()
except Exception as e:
print '# antpage error :',e
Common.traceback_log()
# 查找新上商品,并抓取新上商品详情
def get_newitems(self, items):
result_items = []
for item in items:
item_status, item_val, o_val = item
item_juid = item_val[1]
result_items.append({"item_juId":str(item_juid),"val":o_val,"r_val":item_val})
new_item_list = self.worker.selectNewItems(result_items)
print '# new items num:',len(new_item_list)
# 抓取新上商品
itemcrawl_type = 'new'
# 附加信息
a_val = (self.begin_time,)
items = self.crawlNewItems(new_item_list,itemcrawl_type,a_val)
# 保存新上商品信息到Redis
new_items = []
for item in items:
iteminfoSql = item
item_juid = iteminfoSql[1]
new_items.append({"item_juId":item_juid,"r_val":iteminfoSql})
self.worker.putItemDB(new_items)
# 抓取新上的商品详情
def crawlNewItems(self, _new_items, itemcrawl_type, a_val):
print '# crawl Group Items start:',time.strftime('%Y-%m-%d %H:%M:%S', time.localtime(time.time()))
# 多线程 控制并发的线程数
max_th = Config.item_max_th
if len(_new_items) > max_th:
m_itemsObj = JHSGroupItemCrawlerM(itemcrawl_type, max_th, a_val)
else:
m_itemsObj = JHSGroupItemCrawlerM(itemcrawl_type, len(_new_items), a_val)
m_itemsObj.createthread()
m_itemsObj.putItems(_new_items)
m_itemsObj.run()
_items = m_itemsObj.items
print '# insert new item num:',len(_items)
print '# crawl Group Items end:',time.strftime('%Y-%m-%d %H:%M:%S', time.localtime(time.time()))
return _items
# 商品团频道
def categoryListTEMP(self):
page = self.crawler.getData(Config.ju_home_today, Config.ju_home)
if not page or page == '': print '# not get today page'
category_list = []
m = re.search(r'<div class="J_CatLeft layout-left">.+?<table>(.+?)</table>.+?</div>',page,flags=re.S)
if m:
category_list = self.categoryListType1(m.group(1))
else:
m = re.search(r'<div class="catbg">\s+<div class="ju-wrapper">\s+<div class="cat-menu-h".+?>.+?<ul class="clearfix">(.+?)</ul>',page,flags=re.S)
if m:
category_list = self.categoryListType2(m.group(1))
return category_list
def categoryListType1(self,page):
category_list = []
m = re.search(r'<tr class="h2">.+?</tr>(.+?)<tr class="h2">',page,flags=re.S)
if m:
cate_list = m.group(1)
p = re.compile(r'<a.+?href="(.+?)".+?>(.+?)</a>',flags=re.S)
for cate in p.finditer(cate_list):
category_list.append((cate.group(1),cate.group(2).strip()))
return category_list
def categoryListType2(self,page):
category_list = []
p = re.compile(r'<a.+?href="(.+?)".+?>(.+?)</a>',flags=re.S)
for cate in p.finditer(page):
category_list.append((cate.group(1),cate.group(2).strip()))
return category_list
def getAjaxurlList(self,page_val,refers):
url_list = []
page, c_name, c_id = page_val
p = re.compile(r'<.+?data-ajaxurl="(.+?)".+?>(.+?)</div>',flags=re.S)
i = 0
for a_info in p.finditer(page):
c_subNav = c_name
a_url = a_info.group(1).replace('amp;','')
info = a_info.group(2)
m = re.search(r'<span class="l-f-tbox">(.+?)</span>',info,flags=re.S)
if m:
c_subNav = m.group(1).strip()
a_val = (c_id,c_name,refers,c_subNav)
url_list.append((a_url,refers,a_val))
i += 1
return url_list
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()
