def crawl_item_details(queue, i):
shopall = ShopAllItemDb1()
agentIp = Utils.GetAgentIp()
# agentIp = "127.0.0.1:8888"
driver = PhantomDriver(1, agentIp, 60).return_driver()
while (True):
try:
nid = None
nid = queue.get(block=False)
except:
pass
try:
if nid:
item = CrawlItem.crawl(driver, nid)
if item:
shopall.insert_or_update_details(item)
else:
#如果抓取失败就更换代理ip并重新放到队列中
agentIp = Utils.GetAgentIp()
try:
queue.put(nid, block=False)
except:
pass
print("进程%s抓取%s宝贝详情失败,重试 Ip:%s" % (i, nid, agentIp))
except Exception, e:
agentIp = Utils.GetAgentIp()
driver.quit()
driver = PhantomDriver(1, agentIp, 60).return_driver()
try:
queue.put(nid, block=False)
except:
pass
print("进程%s抓取%s宝贝详情失败:%s,退出浏览器重试" % (i, nid, e.message))
driver.quit()
def crawl_shop_all_item(self):
agentIp = Utils.GetAgentIp()
# agentIp="127.0.0.1:8888"
driver = PhantomDriver(2, agentIp, 60)
parms_url = "{shop_url}/i/asynSearch.htm?_ksTS={now}569_240&callback=jsonp241&mid=w-14766145001-0&wid=14766145001&path=/search.htm&search=y&orderType=hotsell_desc&scene=taobao_shop&pageNo={page_num}"
url = "{shop_url}/search.htm?&search=y&orderType=hotsell_desc&scene=taobao_shop".format(shop_url=self.shop_url)
print (url)
result = driver.download_no_quit(url)
shop_id = -1
source = result['page_source']
if result['ok']:
html = etree.HTML(source)
shop_items = []
if html is not None and 'page-info' in source and html.xpath("//span[contains(@class,'page-info')]/text()"):
total = int(html.xpath("//span[contains(@class,'page-info')]/text()")[0].split('/')[1])
if html.xpath("//meta[@name='microscope-data']"):
for meta in html.xpath("//meta[@name='microscope-data']")[0].get('content').split(';'):
if 'shopid' in meta.lower():
shop_id = meta.split("=")[1]
self.shopall.format_data(shop_id,False)
shop_items.extend(self.parse_items(html, shop_id))
driver = PhantomDriver(1, agentIp, 60)
for i in range(1, total):
page_num = i + 1
print("page%s" % page_num)
url = parms_url.format(shop_url=self.shop_url, now=int(time.time()), page_num=page_num)
result = driver.download_no_quit(url)
if result['ok']:
html = etree.HTML(result['page_source'])
if result['ok'] and 'page-info' in source and html.xpath("//span[contains(@class,'page-info')]/text()"):
results = self.parse_items(html, shop_id)
shop_items.extend(results)
self.shopall.insert_or_update(shop_items)
else:
print("无法获取%s" % agentIp)
return -1
return shop_id
