def parseList(self, response):
source = response.meta['source']
wx_account = response.meta['wx_account']
url = response.meta['url']
body = EncodeUtil.toUnicode(response.body)
# 判断被禁止 提示需要重启路由 清理cookie
if response.status == 302:
# 更新状态为更新失败
self.logDao.warn(u'您的访问过于频繁,重新拨号')
self.wxSourceDao.updateStatus(wx_account, 'updateFail')
# 获取Ip # 同时空线程30s
NetworkUtil.getNewIp()
TimerUtil.sleep(30)
else:
self.logDao.info(u'开始解析:' + wx_account)
# 进行解析
selector = Selector(text=body)
results = selector.xpath('//*[@id="main"]/div[4]/ul/li')
self.logDao.info(u'列表长度:' + str(len(results)))
hasCatch = False
for result in results:
wx_name = result.xpath('//*[@id="sogou_vr_11002301_box_0"]/div/div[2]/p[1]/a/text()').extract_first()
wx_account_ = result.xpath('//p[@class="info"]/label/text()').extract_first()
wx_url = result.xpath('//p[@class="tit"]/a/@href').extract_first()
if wx_account_ == wx_account:
self.logDao.info(u'成功抓取:' + wx_account_)
self.wxSourceDao.updateSource(wx_account, wx_name, wx_url, 'last')
hasCatch = True
break
if not hasCatch:
self.logDao.info(u'没有抓到:' + wx_account_)
self.wxSourceDao.updateStatus(wx_account, 'none')
pass
def start_requests(self):
# 检测网络
while not NetworkUtil.checkNetWork():
# 20s检测一次
TimerUtil.sleep(20)
self.logDao.warn(u'检测网络不可行')
# 检测服务器
while not NetworkUtil.checkService():
# 20s检测一次
TimerUtil.sleep(20)
self.logDao.warn(u'检测服务器不可行')
# 必读 玩物 产品榜 快报 游戏要闻 单品 盘点 花边要闻 游戏快报
cids = ['6', '66', '73', '84', '100', '119', '120', '121', '122']
# 必读
url = 'https://a.jiemian.com/index.php?m=lists&a=ajaxlist&callback=&_=1502103362598&page='
for cid in cids:
for page in range(1, 2):
newUrl = url + str(page) + ('&cid=' + str(cid))
self.logDao.warn(u'进行抓取列表:' + newUrl)
yield scrapy.Request(url=newUrl,
meta={
'request_type': 'jiemian_page_list',
'url': newUrl
},
callback=self.parseArticleList,
dont_filter=True)
def start_requests(self):
# 如果正在爬，就不请求
status = self.getStatus()
if status == 'running':
self.isRunningStop = True
return
self.saveStatus('running')
# 检测网络
while not NetworkUtil.checkNetWork():
# 20s检测一次
TimerUtil.sleep(20)
self.logDao.warn(u'检测网络不可行')
# 检测服务器
while not NetworkUtil.checkService():
# 20s检测一次
TimerUtil.sleep(20)
self.logDao.warn(u'检测服务器不可行')
src_channel = u'第一财经'
sub_channel = u'新闻'
# 进行页面访问
newUrl = 'http://www.yicai.com/news/'
self.logDao.warn(u'进行抓取列表:' + newUrl)
yield scrapy.Request(url=newUrl,
meta={
'request_type': 'diyicaijing_page_list',
'url': newUrl,
'src_channel': src_channel,
'sub_channel': sub_channel
},
callback=self.parseArticleList, dont_filter=True)
def start_requests(self):
# while True:
# 检测网络
while not NetworkUtil.checkNetWork():
# 20s检测一次
TimerUtil.sleep(20)
self.logDao.warn(u'检测网络不可行')
# continue
# 检测服务器
while not NetworkUtil.checkService():
# 20s检测一次
TimerUtil.sleep(20)
self.logDao.warn(u'检测服务器不可行')
# continue
for page in range(1, 2):
r = random.uniform(0, 1)
url = 'http://feed.mix.sina.com.cn/api/roll/get?pageid=372&lid=2431&k=&num=50&callback=&_=1501148356254&page='
newUrl = url + str(page) + '&r=' + str(r)
self.logDao.info(u"开始抓取列表：" + newUrl)
yield scrapy.Request(url=newUrl,
meta={
'request_type': 'sina_list',
'url': newUrl
},
callback=self.parseList2)
def start_requests(self):
# while True:
# 检测网络
while not NetworkUtil.checkNetWork():
# 20s检测一次
TimerUtil.sleep(20)
self.logDao.warn(u'检测网络不可行')
# continue
# 检测服务器
while not NetworkUtil.checkService():
# 20s检测一次
TimerUtil.sleep(20)
self.logDao.warn(u'检测服务器不可行')
# continue
# 进行爬虫
url = 'http://roll.news.sina.com.cn/interface/rollnews_ch_out_interface.php?col=96&spec=&type=&ch=01&k=&offset_page=0&offset_num=0&num=220&asc=&page=1'
# url = 'http://tech.sina.com.cn/t/2017-07-24/doc-ifyihrit1274195.shtml'
r = random.uniform(0, 1)
newUrl = url + ('&r=' + str(r))
self.logDao.info(u"开始抓取列表：" + newUrl)
yield scrapy.Request(url=newUrl, meta={'request_type': 'sina_list', 'url': newUrl}, callback=self.parseList)
# 补缺补漏
url = 'http://feed.mix.sina.com.cn/api/roll/get?pageid=372&lid=2431&k=&num=50&page=1&callback=&_=1501148356254'
r = random.uniform(0, 1)
newUrl = url + ('&r=' + str(r))
self.logDao.info(u"开始抓取列表：" + newUrl)
yield scrapy.Request(url=newUrl, meta={'request_type': 'sina_list', 'url': newUrl}, callback=self.parseList2)
def start_requests(self):
# 如果正在爬，就不请求
status = self.getStatus()
if status == 'running':
self.isRunningStop = True
return
self.saveStatus('running')
# 检测网络
while not NetworkUtil.checkNetWork():
# 20s检测一次
TimerUtil.sleep(20)
self.logDao.warn(u'检测网络不可行')
# 检测服务器
while not NetworkUtil.checkService():
# 20s检测一次
TimerUtil.sleep(20)
self.logDao.warn(u'检测服务器不可行')
# 进行页面访问
newUrl = 'http://tech.qq.com/l/scroll.htm'
self.logDao.warn(u'进行抓取列表:' + newUrl)
yield scrapy.Request(url=newUrl,
meta={
'request_type': 'demoName_page_list',
'url': newUrl
},
callback=self.parseArticleList,
dont_filter=True)
def start_requests(self):
# TODO..加上while可能有问题，有些抓不到
# 检测网络
while not NetworkUtil.checkNetWork():
# 20s检测一次
TimerUtil.sleep(20)
self.logDao.warn(u'检测网络不可行')
# 检测服务器
while not NetworkUtil.checkService():
# 20s检测一次
TimerUtil.sleep(20)
self.logDao.warn(u'检测服务器不可行')
# 进行爬虫
# 获取源 可用的，且（是更新失败的，或者最新的同时更新时间跟当前相比大于40分钟）
sources = self.wxSourceDao.queryEnable_special(isRandom=True, wx_accounts=['CINNO_CreateMore'])
for source in sources:
# 更新当前条状态为 更新中，如果更新失败或者被绊则更新为更新失败，更新成功之后设置为成功
(wx_name, wx_account, wx_url, wx_avatar, update_status, is_enable, update_time) = source
# 更新状态为更新中
self.wxSourceDao.updateStatus(wx_account, 'updating')
# 进行页面访问
url = 'http://weixin.sogou.com/weixin?type=1&s_from=input&ie=utf8&_sug_=n&_sug_type_=&query='
newUrl = url + wx_account
self.logDao.warn(u'进行抓取:' + newUrl)
yield scrapy.Request(url=newUrl,
meta={'request_type': 'weixin_source', 'url': newUrl,
'wx_account': wx_account, 'source': source},
callback=self.parseList, dont_filter=True)
def parseArticleList(self, response):
body = EncodeUtil.toUnicode(response.body)
selector = Selector(text=body)
source_url = response.meta['source_url']
print source_url
title = selector.xpath('//title/text()').extract_first('').strip(u' ')
isN = u"请输入验证码" == title
if isN or response.status == 302:
self.logDao.info(u'访问过多被禁止,重新拨号')
# 获取Ip # 同时空线程30s
NetworkUtil.getNewIp()
TimerUtil.sleep(50)
NetworkUtil.openWebbrowser(source_url)
else:
source = response.meta['source']
wx_account = response.meta['wx_account']
wx_account_id = response.meta['wx_account_id']
self.logDao.info(u'开始解析列表:' + wx_account)
# 进行解析
articleJS = selector.xpath('//script/text()').extract()
for js in articleJS:
if 'var msgList = ' in js:
p8 = re.compile('var\s*msgList\s*=.*;')
matchList = p8.findall(js)
for match in matchList:
match = match.lstrip('var msgList = ').rstrip(';')
# 格式化
articles = demjson.decode(match) or {}
articles = articles['list'] or []
self.logDao.info(u'匹配到文章列表' + wx_account)
for article in articles:
app_msg_ext_info = article.get(
'app_msg_ext_info') or {}
desc = app_msg_ext_info.get('digest') or ''
title = app_msg_ext_info.get('title') or ''
# 如果存在则不抓取
if self.checkDao.checkExist(title, wx_account, 1):
self.logDao.info(u'已经存在' + wx_account + ':' +
title)
continue
detailUrl = app_msg_ext_info['content_url'] or ''
if not detailUrl:
continue
detailUrl = "http://mp.weixin.qq.com" + detailUrl
detailUrl = detailUrl.replace("amp;", "")
self.logDao.info(u'抓取' + wx_account + ':' + title +
':' + detailUrl)
yield scrapy.Request(url=detailUrl,
meta={
'request_type':
'weixin_detail',
'wx_account': wx_account,
"source": source,
"title": title,
'wx_account_id':
wx_account_id,
"source_url": detailUrl
},
callback=self.parseArticle)
def start_requests(self):
# 如果在晚上12点到早上6点不爬
hour = datetime.datetime.now().hour
if 0 <= hour <= 6:
self.logDao.info(u'这个时间不爬。0-6点')
return
# 如果正在爬，就不请求
status = self.getStatus()
if status == 'running':
return
self.saveStatus('running')
# 检测网络
while not NetworkUtil.checkNetWork():
# 20s检测一次
TimerUtil.sleep(20)
self.logDao.warn(u'检测网络不可行')
# 检测服务器
while not NetworkUtil.checkService():
# 20s检测一次
TimerUtil.sleep(20)
self.logDao.warn(u'检测服务器不可行')
# 获取源 可用有值
sources = self.wxSourceDao.queryWxUrl(isRandom=True)
# 排序优先
update_time, brokenAccounts = self.getBrokenAccounts()
firstGroup = []
secondGroup = []
for source in sources:
(id, wx_name, wx_account, wx_url, wx_avatar, update_status,
is_enable, update_time) = source
if wx_account in brokenAccounts:
firstGroup.append(source)
else:
secondGroup.append(source)
sources = firstGroup + secondGroup
self.wxSources = sources
for source in sources:
(id, wx_name, wx_account, wx_url, wx_avatar, update_status,
is_enable, update_time) = source
# 进行页面访问
newUrl = wx_url
self.logDao.warn(u'进行抓取:' + newUrl)
yield scrapy.Request(url=newUrl,
meta={
'request_type': 'weixin_page_list',
'source_url': newUrl,
'wx_account': wx_account,
'source': source,
'wx_account_id': id
},
callback=self.parseArticleList,
dont_filter=True)
def start_requests(self):
# while True:
# 检测网络
if not NetworkUtil.checkNetWork():
# 20s检测一次
TimerUtil.sleep(20)
self.logDao.warn(u'检测网络不可行')
# continue
# 检测服务器
if not NetworkUtil.checkService():
# 20s检测一次
TimerUtil.sleep(20)
self.logDao.warn(u'检测服务器不可行')
# continue
if self.request_stop:
# 拨号生效时间不定，所以需要间隔一段时间再重试
timeSpace = time.time() - self.request_stop_time
if timeSpace / 60 <= 2:
# 当时间间隔小于 2分钟 就不请求
# continue
pass
else:
self.request_stop = False
# 进行爬虫
url = 'http://roll.news.sina.com.cn/interface/rollnews_ch_out_interface.php?col=30&spec=&type=&ch=05&k' \
'=&offset_page=0&offset_num=0&num=60&asc=&page='
for page in range(0, 11):
if self.request_stop:
self.logDao.warn(u'出现被绊或者出现网络异常，退出循环')
# 当网络出现被绊的情况，就需要停止所有的请求等待IP更换
break
r = random.uniform(0, 1)
newUrl = url + str(page)
newUrl += ('&r=' + str(r))
self.logDao.info(u"开始抓取列表：" + newUrl)
yield scrapy.Request(url=newUrl,
meta={'url': newUrl},
callback=self.parseList)
# 跑空线程2秒
TimerUtil.sleep(2)
if self.request_stop:
# 需要发起通知 进行重新拨号
self.logDao.warn(u'发送重新拨号信号，请等待2分钟会尝试重新抓取')
self.request_stop_time = time.time()
pass
else:
# 正常抓好之后，当前跑空线程10分钟，不影响一些还没请求完成的request
self.logDao.info(u'请求了一轮了，但是可能还有没有请求完成，睡一会10分钟')
TimerUtil.sleep(10 * 60)
pass
def start_requests(self):
# 检测网络
while not NetworkUtil.checkNetWork():
# 20s检测一次
TimerUtil.sleep(20)
self.logDao.warn(u'检测网络不可行')
# 检测服务器
while not NetworkUtil.checkService():
# 20s检测一次
TimerUtil.sleep(20)
self.logDao.warn(u'检测服务器不可行')
src_channel = u'网易科技'
sub_channel = u'科技'
# 进行页面访问
newUrl = 'http://tech.163.com/special/00094IHV/news_json.js?' + str(
random.uniform(0, 1))
self.logDao.warn(u'进行抓取列表:' + newUrl)
yield scrapy.Request(url=newUrl,
meta={
'request_type': 'wangyi_page_list',
'url': newUrl,
'src_channel': src_channel,
'sub_channel': sub_channel
},
callback=self.parseArticleList,
dont_filter=True)
src_channel = u'网易财经'
sub_channel = u'财经'
# 进行页面访问
newUrl = 'http://money.163.com/special/00251G8F/news_json.js?' + str(
random.uniform(0, 1))
self.logDao.warn(u'进行抓取列表:' + newUrl)
yield scrapy.Request(url=newUrl,
meta={
'request_type': 'wangyi_page_list',
'url': newUrl,
'src_channel': src_channel,
'sub_channel': sub_channel
},
callback=self.parseArticleList,
dont_filter=True)
def start_requests(self):
# 检测网络
while not NetworkUtil.checkNetWork():
# 20s检测一次
TimerUtil.sleep(20)
self.logDao.warn(u'检测网络不可行')
# 检测服务器
while not NetworkUtil.checkService():
# 20s检测一次
TimerUtil.sleep(20)
self.logDao.warn(u'检测服务器不可行')
# 进行页面访问
src_channel = u'腾讯科技'
sub_channel = u'科技'
newUrl = 'http://tech.qq.com/l/scroll.htm'
self.logDao.warn(u'进行抓取列表:' + newUrl)
yield scrapy.Request(url=newUrl,
meta={
'request_type': 'tengxun_page_list',
'url': newUrl,
'src_channel': src_channel,
'sub_channel': sub_channel
},
callback=self.parseArticleList, dont_filter=True)
# 进行页面访问
src_channel = u'腾讯财经'
sub_channel = u'财经要闻'
newUrl = 'http://finance.qq.com/'
self.logDao.warn(u'进行抓取列表:' + newUrl)
yield scrapy.Request(url=newUrl,
meta={
'request_type': 'tengxun_page_list',
'url': newUrl,
'src_channel': src_channel,
'sub_channel': sub_channel
},
callback=self.parseArticleList2, dont_filter=True)
def start_requests(self):
# unKnow = ["didalive", "HIS_Technology", "ad_helper", "zhongduchongdu"]; 是搜索不到的
# TODO..加上while可能有问题，有些可能抓不到
# 如果正在爬，就不请求
# 检测网络
while not NetworkUtil.checkNetWork():
# 20s检测一次
TimerUtil.sleep(20)
self.logDao.warn(u'检测网络不可行')
# 检测服务器
while not NetworkUtil.checkService():
# 20s检测一次
TimerUtil.sleep(20)
self.logDao.warn(u'检测服务器不可行')
# 获取源 可用有值
sources = self.wxSourceDao.queryWxUrl_special(isRandom=True,
wx_accounts=['qqtech'])
for source in sources:
(id, wx_name, wx_account, wx_url, wx_avatar, update_status,
is_enable, update_time) = source
# 进行页面访问
newUrl = wx_url
self.logDao.warn(u'进行抓取:' + newUrl)
yield scrapy.Request(url=newUrl,
meta={
'request_type': 'weixin_page_list',
'source_url': newUrl,
'wx_account': wx_account,
'source': source,
'wx_account_id': id
},
callback=self.parseArticleList,
dont_filter=True)
def start_requests(self):
# 如果正在爬，就不请求
status = self.getStatus()
if status == 'running':
self.isRunningStop = True
return
self.saveStatus('running')
# 检测网络
while not NetworkUtil.checkNetWork():
# 20s检测一次
TimerUtil.sleep(20)
self.logDao.warn(u'检测网络不可行')
# 检测服务器
while not NetworkUtil.checkService():
# 20s检测一次
TimerUtil.sleep(20)
self.logDao.warn(u'检测服务器不可行')
src_channel = u'搜狐科技'
sub_channel = u'科技'
for page in range(1, 4):
# 进行页面访问
url = 'http://v2.sohu.com/public-api/feed?scene=CHANNEL&sceneId=30&size=20&callback=&_=1502075449669&page='
newUrl = url + str(page)
self.logDao.warn(u'进行抓取列表:' + newUrl)
yield scrapy.Request(url=newUrl,
meta={
'request_type': 'sohu_page_list',
'url': newUrl,
'src_channel': src_channel,
'sub_channel': sub_channel
},
callback=self.parseArticleList,
dont_filter=True)
def start_requests(self):
# 如果正在爬，就不请求
status = self.getStatus()
if status == 'running':
self.isRunningStop = True
return
self.saveStatus('running')
# 检测网络
while not NetworkUtil.checkNetWork():
# 20s检测一次
TimerUtil.sleep(20)
self.logDao.warn(u'检测网络不可行')
# 检测服务器
while not NetworkUtil.checkService():
# 20s检测一次
TimerUtil.sleep(20)
self.logDao.warn(u'检测服务器不可行')
# 必读 玩物 产品榜 快报 游戏要闻 单品 盘点 花边要闻 游戏快报
cids = [{
'src_channel': u'界面科技',
'sub_channel': u'必读',
'num': '6'
}, {
'src_channel': u'界面科技',
'sub_channel': u'玩物',
'num': '66'
}, {
'src_channel': u'界面科技',
'sub_channel': u'产品榜',
'num': '73'
}, {
'src_channel': u'界面科技',
'sub_channel': u'快报',
'num': '84'
}, {
'src_channel': u'界面游戏',
'sub_channel': u'游戏要闻',
'num': '100'
}, {
'src_channel': u'界面游戏',
'sub_channel': u'单品',
'num': '119'
}, {
'src_channel': u'界面游戏',
'sub_channel': u'盘点',
'num': '120'
}, {
'src_channel': u'界面游戏',
'sub_channel': u'花边要闻',
'num': '121'
}, {
'src_channel': u'界面游戏',
'sub_channel': u'游戏快报',
'num': '122'
}]
# 必读
url = 'https://a.jiemian.com/index.php?m=lists&a=ajaxlist&callback=&_=1502103362598&page='
for cid in cids:
for page in range(1, 2):
cidNum = cid.get('num')
src_channel = cid.get('src_channel')
sub_channel = cid.get('sub_channel')
newUrl = url + str(page) + ('&cid=' + cidNum)
self.logDao.warn(u'进行抓取列表:' + newUrl)
yield scrapy.Request(url=newUrl,
meta={
'request_type': 'jiemian_page_list',
'url': newUrl,
'src_channel': src_channel,
'sub_channel': sub_channel
},
callback=self.parseArticleList,
dont_filter=True)
def start_requests(self):
# while True:
# 检测网络
while not NetworkUtil.checkNetWork():
# 20s检测一次
TimerUtil.sleep(20)
self.logDao.warn(u'检测网络不可行')
# continue
# 检测服务器
while not NetworkUtil.checkService():
# 20s检测一次
TimerUtil.sleep(20)
self.logDao.warn(u'检测服务器不可行')
# continue
src_channel = u'新浪科技'
sub_channel = u'科技'
# 进行爬虫
url = 'http://roll.news.sina.com.cn/interface/rollnews_ch_out_interface.php?col=96&spec=&type=&ch=01&k=&offset_page=0&offset_num=0&num=220&asc=&page=1'
r = random.uniform(0, 1)
newUrl = url + ('&r=' + str(r))
self.logDao.info(u"开始抓取列表：" + newUrl)
yield scrapy.Request(url=newUrl,
meta={
'request_type': 'sina_list',
'url': newUrl,
'src_channel': src_channel,
'sub_channel': sub_channel
},
callback=self.parseList)
src_channel = u'新浪科技'
sub_channel = u'科技'
# 补缺补漏
url = 'http://feed.mix.sina.com.cn/api/roll/get?pageid=372&lid=2431&k=&num=50&page=1&callback=&_=1501148356254'
r = random.uniform(0, 1)
newUrl = url + ('&r=' + str(r))
self.logDao.info(u"开始抓取列表：" + newUrl)
yield scrapy.Request(url=newUrl,
meta={
'request_type': 'sina_list',
'url': newUrl,
'src_channel': src_channel,
'sub_channel': sub_channel
},
callback=self.parseList2)
# 新浪财经 要闻
src_channel = u'新浪财经'
sub_channel = u'要闻'
url = 'http://finance.sina.com.cn/'
newUrl = url
self.logDao.info(u"开始抓取列表：" + newUrl)
yield scrapy.Request(url=newUrl,
meta={
'request_type': 'sina_list',
'url': newUrl,
'src_channel': src_channel,
'sub_channel': sub_channel
},
callback=self.parseList3)
def parseArticle(self, response):
body = EncodeUtil.toUnicode(response.body)
selector = Selector(text=body)
title = selector.xpath('//title/text()').extract_first('').strip(u' ')
source_url = response.meta['source_url']
wx_account = response.meta['wx_account']
isN = u"请输入验证码" == title
if isN or response.status == 302:
self.logDao.info(u'访问过多被禁止,重新拨号')
# 存起来
self.brokenAccounts.append(wx_account)
# 获取Ip # 同时空线程30s
NetworkUtil.getNewIp()
TimerUtil.sleep(80)
NetworkUtil.openWebbrowser(source_url)
else:
title = response.meta['title']
source_url = response.meta['source_url']
wx_account_id = response.meta['wx_account_id']
self.logDao.info(u'开始解析文章' + wx_account + ':' + title + ':' +
source_url)
self.logDao.info(u'开始解析文章：' + source_url)
# 进行解析
post_date = selector.xpath(
'//*[@id="post-date"]/text()').extract_first('')
try:
post_date = time.strftime("%Y-%m-%d %H:%M:%S",
time.strptime(post_date, "%Y-%m-%d"))
except Exception:
pass
styles = selector.xpath('//style/text()').extract()
styles = CssUtil.compressCss(styles).replace('\'', '"').replace(
'\\', '\\\\')
styles = CssUtil.clearUrl(styles)
styles = CssUtil.clearBackgroundColor(styles, ['#f3f3f3'])
post_user = selector.xpath(
'//*[@id="post-user"]/text()').extract_first('')
content_html = selector.xpath('//*[@id="js_content"]')
if not len(content_html):
self.logDao.info(u'不存在内容：' + source_url)
return
# 去除内部不需要的标签
content_items = content_html.xpath('*')
if not len(content_items):
self.logDao.info(u'不存在内容：' + source_url)
return
# content_items_new = []
# for item in content_items:
# itemStr = item.extract()
# if u'订阅微信' in itemStr:
# continue
# content_items_new.append(item)
# content_items = content_items_new
# 得到纯文本
content_txt = []
for item in content_items:
# 文本
allTxt = item.xpath('.//text()').extract()
allTxt = ''.join(allTxt).replace('\t', '')
# 加入
content_txt.append(allTxt)
content_txt = '\n'.join(content_txt)
# 组装新的内容标签
outHtml = """<div class="rich_media_content " id="js_content">${++content++}</div>"""
content_items = content_items.extract()
content_items = ''.join(content_items)
content_html = outHtml.replace('${++content++}', content_items)
selector = Selector(text=content_html)
# 解析文档中的所有图片url，然后替换成标识
image_urls = []
imgs = selector.xpath('descendant::img')
for img in imgs:
# 图片可能放在src 或者data-src
image_url = img.xpath('@src | @data-src').extract_first('')
if image_url and image_url.startswith('http'):
self.logDao.info(u'得到图片：' + image_url)
image_urls.append({
'url': image_url,
})
self.logDao.info(wx_account + u'得到文章：' + title + ":" + post_date +
':' + post_user)
self.logDao.info(u'得到文章：' + source_url)
# 得到hashCode1
hash_code = self.checkDao.getHashCode(title, wx_account, 1)
self.saveFile(hash_code, body)
# 去除 image 的 alt title
selector = Selector(text=content_html)
imgAltTitles = selector.xpath('//img/@alt|//img/@title').extract()
# 处理提示块img的 alt title, 关注//img/@alt|//img/@title
for imgAltTitle in imgAltTitles:
if imgAltTitle.strip(' '):
content_html = content_html.replace(imgAltTitle, '')
contentItem = ContentItem()
contentItem['content_txt'] = content_txt
contentItem['image_urls'] = image_urls
contentItem['title'] = title
contentItem['source_url'] = source_url
contentItem['post_date'] = post_date
contentItem['sub_channel'] = ''
contentItem['post_user'] = post_user
contentItem['tags'] = ''
contentItem['styles'] = styles
contentItem['content_html'] = content_html
contentItem['hash_code'] = hash_code
contentItem['info_type'] = 1
contentItem['src_source_id'] = 1
contentItem['src_account_id'] = wx_account_id
contentItem['src_channel'] = '微信公众号'
contentItem['src_ref'] = ''
contentItem['wx_account'] = wx_account
return contentItem
def start_requests(self):
# 如果正在爬，就不请求
status = self.getStatus()
if status == 'running':
self.isRunningStop = True
return
self.saveStatus('running')
# 检测网络
while not NetworkUtil.checkNetWork():
# 20s检测一次
TimerUtil.sleep(20)
self.logDao.warn(u'检测网络不可行')
# 检测服务器
while not NetworkUtil.checkService():
# 20s检测一次
TimerUtil.sleep(20)
self.logDao.warn(u'检测服务器不可行')
src_channel = u'凤凰财经'
sub_channel = u'电子竞技'
url = 'http://games.ifeng.com/listpage/17886/1/list.shtml'
styleUrlDefault = [
'http://p2.ifengimg.com/a/2016/0523/esports.css',
'http://y1.ifengimg.com/package/t_20130820__15953/css/pl_detail_v8.css'
]
newUrl = url
self.logDao.warn(u'进行抓取列表:' + newUrl)
yield scrapy.Request(url=newUrl,
meta={
'request_type': 'fenghuang_page_list',
'url': newUrl,
'src_channel': src_channel,
'sub_channel': sub_channel,
'styleUrlDefault': styleUrlDefault
},
callback=self.parseArticleList3,
dont_filter=True)
sub_channel = u'产品资讯'
url = 'http://games.ifeng.com/listpage/27456/1/list.shtml'
styleUrlDefault = [
'http://p0.ifengimg.com/fe/responsiveDetail/styles/pc_c38f5a0e.css'
]
newUrl = url
self.logDao.warn(u'进行抓取列表:' + newUrl)
yield scrapy.Request(url=newUrl,
meta={
'request_type': 'fenghuang_page_list',
'url': newUrl,
'src_channel': src_channel,
'sub_channel': sub_channel,
'styleUrlDefault': styleUrlDefault
},
callback=self.parseArticleList,
dont_filter=True)
sub_channel = u'热点资讯'
url = 'http://games.ifeng.com/listpage/27455/1/list.shtml'
styleUrlDefault = [
'http://p0.ifengimg.com/fe/responsiveDetail/styles/pc_c38f5a0e.css'
]
newUrl = url
self.logDao.warn(u'进行抓取列表:' + newUrl)
yield scrapy.Request(url=newUrl,
meta={
'request_type': 'fenghuang_page_list',
'url': newUrl,
'src_channel': src_channel,
'sub_channel': sub_channel,
'styleUrlDefault': styleUrlDefault
},
callback=self.parseArticleList,
dont_filter=True)
src_channel = u'凤凰科技'
sub_channel = u'资讯'
url = 'http://tech.ifeng.com/listpage/800/0/1/rtlist.shtml'
styleUrlDefault = [
'http://p0.ifengimg.com/fe/responsiveDetail/styles/pc_c38f5a0e.css'
]
newUrl = url
self.logDao.warn(u'进行抓取列表:' + newUrl)
yield scrapy.Request(url=newUrl,
meta={
'request_type': 'fenghuang_page_list',
'url': newUrl,
'src_channel': src_channel,
'sub_channel': sub_channel,
'styleUrlDefault': styleUrlDefault
},
callback=self.parseArticleList2,
dont_filter=True)