def crawl_collection(self):
if self.is_tmall:
c_url = "http://count.tbcdn.cn/counter3?callback=jsonp126&keys=ICCP_1_" + self.num_id
collectionData = self.crawl_page(c_url)
if collectionData:
self.collection = int(
get_num_val(collectionData, "ICCP_1_" + self.num_id))
else:
logger.warn("Can not parse tmall item collection %s",
self.item_id)
else:
counterApi = get_val(self.data, "counterApi")
if counterApi:
counterApi = get_val(self.data,
"counterApi").replace(r'''\/''', "/")
counterData = self.crawl_page(
counterApi +
"&callback=DT.mods.SKU.CountCenter.saveCounts")
try:
self.collection = int(
get_num_val(counterData, 'ICCP_1_' + self.num_id))
self.browse = int(
get_num_val(counterData, 'ICVT_7_' + self.num_id))
except:
self.collection = 0
self.browse = 0
def crawl_volume(self):
if self.is_tmall:
apiItemInfoUrl = get_val(self.data, "initApi").replace(r'''\/''', "/")
self.tmallInitApi = self.crawl_page(apiItemInfoUrl)
try:
self.tmallInitApijson = loads(self.tmallInitApi.decode('gb18030').encode('utf8'))
except:
logger.info("parse tmall api json failed %s : %s", self.item_id, traceback.format_exc())
if self.tmallInitApijson:
try:
self.volume = self.tmallInitApijson['defaultModel']['sellCountDO']['sellCount']
except:
logger.warn("try to get volume from api failed %s", self.item_id)
if self.volume < 0:
try:
self.volume = int(get_val(self.tmallInitApi, "sellCount"))
except:
self.volume = 0
logger.warn("Can not parse tmall item volume %s", self.item_id)
else:
apiItemInfoVal = get_val(self.data, "apiItemInfo")
if apiItemInfoVal:
apiItemInfoUrl = get_val(self.data, "apiItemInfo").replace(r'''\/''', "/")
itemInfoData = self.crawl_page(apiItemInfoUrl)
try:
self.volume = int(get_num_val(itemInfoData, 'quanity'))
self.confirmVolume = int(get_num_val(itemInfoData, 'confirmGoods'))
except:
self.volume = 0
logger.warn("Can not parse taobao item volume %s", self.item_id)
else:
self.volume = 0
def crawl_title(self):
try:
self.data = self.crawl_page(self.url)
if not self.data:
logger.warn("download %s %s page failed, possible network connection failure", self.item_id, self.num_id)
return
# check tmall
if not self.is_tmall and len(self.data) < 256 and self.url.find('item.taobao.com') > 0 and self.data.find("window.location.href='http://detail.tmall.com/item.htm'+window.location.search") > 0:
self.data = self.crawl_page(self.url.replace('item.taobao.com', 'detail.tmall.com'))
if self.check_offline():
self.is_offline = True
self.html_obj = parse_html(self.data, encoding="gb18030")
title = self.html_obj.xpath("//html/head/title/text()")
if title and title[0].find(u"转卖") > 0:
self.is_offline = True
if title:
self.title = title[0].encode('utf8').replace("-淘宝网", "").replace("-tmall.com天猫", "")
#tblogo = self.html_obj.xpath("//*[@id='shop-logo']")
tmalllogo = self.html_obj.xpath("//*[@id='mallLogo']")
if not tmalllogo:
tmalllogo = self.html_obj.xpath("//*[@id='simple-logo']")
if not self.is_tmall and tmalllogo:
self.is_tmall = True
self.thumbImages = self.html_obj.xpath("//ul[@id='J_UlThumb']//img/@src")
if not len(self.thumbImages):
try:
# try load thumb images for tmall page
self.thumbImages = [IMAGESTYLE_RE.subn(r'\g<1>', x)[0] for x in self.html_obj.xpath("//ul[@id='J_UlThumb']//li/@style")]
# taobao @src to @data-src
if not len(self.thumbImages):
self.thumbImages = self.html_obj.xpath("//ul[@id='J_UlThumb']//img/@data-src")
except:
logger.warn("No thumbs found %s", self.item_id)
if self.is_tmall:
self.cid = get_val(self.data, "categoryId").split('&')[0]
else:
self.cid = get_val(self.data, "cid")
logger.info("Got %s %s html success", self.item_id, self.num_id)
except:
logger.error("crawling %s %s unknown exception %s", self.item_id, self.num_id, traceback.format_exc(), extra={'tags':['crawlItemException',]})
raise
def crawl_desc(self):
try:
self.descUrl = get_val(self.data, "apiItemDesc").replace(r'''\/''', "/")
except:
try:
self.descUrl = get_val(self.data, "ItemDesc").replace(r'''\/''', "/")
except:
if not self.data.find(u"暂无描述".encode('gb18030')) > 0:
self.descUrl = DESCURL_RE.search(self.data).group(0)
# find http://dsc.taobaocdn.com/[^.]+\.desc.* as descUrl
if self.descUrl:
self.descContent = self.crawl_page(self.descUrl)
logger.debug("Got %s desc details %s", self.item_id, len(self.descContent))
else:
logger.warn("%s desc url not found", self.item_id)
def crawl_taobao_rate(self, from_rate_id):
rateListUrlBase = get_val(self.data, "data-listApi")
if rateListUrlBase:
rateListUrlBase = rateListUrlBase.replace(r'''\/''', "/")
maxPage = 1
page = 1
total = 0
while page <= maxPage:
if page >= FLAGS.mostPage:
break
page1Result = self.crawl_taobao_rate_page(rateListUrlBase, page)
if not page1Result:
return
results = page1Result['comments']
maxPage = page1Result['maxPage']
if not results:
break
self.taobao_comments_to_pb(results)
page += 1
total += len(results)
if from_rate_id:
if self.cut_comments(from_rate_id):
break
if self.max_comments > 0 and self.max_comments < total:
break
logger.debug("Got %s %s comments", self.item_id, len(self.comments))
def crawl_taobao_rate(self, from_rate_id):
rateListUrlBase = get_val(self.data, "data-listApi")
if rateListUrlBase:
rateListUrlBase = rateListUrlBase.replace(r'''\/''', "/")
maxPage = 1
page = 1
total = 0
while page <= maxPage:
if page >= FLAGS.mostPage:
break
page1Result = self.crawl_taobao_rate_page(rateListUrlBase, page)
if not page1Result:
return
results = page1Result['comments']
maxPage = page1Result['maxPage']
if not results:
break
self.taobao_comments_to_pb(results, page)
page += 1
total += len(results)
if from_rate_id:
if self.cut_comments(from_rate_id):
break
if self.max_comments > 0 and self.max_comments < total:
break
logger.debug("Got %s %s comments", self.item_id, len(self.comments))
def crawler(sql):
db = get_db_engine()
items = list(db.execute(sql))
# debug
if FLAGS.debug_parser:
import pdb
pdb.set_trace()
for item in items:
shop_id = item[0]
shop_type = item[1]
item_id = item[2]
url = item[3]
try:
htm = get_item_htm(item_id, url, db)
if shop_type == 1:
htm_obj = parse_html(htm, encoding='gb18030')
discount_url = htm_obj.xpath("//div[@id='promote']/@data-default")
if discount_url and len(discount_url) > 0:
item_headers = {'Referer': url, 'User-Agent': DEFAULT_UA}
disc_content = download(discount_url[0], item_headers)
if disc_content.strip():
disc_obj = parse_html(disc_content, encoding='gb18030')
content = disc_obj.xpath("//div[@id='J_MjsData']/h3/text()")[0].strip()
dates = disc_obj.xpath("//div[@id='J_MjsData']/h3/span[@class='tb-indate']/text()")[0].strip()
st = dates.encode('utf-8').replace("--","—").split("—")
start_time = datetime.datetime.strptime(st[0].strip().replace('年','-').replace("月","-").replace("日",""),'%Y-%m-%d')
end_time = datetime.datetime.strptime(st[1].strip().replace('年','-').replace("月","-").replace("日",""),'%Y-%m-%d')
db.execute("replace into shop_discount (shop_id,content,start_time,end_time,discount_url,create_time,last_update_time) values (%s,%s,%s,%s,%s,now(),now())",
shop_id, content.encode('utf-8'), start_time, end_time, discount_url[0])
logger.info("taobao shop %s get discount success", shop_id)
else:
logger.warning("taobao shop %s:%s not discount.", shop_id, url)
else:
logger.warning("taobao shop %s:%s not discount.", shop_id, url)
elif shop_type == 2:
d_url = get_val(htm, "initApi")
if d_url:
item_headers = {'Referer': url, 'User-Agent': DEFAULT_UA}
disc_content = download(d_url, item_headers)
cjson = loads(disc_content.decode('gb18030').encode('utf8'))
shop_prom = cjson['defaultModel']['itemPriceResultDO']['tmallShopProm']
if shop_prom:
st = int(shop_prom['startTime'])/1000
et = int(shop_prom['endTime'])/1000
start_time = time.strftime("%Y-%m-%d", time.localtime(st))
end_time = time.strftime("%Y-%m-%d", time.localtime(et))
content = shop_prom['promPlan'][0]['msg']
db.execute("replace into shop_discount (shop_id,content,start_time,end_time,discount_url,create_time,last_update_time) values (%s,%s,%s,%s,%s,now(),now())",
shop_id, content.encode('utf-8'), start_time, end_time, d_url)
logger.info("tmall shop %s get discount success", shop_id)
else:
logger.warning("taobao shop %s:%s not discount.", shop_id, url)
except:
logger.error("shop %s:%s xpath failed:%s", shop_id, url, traceback.format_exc())
def crawl_desc(self):
try:
self.descUrl = get_val(self.data,
"apiItemDesc").replace(r'''\/''', "/")
except:
try:
self.descUrl = get_val(self.data,
"ItemDesc").replace(r'''\/''', "/")
except:
if not self.data.find(u"暂无描述".encode('gb18030')) > 0:
self.descUrl = DESCURL_RE.search(self.data).group(0)
# find http://dsc.taobaocdn.com/[^.]+\.desc.* as descUrl
if self.descUrl:
self.descContent = self.crawl_page(self.descUrl)
logger.info("Got %s %s desc details %s", self.item_id, self.num_id,
len(self.descContent))
else:
logger.warn("%s desc url not found", self.item_id)
def crawl_collection(self):
if self.is_tmall:
c_url = "http://count.tbcdn.cn/counter3?callback=jsonp126&keys=ICCP_1_" + self.num_id
collectionData = self.crawl_page(c_url)
if collectionData:
self.collection = int(get_num_val(collectionData, "ICCP_1_" + self.num_id))
else:
logger.warn("Can not parse tmall item collection %s", self.item_id)
else:
counterApi = get_val(self.data, "counterApi")
if counterApi:
counterApi = get_val(self.data, "counterApi").replace(r'''\/''', "/")
counterData = self.crawl_page(counterApi + "&callback=DT.mods.SKU.CountCenter.saveCounts")
try:
self.collection = int(get_num_val(counterData, 'ICCP_1_' + self.num_id))
self.browse = int(get_num_val(counterData, 'ICVT_7_' + self.num_id))
except:
self.collection = 0
self.browse = 0
def crawl_volume(self):
if self.is_tmall:
apiItemInfoUrl = get_val(self.data,
"initApi").replace(r'''\/''', "/")
self.tmallInitApi = self.crawl_page(apiItemInfoUrl)
try:
self.tmallInitApijson = loads(
self.tmallInitApi.decode('gb18030').encode('utf8'))
except:
logger.info("parse tmall api json failed %s : %s",
self.item_id, traceback.format_exc())
if self.tmallInitApijson:
try:
self.volume = self.tmallInitApijson['defaultModel'][
'sellCountDO']['sellCount']
except:
logger.warn("try to get volume from api failed %s",
self.item_id)
if self.volume < 0:
try:
self.volume = int(get_val(self.tmallInitApi, "sellCount"))
except:
self.volume = 0
logger.warn("Can not parse tmall item volume %s",
self.item_id)
else:
apiItemInfoVal = get_val(self.data, "apiItemInfo")
if apiItemInfoVal:
apiItemInfoUrl = get_val(self.data, "apiItemInfo").replace(
r'''\/''', "/")
itemInfoData = self.crawl_page(apiItemInfoUrl)
try:
self.volume = int(get_num_val(itemInfoData, 'quanity'))
self.confirmVolume = int(
get_num_val(itemInfoData, 'confirmGoods'))
except:
self.volume = 0
logger.warn("Can not parse taobao item volume %s",
self.item_id)
else:
self.volume = 0
def crawl_stock(self):
if self.is_tmall:
try:
self.stock = int(get_num_val(self.tmallInitApi, "icTotalQuantity"))
except:
logger.warn("Can not parse tmall item stock %s", self.item_id)
else:
try:
if self.dynamicStockData:
self.stock = int(get_val(self.dynamicStockData, "stock").strip())
else:
logger.warn("Can not parse taobao item stock %s", self.item_id)
except:
logger.error("Can not parse tmall item stock %s", self.item_id)
def crawl_stock(self):
if self.is_tmall:
try:
self.stock = int(
get_num_val(self.tmallInitApi, "icTotalQuantity"))
except:
logger.warn("Can not parse tmall item stock %s", self.item_id)
else:
try:
if self.dynamicStockData:
self.stock = int(
get_val(self.dynamicStockData, "stock").strip())
else:
logger.warn("Can not parse taobao item stock %s",
self.item_id)
except:
logger.error("Can not parse tmall item stock %s", self.item_id)
def crawl_price(self):
self.bidPrice = self.html_obj.xpath(
"//input[@name='current_price']/@value")
self.originPrice = self.html_obj.xpath(
"//strong[@id='J_StrPrice']/em[@class='tb-rmb-num']/text()")
if not self.originPrice:
self.originPrice = self.html_obj.xpath(
"//strong[@class='J_originalPrice']/text()")
self.promoteUrl2 = get_val(self.data, "apiPromoData")
if self.promoteUrl2:
self.promoteUrl2 = self.promoteUrl2.replace(r'''\/''', "/")
price = ""
if self.is_tmall and self.tmallInitApi and self.tmallInitApijson:
try:
priceInfo = self.tmallInitApijson['defaultModel'][
'itemPriceResultDO']['priceInfo']
if priceInfo:
if priceInfo.has_key('def'):
defaultPriceInfo = priceInfo['def']
else:
defaultPriceInfo = priceInfo[priceInfo.keys()[0]]
# 2013-11-22 改为获取真实促销价格，而不是扣除佣金后的价格
if defaultPriceInfo.has_key(
'promotionList'
) and defaultPriceInfo['promotionList']:
price = defaultPriceInfo['promotionList'][0]['price']
if not price:
if defaultPriceInfo.has_key('price'):
price = defaultPriceInfo['price']
if not price:
if defaultPriceInfo.has_key('promPrice'):
price = defaultPriceInfo['promPrice']['price']
elif defaultPriceInfo.has_key(
'promotionList'
) and defaultPriceInfo['promotionList']:
price = str(
min([
float(x.get('price', '100000000.0'))
for x in defaultPriceInfo['promotionList']
]))
except:
logger.warn("Parse tmall json price failed, %s", self.item_id)
if not price:
if self.promoteUrl2:
self.promoteContent = self.crawl_page(
self.promoteUrl2).replace('"', '"')
tag = "low:"
if self.promoteContent.find(tag) > 0:
pos = self.promoteContent.find(tag) + len(tag)
pos2 = self.promoteContent.find(',', pos)
price = self.promoteContent[pos:pos2]
if not price:
price = get_num_val(self.promoteContent, 'price')
else:
self.promoteUrl = "http://marketing.taobao.com/home/promotion/item_promotion_list.do?itemId=%s" % self.num_id
self.promoteContent = self.crawl_page(self.promoteUrl)
if self.promoteContent:
self.promoteContent = self.promoteContent.replace(
'"', '"')
tag = "promPrice":""
if self.promoteContent.find(tag) > 0:
pos = self.promoteContent.find(tag) + len(tag)
pos2 = self.promoteContent.find('"', pos)
price = self.promoteContent[pos:pos2]
if not price:
tbPrice = self.html_obj.xpath("//strong[@class='tb-price']/text()")
tbPrice1 = self.html_obj.xpath("//span[@class='tb-price']/text()")
if tbPrice and not tbPrice[0].strip():
price = tbPrice[0].strip()
elif tbPrice1 and not tbPrice1[0].strip():
price = tbPrice1[0].strip()
if price.find("-") > 0:
price = price.split('-')[0].strip()
if not price:
rg_m = re.compile('price:\"[0-9]+[.][0-9]+\"', re.IGNORECASE
| re.DOTALL).search(self.dynamicStockData)
if rg_m:
price_str = rg_m.group(0).split(":")[1].replace("\"", "")
price = Decimal(price_str)
# 2013-09-03 get price url
if not price:
#这里稍微有点麻烦,主要针对string进行处理
pirce_url = "http://ajax.tbcdn.cn/json/umpStock.htm?itemId=%s&p=1" % self.num_id
response = download(pirce_url, self.headers)
rg = re.compile('price:\"[0-9]+[.][0-9]+\"',
re.IGNORECASE | re.DOTALL)
m = rg.search(response.decode('gb18030').encode('utf8'))
if m:
price_str = m.group(0).split(":")[1].replace("\"", "")
price = Decimal(price_str)
# not chuxiao price, set origin price
if not price:
if self.originPrice:
price = self.originPrice[0].strip()
elif self.bidPrice:
price = self.bidPrice[0].strip()
if price.find("-") > 0:
price = price.split('-')[0].strip()
self.price = float(price)
logger.debug("%s price is %s", self.item_id, self.price)
def crawl(self):
try:
self.data = self.crawl_page(self.url)
if FLAGS.debug_parser:
import pdb; pdb.set_trace()
# check tmall
if not self.is_tmall and len(self.data) < 256 and self.url.find('item.taobao.com') > 0 and self.data.find("window.location.href='http://detail.tmall.com/item.htm'+window.location.search") > 0:
self.data = self.crawl_page(self.url.replace('item.taobao.com', 'detail.tmall.com'))
if self.check_offline():
self.is_offline = True
self.html_obj = parse_html(self.data, encoding="gb18030")
title = self.html_obj.xpath("//html/head/title/text()")
if title and title[0].find(u"转卖") > 0:
self.is_offline = True
self.detailDiv = self.html_obj.xpath("//div[@id='detail']")
self.buyButton = self.html_obj.xpath("//a[@id='J_LinkBuy']")
self.originPrice = self.html_obj.xpath("//strong[@id='J_StrPrice']/em[@class='tb-rmb-num']/text()")
if not self.originPrice:
self.originPrice = self.html_obj.xpath("//strong[@class='J_originalPrice']/text()")
#self.bidPrice = self.html_obj.xpath("//li[contains(concat(' ',normalize-space(@class),' '),' detail-price ')]/strong/text()")
self.bidPrice = self.html_obj.xpath("//input[@name='current_price']/@value")
self.thumbImages = self.html_obj.xpath("//ul[@id='J_UlThumb']//img/@src")
if not len(self.thumbImages):
try:
# try load thumb images for tmall page
self.thumbImages = [IMAGESTYLE_RE.subn(r'\g<1>', x)[0] for x in self.html_obj.xpath("//ul[@id='J_UlThumb']//li/@style")]
# taobao @src to @data-src
if not len(self.thumbImages):
self.thumbImages = self.html_obj.xpath("//ul[@id='J_UlThumb']//img/@data-src")
except:
logger.warn("No thumbs found %s", self.item_id)
tblogo = self.html_obj.xpath("//*[@id='shop-logo']")
tmalllogo = self.html_obj.xpath("//*[@id='mallLogo']")
if not self.is_tmall and tmalllogo:
self.is_tmall = True
if self.is_tmall:
self.cid = get_val(self.data, "categoryId").split('&')[0]
apiItemInfoUrl = get_val(self.data, "initApi").replace(r'''\/''', "/")
self.tmallInitApi = self.crawl_page(apiItemInfoUrl)
try:
self.tmallInitApijson = loads(self.tmallInitApi.decode('gb18030').encode('utf8'))
except:
logger.info("parse tmall api json failed %s : %s", self.item_id, traceback.format_exc())
if self.tmallInitApijson:
try:
self.volume = self.tmallInitApijson['defaultModel']['sellCountDO']['sellCount']
except:
logger.warn("try to get volume from api failed %s", self.item_id)
if self.volume < 0:
try:
self.volume = int(get_val(self.tmallInitApi, "sellCount"))
except:
logger.warn("Can not parse item volume %s", self.item_id)
# 库存 ：icTotalQuantity
""""
reviewInfoUrl = get_val(self.data, "apiMallReviews").replace(r'''\/''', "/")
reviewInfoData = self.crawl_page(reviewInfoUrl)
m = RATECOUNT_RE.match(reviewInfoData)
if m:
self.reviewCount = m.group(1)
else:
self.reviewCount = None
"""
else:
self.cid = get_val(self.data, "cid")
apiItemInfoVal = get_val(self.data, "apiItemInfo")
if apiItemInfoVal:
apiItemInfoUrl = get_val(self.data, "apiItemInfo").replace(r'''\/''', "/")
itemInfoData = self.crawl_page(apiItemInfoUrl)
try:
self.volume = int(get_num_val(itemInfoData, 'quanity'))
except:
self.volume = -1
else:
self.volume = -1
#interval = get_val(data2, 'interval')
# 库存 skudata = get_val(self.data, 'valItemInfo').replace(r'''\/''', "/")
"""
reviewInfoUrl = get_val(self.data, "data-commonApi").replace(r'''\/''', "/")
reviewInfoData = self.crawl_page(reviewInfoUrl)
self.reviewCount = get_val(reviewInfoData, 'total')
"""
except:
logger.error("crawling %s unknown exception %s", self.item_id, traceback.format_exc(), extra={'tags':['crawlItemException',]})
raise
def crawl_price(self):
self.promoteUrl2 = get_val(self.data, "apiPromoData")
if self.promoteUrl2:
self.promoteUrl2 = self.promoteUrl2.replace(r'''\/''', "/")
price = ""
if self.is_tmall and self.tmallInitApi and self.tmallInitApijson:
try:
priceInfo = self.tmallInitApijson['defaultModel']['itemPriceResultDO']['priceInfo']
if priceInfo:
if priceInfo.has_key('def'):
defaultPriceInfo = priceInfo['def']
else:
defaultPriceInfo = priceInfo[priceInfo.keys()[0]]
if defaultPriceInfo.has_key('promPrice'):
price = defaultPriceInfo['promPrice']['price']
elif defaultPriceInfo.has_key('promotionList') and defaultPriceInfo['promotionList']:
price = str(min([float(x.get('price','100000000.0')) for x in defaultPriceInfo['promotionList']]))
else:
price = defaultPriceInfo['price']
except:
logger.warn("Parse tmall json price failed, %s", self.item_id)
if not price:
if self.promoteUrl2:
self.promoteContent = self.crawl_page(self.promoteUrl2).replace('"', '"')
tag = "low:"
if self.promoteContent.find(tag) > 0:
pos = self.promoteContent.find(tag) + len(tag)
pos2 = self.promoteContent.find(',', pos)
price = self.promoteContent[pos:pos2]
if not price:
price = get_num_val(self.promoteContent, 'price')
else:
self.promoteUrl = "http://marketing.taobao.com/home/promotion/item_promotion_list.do?itemId=%s" % self.num_id
self.promoteContent = self.crawl_page(self.promoteUrl).replace('"', '"')
tag = "promPrice":""
if self.promoteContent.find(tag) > 0:
pos = self.promoteContent.find(tag) + len(tag)
pos2 = self.promoteContent.find('"', pos)
price = self.promoteContent[pos:pos2]
if not price:
tbPrice = self.html_obj.xpath("//strong[@class='tb-price']/text()")
tbPrice1 = self.html_obj.xpath("//span[@class='tb-price']/text()")
if tbPrice and not tbPrice[0].strip():
price = tbPrice[0].strip()
elif tbPrice1 and not tbPrice1[0].strip():
price = tbPrice1[0].strip()
if price.find("-") > 0:
price = price.split('-')[0].strip()
# 2013-09-03 get price url
if not price:
#这里稍微有点麻烦,主要针对string进行处理
pirce_url = "http://ajax.tbcdn.cn/json/umpStock.htm?itemId=%s&p=1" % self.num_id
response = download(pirce_url, self.headers)
rg = re.compile('price:\"[0-9]+[.][0-9]+\"', re.IGNORECASE|re.DOTALL)
m = rg.search(response.decode('gb18030').encode('utf8'))
if m:
price_str = m.group(0).split(":")[1].replace("\"", "")
price = Decimal(price_str)
# not chuxiao price, set origin price
if not price:
if self.originPrice:
price = self.originPrice[0].strip()
elif self.bidPrice:
price = self.bidPrice[0].strip()
if price.find("-") > 0:
price = price.split('-')[0].strip()
self.price = float(price)
logger.debug("%s price is %s", self.item_id, self.price)
def crawl_title(self):
try:
self.data = self.crawl_page(self.url)
if not self.data:
logger.warn(
"download %s %s page failed, possible network connection failure",
self.item_id, self.num_id)
return
# check tmall
if not self.is_tmall and len(self.data) < 256 and self.url.find(
'item.taobao.com'
) > 0 and self.data.find(
"window.location.href='http://detail.tmall.com/item.htm'+window.location.search"
) > 0:
self.data = self.crawl_page(
self.url.replace('item.taobao.com', 'detail.tmall.com'))
if self.check_offline():
self.is_offline = True
self.html_obj = parse_html(self.data, encoding="gb18030")
title = self.html_obj.xpath("//html/head/title/text()")
if title and title[0].find(u"转卖") > 0:
self.is_offline = True
if title:
self.title = title[0].encode('utf8').replace(
"-淘宝网", "").replace("-tmall.com天猫", "")
#tblogo = self.html_obj.xpath("//*[@id='shop-logo']")
tmalllogo = self.html_obj.xpath("//*[@id='mallLogo']")
if not tmalllogo:
tmalllogo = self.html_obj.xpath("//*[@id='simple-logo']")
if not self.is_tmall and tmalllogo:
self.is_tmall = True
self.thumbImages = self.html_obj.xpath(
"//ul[@id='J_UlThumb']//img/@src")
if not len(self.thumbImages):
try:
# try load thumb images for tmall page
self.thumbImages = [
IMAGESTYLE_RE.subn(r'\g<1>', x)[0] for x in
self.html_obj.xpath("//ul[@id='J_UlThumb']//li/@style")
]
# taobao @src to @data-src
if not len(self.thumbImages):
self.thumbImages = self.html_obj.xpath(
"//ul[@id='J_UlThumb']//img/@data-src")
except:
logger.warn("No thumbs found %s", self.item_id)
if self.is_tmall:
self.cid = get_val(self.data, "categoryId").split('&')[0]
else:
self.cid = get_val(self.data, "cid")
logger.info("Got %s %s html success", self.item_id, self.num_id)
except:
logger.error("crawling %s %s unknown exception %s",
self.item_id,
self.num_id,
traceback.format_exc(),
extra={'tags': [
'crawlItemException',
]})
raise
def crawl(self):
try:
self.data = self.crawl_page(self.url)
if FLAGS.debug_parser:
import pdb
pdb.set_trace()
# check tmall
if not self.is_tmall and len(self.data) < 256 and self.url.find(
'item.taobao.com'
) > 0 and self.data.find(
"window.location.href='http://detail.tmall.com/item.htm'+window.location.search"
) > 0:
self.data = self.crawl_page(
self.url.replace('item.taobao.com', 'detail.tmall.com'))
if self.check_offline():
self.is_offline = True
self.html_obj = parse_html(self.data, encoding="gb18030")
title = self.html_obj.xpath("//html/head/title/text()")
if title and title[0].find(u"转卖") > 0:
self.is_offline = True
self.detailDiv = self.html_obj.xpath("//div[@id='detail']")
self.buyButton = self.html_obj.xpath("//a[@id='J_LinkBuy']")
self.originPrice = self.html_obj.xpath(
"//strong[@id='J_StrPrice']/em[@class='tb-rmb-num']/text()")
if not self.originPrice:
self.originPrice = self.html_obj.xpath(
"//strong[@class='J_originalPrice']/text()")
#self.bidPrice = self.html_obj.xpath("//li[contains(concat(' ',normalize-space(@class),' '),' detail-price ')]/strong/text()")
self.bidPrice = self.html_obj.xpath(
"//input[@name='current_price']/@value")
self.thumbImages = self.html_obj.xpath(
"//ul[@id='J_UlThumb']//img/@src")
if not len(self.thumbImages):
try:
# try load thumb images for tmall page
self.thumbImages = [
IMAGESTYLE_RE.subn(r'\g<1>', x)[0] for x in
self.html_obj.xpath("//ul[@id='J_UlThumb']//li/@style")
]
# taobao @src to @data-src
if not len(self.thumbImages):
self.thumbImages = self.html_obj.xpath(
"//ul[@id='J_UlThumb']//img/@data-src")
except:
logger.warn("No thumbs found %s", self.item_id)
tblogo = self.html_obj.xpath("//*[@id='shop-logo']")
tmalllogo = self.html_obj.xpath("//*[@id='mallLogo']")
if not self.is_tmall and tmalllogo:
self.is_tmall = True
if self.is_tmall:
self.cid = get_val(self.data, "categoryId").split('&')[0]
apiItemInfoUrl = get_val(self.data,
"initApi").replace(r'''\/''', "/")
self.tmallInitApi = self.crawl_page(apiItemInfoUrl)
try:
self.tmallInitApijson = loads(
self.tmallInitApi.decode('gb18030').encode('utf8'))
except:
logger.info("parse tmall api json failed %s : %s",
self.item_id, traceback.format_exc())
if self.tmallInitApijson:
try:
self.volume = self.tmallInitApijson['defaultModel'][
'sellCountDO']['sellCount']
except:
logger.warn("try to get volume from api failed %s",
self.item_id)
if self.volume < 0:
try:
self.volume = int(
get_val(self.tmallInitApi, "sellCount"))
except:
logger.warn("Can not parse item volume %s",
self.item_id)
# 库存 ：icTotalQuantity
""""
reviewInfoUrl = get_val(self.data, "apiMallReviews").replace(r'''\/''', "/")
reviewInfoData = self.crawl_page(reviewInfoUrl)
m = RATECOUNT_RE.match(reviewInfoData)
if m:
self.reviewCount = m.group(1)
else:
self.reviewCount = None
"""
else:
self.cid = get_val(self.data, "cid")
apiItemInfoVal = get_val(self.data, "apiItemInfo")
if apiItemInfoVal:
apiItemInfoUrl = get_val(self.data, "apiItemInfo").replace(
r'''\/''', "/")
itemInfoData = self.crawl_page(apiItemInfoUrl)
try:
self.volume = int(get_num_val(itemInfoData, 'quanity'))
except:
self.volume = -1
else:
self.volume = -1
#interval = get_val(data2, 'interval')
# 库存 skudata = get_val(self.data, 'valItemInfo').replace(r'''\/''', "/")
"""
reviewInfoUrl = get_val(self.data, "data-commonApi").replace(r'''\/''', "/")
reviewInfoData = self.crawl_page(reviewInfoUrl)
self.reviewCount = get_val(reviewInfoData, 'total')
"""
except:
logger.error("crawling %s unknown exception %s",
self.item_id,
traceback.format_exc(),
extra={'tags': [
'crawlItemException',
]})
raise
def crawler(sql):
db = get_db_engine()
shops = list(db.execute(sql))
# debug
if FLAGS.debug_parser:
import pdb
pdb.set_trace()
for shop in shops:
shop_id = shop[0]
url = str(shop[1])
type = shop[2]
if url[-1] != '/':
url += "/"
try:
shop_headers = {'Referer': url, 'User-Agent': DEFAULT_UA}
dongtai_url = url + "dongtai.htm"
dongtai_data = download(dongtai_url, shop_headers)
if dongtai_data:
dongtai_obj = parse_html(dongtai_data, encoding="gb18030")
dongtai_title = dongtai_obj.xpath("//title/text()")[0].encode('utf-8')
if '店铺动态' in dongtai_title:
microscope_data = dongtai_obj.xpath("//*[@name='microscope-data']/@content")
userId = get_val(str(microscope_data), "userId")
if userId:
dongtai_headers = {'Referer': dongtai_url, 'User-Agent': DEFAULT_UA}
promotion_url = "http://shuo.taobao.com/feed/v_front_feeds.htm?_input_charset=utf-8&page=1" \
"&userId=%s&vfeedTabId=115" % userId
promotion_data = download(promotion_url, dongtai_headers)
if promotion_data:
promotion_obj = parse_html(promotion_data, encoding="gb18030")
i = 0
while i < 10:
feedInfo = promotion_obj.xpath("//div[@class='fd-item show-detail']//div[@class='fd-text']//span[@class='J_FeedInfo']/text()")[i].encode('utf-8')
if '店铺促销中' in feedInfo or '限时折扣' in feedInfo or '折扣限时' in feedInfo:
#title = promotion_obj.xpath("//div[@class='fd-item show-detail']//div[@class='fd-container']//dt//a/text()")[i]
link = promotion_obj.xpath("//div[@class='fd-item show-detail']//div[@class='fd-container']//dd//a[@class='fd-view-detail']/@href")[i]
promotion_price = promotion_obj.xpath("//div[@class='fd-item show-detail']//div[@class='fd-container']//dd//span[@class='g_price']/strong/text()")[i]
price = promotion_obj.xpath("//div[@class='fd-item show-detail']//div[@class='fd-container']//dd//span[@class='g_price g_price-original']/strong/text()")[i]
promotion_time = promotion_obj.xpath(u"//div[@class='fd-item show-detail']//div[@class='fd-container']//dd[contains(text(),'起止日期')]/text()")[i]
pt = promotion_time.encode('utf-8').replace("起止日期：","").split(" - ")
start_time = pt[0].replace(".", "-")
end_time = pt[1].replace(".", "-")
if '2013' not in pt[1] or '2014' not in pt[1]:
end_time = '2013-' + end_time
if start_time > end_time:
end_time = end_time.replace("2013", "2014")
num_id = get_numiid(link, dongtai_headers)
if num_id:
sql = "select id from shop_promotion where shop_id=%s and num_id=%s" % (shop_id, num_id)
re = list(db.execute(sql))
if not re:
db.execute("insert into shop_promotion (shop_id, num_id, price, "
"promotion_price, start_time, end_time, create_time, "
"last_update_time) values (%s,'%s',%s,%s,'%s','%s',now(),now())"
% (shop_id, num_id, price.replace(',', ''), promotion_price.replace(',', ''), start_time, end_time))
else:
logger.error("shop %s:%s crawler num_id failed", shop_id, url)
i += 1
logger.info("shop %s:%s crawler promotiom item num=%s", shop_id, url, i)
else:
logger.warning("shop %s:%s not promotion info", shop_id, url)
else:
logger.error("shop %s:%s crawler userId failed", shop_id, url)
else:
logger.error("shop %s:%s not dongtai page", shop_id, url)
except:
logger.error("shop %s:%s crawler failed %s", shop_id, url, traceback.format_exc())
def crawler(sql):
db = get_db_engine()
shops = list(db.execute(sql))
# debug
if FLAGS.debug_parser:
import pdb
pdb.set_trace()
for shop in shops:
shop_id = shop[0]
url = str(shop[1])
type = shop[2]
if url[-1] != '/':
url += "/"
try:
shop_headers = {'Referer': url, 'User-Agent': DEFAULT_UA}
dongtai_url = url + "dongtai.htm"
dongtai_data = download(dongtai_url, shop_headers)
if dongtai_data:
dongtai_obj = parse_html(dongtai_data, encoding="gb18030")
dongtai_title = dongtai_obj.xpath("//title/text()")[0].encode(
'utf-8')
if '店铺动态' in dongtai_title:
microscope_data = dongtai_obj.xpath(
"//*[@name='microscope-data']/@content")
userId = get_val(str(microscope_data), "userId")
if userId:
dongtai_headers = {
'Referer': dongtai_url,
'User-Agent': DEFAULT_UA
}
promotion_url = "http://shuo.taobao.com/feed/v_front_feeds.htm?_input_charset=utf-8&page=1" \
"&userId=%s&vfeedTabId=115" % userId
promotion_data = download(promotion_url,
dongtai_headers)
if promotion_data:
promotion_obj = parse_html(promotion_data,
encoding="gb18030")
i = 0
while i < 10:
feedInfo = promotion_obj.xpath(
"//div[@class='fd-item show-detail']//div[@class='fd-text']//span[@class='J_FeedInfo']/text()"
)[i].encode('utf-8')
if '店铺促销中' in feedInfo or '限时折扣' in feedInfo or '折扣限时' in feedInfo:
#title = promotion_obj.xpath("//div[@class='fd-item show-detail']//div[@class='fd-container']//dt//a/text()")[i]
link = promotion_obj.xpath(
"//div[@class='fd-item show-detail']//div[@class='fd-container']//dd//a[@class='fd-view-detail']/@href"
)[i]
promotion_price = promotion_obj.xpath(
"//div[@class='fd-item show-detail']//div[@class='fd-container']//dd//span[@class='g_price']/strong/text()"
)[i]
price = promotion_obj.xpath(
"//div[@class='fd-item show-detail']//div[@class='fd-container']//dd//span[@class='g_price g_price-original']/strong/text()"
)[i]
promotion_time = promotion_obj.xpath(
u"//div[@class='fd-item show-detail']//div[@class='fd-container']//dd[contains(text(),'起止日期')]/text()"
)[i]
pt = promotion_time.encode(
'utf-8').replace("起止日期：",
"").split(" - ")
start_time = pt[0].replace(".", "-")
end_time = pt[1].replace(".", "-")
if '2013' not in pt[1] or '2014' not in pt[
1]:
end_time = '2013-' + end_time
if start_time > end_time:
end_time = end_time.replace(
"2013", "2014")
num_id = get_numiid(link, dongtai_headers)
if num_id:
sql = "select id from shop_promotion where shop_id=%s and num_id=%s" % (
shop_id, num_id)
re = list(db.execute(sql))
if not re:
db.execute(
"insert into shop_promotion (shop_id, num_id, price, "
"promotion_price, start_time, end_time, create_time, "
"last_update_time) values (%s,'%s',%s,%s,'%s','%s',now(),now())"
% (shop_id, num_id,
price.replace(',', ''),
promotion_price.replace(
',', ''), start_time,
end_time))
else:
logger.error(
"shop %s:%s crawler num_id failed",
shop_id, url)
i += 1
logger.info(
"shop %s:%s crawler promotiom item num=%s",
shop_id, url, i)
else:
logger.warning("shop %s:%s not promotion info",
shop_id, url)
else:
logger.error("shop %s:%s crawler userId failed",
shop_id, url)
else:
logger.error("shop %s:%s not dongtai page", shop_id, url)
except:
logger.error("shop %s:%s crawler failed %s", shop_id, url,
traceback.format_exc())
