def step2bbs(self, params):
Logger.getlogging().info("Dm5Commnets.STEP_2")
# 将STEP_1中的docurl传下来
docurl = params.customized['docurl']
comments_count = self.r.parse(ur'(\d+)个回复', params.content)[0]
# 判断增量
cmtnum = URLStorage.getcmtnum(params.originalurl)
if cmtnum >= comments_count:
return
URLStorage.setcmtnum(params.originalurl, comments_count)
# 总数除以page_size，然后加1，可得到评论总页数comments_count
pagenum = 0
xparser = XPathUtility(params.content)
if not xparser.xpath('//*[@class="inkk ma5"]'):
Logger.getlogging().warning('{0}:30001'.format(params.originalurl))
return
pageList = xparser.xpath('//*[@id="search_fy"]/a/text()')
if not pageList:
pagenum = 1
else:
pagenum = int(pageList[-2])
for page in range(1, pagenum + 1, 1):
comment_url = Dm5Commnets.COMMENT_URL.format(docurl=docurl,
page=page)
self.storeurl(comment_url, params.originalurl,
Dm5Commnets.STEP_3_BBS)
def step2bbs(self, params):
Logger.getlogging().info("Ea3wcomments.STEP_2")
commentinfo_url = params.customized['commentinfo_url'] + "&load=all"
xparser = XPathUtility(params.content)
comments_count = xparser.getnumber('//div[@class="at-comment"]/a/span')
# 保存页面评论量
cmtnum = URLStorage.getcmtnum(params.originalurl)
if cmtnum >= comments_count:
return
URLStorage.setcmtnum(params.originalurl, comments_count)
self.storeurl(commentinfo_url, params.originalurl,
Ea3wcomments.STEP_3_BBS)
def step2(self, params):
Logger.getlogging().info("Dm123NewsComments.STEP_2")
classid = params.customized['classid']
id = params.customized['id']
xparser = XPathUtility(params.content)
# 评论总数(当评论不满一页时，直接获取到的comments_count为0)
comments_count = xparser.getnumber('//div/a[1]/b')
# comments_count为0时分两种情况，真的没有评论和有评论
if 0 == comments_count:
commentsinfos = xparser.getcomments('//div[@class="rbvalueout"]')
commentstimes = xparser.getcomments('//span[@class="rbtime"]')
# comments_count重新赋值
comments_count = len(commentsinfos)
if 0 == comments_count:
return
else:
# 判断增量
cmtnum = URLStorage.getcmtnum(params.originalurl)
if cmtnum >= comments_count:
return
URLStorage.setcmtnum(params.originalurl, comments_count)
self.storeurl(params.originalurl, params.originalurl, Dm123NewsComments.STEP_3,
{'is_only_one_page': True, 'commentsinfos': commentsinfos,
'commentstimes': commentstimes})
else:
# 判断增量
cmtnum = URLStorage.getcmtnum(params.originalurl)
if cmtnum >= comments_count:
return
URLStorage.setcmtnum(params.originalurl, comments_count)
# 评论页数
page_count = int(math.ceil(float(comments_count) / self.page_size))
for page in range(0, int(page_count), 1):
comment_url = Dm123NewsComments.COMMENT_URL.format(page=page, classid=classid, id=id)
self.storeurl(comment_url, params.originalurl, Dm123NewsComments.STEP_3,
{'is_only_one_page': False})
def step2bbs(self, params):
Logger.getlogging().info("BaozouNewsComments.STEP_2")
topic_id = params.customized['topic_id']
commentsinfo = json.loads(params.content)
comments_count = commentsinfo['total_count']
# 保存页面评论量
cmtnum = URLStorage.getcmtnum(params.originalurl)
if cmtnum >= comments_count:
return
URLStorage.setcmtnum(params.originalurl, comments_count)
for index in range(1, int(commentsinfo['total_pages']) + 1, 1):
commentinfo_url = BaozouNewsComments.COMMENT_URL.format(
topic_id=topic_id, page=index)
self.storeurl(commentinfo_url, params.originalurl,
BaozouNewsComments.STEP_3_BBS)
def step2(self, params):
Logger.getlogging().info("ThirtysixKryptonComments.STEP_2")
# 将STEP_1中的cid传下来
cid = params.customized['cid']
jsoncontent = json.loads(params.content)
comments_count = jsoncontent['data']['total_items']
page_count = jsoncontent['data']['total_pages']
# 判断增量
cmtnum = URLStorage.getcmtnum(params.originalurl)
if cmtnum >= comments_count:
return
URLStorage.setcmtnum(params.originalurl, comments_count)
for page in range(1, page_count+1, 1):
commentinfo_url = ThirtysixKryptonComments.COMMENT_URL.format(cid, self.page_size, page)
self.storeurl(commentinfo_url, params.originalurl, ThirtysixKryptonComments.STEP_3)
def step2bbs(self, params):
Logger.getlogging().info("Tmtpostcommnets.STEP_2")
tid = params.customized['tid']
commentsinfo = json.loads(params.content)
comments_count = commentsinfo['cursor']['total']
# 保存页面评论量
cmtnum = URLStorage.getcmtnum(params.originalurl)
if cmtnum >= comments_count:
return
URLStorage.setcmtnum(params.originalurl, comments_count)
for index in range(0,
int(math.ceil(float(comments_count) / self.limit)),
1):
self.offset = index * self.limit
commentinfo_url = Tmtpostcommnets.COMMENT_URL.format(
tid=tid, limit=self.limit, offset=self.offset)
self.storeurl(commentinfo_url, params.originalurl,
Tmtpostcommnets.STEP_3_BBS)
def step2(self, params):
Logger.getlogging().info("Flash8Comments.STEP_2")
# 将STEP_1中的docurl传下来
docurl = params.customized['docurl']
xparser = XPathUtility(params.content)
commentsinfo = xparser.getstring('//div[@class="page"]/span/font[1]')
# 保存页面评论量
cmtnum = URLStorage.getcmtnum(params.originalurl)
if cmtnum >= int(commentsinfo[0]):
return
URLStorage.setcmtnum(params.originalurl, int(commentsinfo[0]))
# 总数除以page_size，然后加1，可得到评论总页数comments_count
pagecount = xparser.getnumber('//*[@class="pg"]/label/span')
if pagecount == 0:
pagecount = pagecount + 1
for page in range(1, pagecount + 1, 1):
comment_url = Flash8Comments.COMMENT_URL.format(docurl=docurl,
page=page)
self.storeurl(comment_url, params.originalurl,
Flash8Comments.STEP_3, {'page': page})
def step3bbs(self, params):
Logger.getlogging().info("Chinavaluecomments.STEP_3")
# Step3: 通过Step2设置的url，得到所有评论，抽取评论
params.content = params.content[1:len(params.content) - 1]
commentsinfo = json.loads(params.content)
comments_count = commentsinfo['RecordCount']
# 判断增量
cmtnum = URLStorage.getcmtnum(params.originalurl)
if cmtnum >= comments_count:
return
URLStorage.setcmtnum(params.originalurl, comments_count)
comments = []
for index in range(0,len(commentsinfo['CommentObjs'])):
# 提取时间
cmti = CommentInfo()
cmti.content = commentsinfo['CommentObjs'][index]['Content']
tm = TimeUtility.getuniformtime(TimeUtility.getuniformtime(commentsinfo['CommentObjs'][index]['AddTime'], u'%Y-%m-%d %H:%M'))
if URLStorage.storeupdatetime(params.originalurl, tm):
comments.append(cmti)
# 保存获取的评论
if len(comments) > 0:
self.commentstorage.store(params.originalurl, comments)
def step2(self, params):
""""""
uniqid = params.customized['uniqid']
domain = params.customized['domain']
url = params.customized['url']
jsondata = json.loads(params.content)
comments_count = int(jsondata['show']['total_num'])
# 检查评论数是否增加，没有增加，返回；有增加，更新增加后的页面评论量
cmtnum = URLStorage.getcmtnum(params.originalurl)
if cmtnum >= int(comments_count):
return
URLStorage.setcmtnum(params.originalurl, int(comments_count))
#3. 拼出获取所有评论的url
max = int(
math.ceil(float(comments_count) / HuyaComments.DEFAULT_PAGE_SIZE))
for page in range(1, max + 1, 1):
#num = (page - 1)*HuyaComments.DEFAULT_PAGE_SIZE
comments_url = HuyaComments.COMMENTS_URL.format(uniqid=uniqid,
domain=domain,
url=url,
page=page)
self.storeurl(comments_url, params.originalurl,
HuyaComments.STEP_3)
def process(self, params):
Logger.getlogging().info(params.url)
try:
if params.step is SeventeenKComments.STEP_1:
#Step1: 通过得到docurl，得到获取评论的首页url。
#Logger.getlogging().info("proparam.step is None")
# 在视频url中取出docurl，^http://v\.ifeng\.com\/\w+\/\w+/\d{6}\/[0-9a-z-]+\.shtml
# 取URL中的([0-9a-z-]+)参数，此参数为docurl
docurl = self.r.parse(
'^http://bbs\.17k\.com\/thread-(\d+)-\d+-1\.html',
params.originalurl)[0]
#Logger.getlogging().debug(docurl)
# 评论首页URL为http://comment.ifeng.com/getv.php?job=1&docurl=([0-9a-z-]+)&p=1
commentinfo_url = 'http://bbs.17k.com/thread-{docurl}-1-1.html'.format(
docurl=docurl)
self.storeurl(commentinfo_url, params.originalurl,
SeventeenKComments.STEP_2, {'docurl': docurl})
elif params.step == SeventeenKComments.STEP_2:
#将STEP_1中的docurl传下来
docurl = params.customized['docurl']
# Step2: 通过Step1设置url，得到评论的总数，并根据评论总数得到获取其他评论的url。
#Logger.getlogging().info("params.step == 2")
# 打开STEP1中URL，截取"count"：num字段，取出num的值，num字段为评论总数
xparser = XPathUtility(params.content)
commentsinfo = xparser.getnumber(
'//*[@class="hm ptn"]/span[5]')
#Logger.getlogging().debug(comments_count / self.page_size)
#Logger.getlogging().debug(math.ceil(comments_count / self.page_size))
# 保存页面评论量
cmtnum = URLStorage.getcmtnum(params.originalurl)
if cmtnum >= int(commentsinfo):
return
URLStorage.setcmtnum(params.originalurl, int(commentsinfo))
# 总数除以page_size，然后加1，可得到评论总页数comments_count
# 循环http://comment.ifeng.com/getv.php?job=1&docurl=([0-9a-z-]+)&p=comments_count，从一开始循环到上一步操作取到的数值，从而得到所有评论的URL,并保存
pagecount = xparser.getnumber('//*[@class="pg"]/label/span')
for page in range(1, pagecount + 1, 1):
comment_url = SeventeenKComments.COMMENT_URL.format(
docurl=docurl, page=page)
self.storeurl(comment_url, params.originalurl,
SeventeenKComments.STEP_3, {'page': page})
elif params.step == SeventeenKComments.STEP_3:
# Step3: 通过Step2设置的url，得到所有评论，抽取评论
#Logger.getlogging().info("params.step == 3")
page = params.customized['page']
xparser = XPathUtility(params.content)
commentsinfo = xparser.getcomments(
'//*[contains(@id,"postmessage")]')
commentstime = self.r.parse(ur'发表于 (\d+-\d+-\d+ \d+:\d+)</em>',
params.content)
comments = []
#获取评论
# 设置实际的评论量
if page is 1:
statrIndex = 1
else:
statrIndex = 0
for index in range(statrIndex, len(commentstime), 1):
cmti = CommentInfo()
if URLStorage.storeupdatetime(params.originalurl,
commentstime[index] + ':00'):
# 获取增加的评论（根据时间比较）
cmti.content = commentsinfo[index]
comments.append(cmti)
# 保存获取到的评论
if len(comments) > 0:
self.commentstorage.store(params.originalurl, comments)
else:
Logger.getlogging().error(
'proparam.step == {step}'.format(step=params.step))
except Exception, e:
traceback.print_exc()
def process(self, params):
Logger.getlogging().info(params.url)
try:
if params.step is zhulangComments.STEP_1:
#Step1: 通过得到docurl，得到获取评论的首页url参数。
articleId = self.r.parse('http://www.zhulang.com/(\d+)/',
params.originalurl)[0]
# 取得评论的url列表
comments_url = zhulangComments.COMMENT_URL % (articleId, 1)
self.storeurl(comments_url, params.originalurl,
zhulangComments.STEP_2, {'articleId': articleId})
elif params.step == zhulangComments.STEP_2:
# 获得评论参数
articleId = params.customized['articleId']
# 取得总件数
comment_count = float(self.r.getid('total', params.content))
if comment_count == 0:
return
# 判断增量
cmtnum = URLStorage.getcmtnum(params.originalurl)
if cmtnum >= comment_count:
return
URLStorage.setcmtnum(params.originalurl, comment_count)
# 获取页数
page = int(math.ceil(comment_count /
zhulangComments.PAGE_SIZE))
# 获得url列表
for page in range(0, page, 1):
url = zhulangComments.COMMENT_URL % (articleId, page)
self.storeurl(url, params.originalurl,
zhulangComments.STEP_3)
elif params.step == zhulangComments.STEP_3:
# Step3: 通过Step2设置的url，得到所有评论，抽取评论
Logger.getlogging().info("params.step == 3")
# 取得所有评论
comments = self.r.parse(r'<p class=\\"cmt-txt\\">(.+?)<\\/p>',
params.content)
# 取得所有评论时间
commenttimes = self.r.parse(
r'<span class=\\"cmt-time\\">(.+?)<\\/span>',
params.content)
index = 0
commentsInfo = []
# 取得所有评论
for index in range(index, int(len(comments)), 1):
# 提取时间
publicTime = commenttimes[index]
if URLStorage.storeupdatetime(params.originalurl,
publicTime):
cmti = CommentInfo()
x = json.loads('{"comment":"%s"}' %
comments[index].encode('utf8'))
cmti.content = (x['comment'])
commentsInfo.append(cmti)
# 保存获取的评论
if len(commentsInfo) > 0:
self.commentstorage.store(params.originalurl, commentsInfo)
else:
Logger.getlogging().error(
'proparam.step == {step}'.format(step=params.step))
except Exception, e:
traceback.print_exc()
def process(self, params):
Logger.getlogging().info(params.url)
try:
if params.step is Rain8Comments.STEP_1:
#Step1: 通过得到docurl，得到获取评论的首页url参数。
articleId = self.r.parse('http://\w+\.tadu\.com/\w+/(\d+).*', params.originalurl)[0]
# 取得评论的url列表
comments_url = Rain8Comments.COMMENT_URL.format (articleId = articleId,page = 1)
self.storeurl(comments_url, params.originalurl, Rain8Comments.STEP_2, {'articleId': articleId})
elif params.step == Rain8Comments.STEP_2:
# 获得评论参数
articleId = params.customized['articleId']
# 取得总件数
#comment_count = float(self.r.getid('total', params.content))
xparser = XPathUtility(params.content)
countstr = xparser.getstring('//h4')
if self.r.search(u'\d+', countstr):
comment_count = self.r.parse(u'(\d+)', countstr)[1]
if comment_count == 0:
return
# 判断增量
cmtnum = URLStorage.getcmtnum(params.originalurl)
if cmtnum >= comment_count:
return
URLStorage.setcmtnum(params.originalurl, comment_count)
# 获取页数
totalPage = int(math.ceil(float(comment_count) / TaDuComments.PAGE_SIZE))
# 获得url列表
for page in range(1, totalPage+1 , 1):
url = TaDuComments.COMMENT_URL.format(articleId = articleId,page = page)
self.storeurl(url, params.originalurl, TaDuComments.STEP_3)
elif params.step == TaDuComments.STEP_3:
# Step3: 通过Step2设置的url，得到所有评论，抽取评论
Logger.getlogging().info("params.step == 3")
# 取得所有评论
xparser = XPathUtility(params.content)
comments = xparser.getlist('//ul[@class="cmetlist bookreview-cmetlist"]/li/div/div[2]/p')
# 取得所有评论时间
commenttimes = xparser.getlist('//ul[@class="cmetlist bookreview-cmetlist"]/li/div/div[2]/span')
commentsInfo = []
# 取得所有评论
for index in range(0, int(len(comments)), 1):
# 提取时间
publicTime = commenttimes[index][3:]
cmti = CommentInfo()
tm = TimeUtility.getuniformtime(publicTime,'%Y-%m-%d %H:%M')
if URLStorage.storeupdatetime(params.originalurl, tm):
cmti.content = comments[index].strip()
commentsInfo.append(cmti)
# 保存获取的评论
if len(commentsInfo) > 0:
self.commentstorage.store(params.originalurl, commentsInfo)
else:
Logger.getlogging().error('proparam.step == {step}'.format(step = params.step))
except Exception,e:
traceback.print_exc()
def process(self, params):
Logger.getlogging().info(params.url)
try:
if params.step is HongXiuComments.STEP_1:
#Step1: 通过得到docurl，得到获取评论的首页url参数。
bookId = self.r.parse('http://\w+\.hongxiu\.com/\w+/(\d+).*',
params.originalurl)[0]
# 取得评论的url列表
comments_url = HongXiuComments.COMMENT_URL.format(
bookId=bookId, page=1)
self.storeurl(comments_url, params.originalurl,
HongXiuComments.STEP_2, {'bookId': bookId})
elif params.step == HongXiuComments.STEP_2:
# 获得评论参数
bookId = params.customized['bookId']
# 取得总件数
params.content = (params.content).encode('utf-8')
comment_count = self.r.parse(
'strong id="htmlrecordcnt" class="total">(\d+)</strong>条',
params.content)[0]
if comment_count == 0:
return
# 判断增量
cmtnum = URLStorage.getcmtnum(params.originalurl)
if cmtnum >= comment_count:
return
URLStorage.setcmtnum(params.originalurl, comment_count)
# 获取页数
totalPage = int(
math.ceil(
float(comment_count) / HongXiuComments.PAGE_SIZE))
# 获得url列表
for page in range(1, totalPage + 1, 1):
url = HongXiuComments.COMMENT_URL.format(bookId=bookId,
page=page)
self.storeurl(url, params.originalurl,
HongXiuComments.STEP_3)
elif params.step == HongXiuComments.STEP_3:
# Step3: 通过Step2设置的url，得到所有评论，抽取评论
Logger.getlogging().info("params.step == 3")
# 取得所有评论
soup = BeautifulSoup(params.content, 'html5lib')
comments = soup.select('.inner')
# 取得所有评论时间
commenttimes = soup.select('.postTime')
commentsInfo = []
# 取得所有评论
for index in range(0, int(len(comments)), 1):
# 提取时间
publicTime = self.r.parse(
ur'(.*) 发表', commenttimes[index].get_text())[0]
tm = getuniformtime(publicTime)
if URLStorage.storeupdatetime(params.originalurl, tm):
cmti = CommentInfo()
cmti.content = self.r.parse(
ur'发表([\s\S]*)',
comments[index].get_text().strip())[0]
commentsInfo.append(cmti)
# 保存获取的评论
if len(commentsInfo) > 0:
self.commentstorage.store(params.originalurl, commentsInfo)
else:
Logger.getlogging().error(
'proparam.step == {step}'.format(step=params.step))
except Exception, e:
traceback.print_exc()
def process(self, params):
try:
if params.step is BookComments.STEP_1:
#Step1: 通过原始url得到Key，得到获取评论的首页url
urlsplit = params.originalurl.split('/')
if len(urlsplit[-1].strip()) > 0:
key = urlsplit[-1]
else:
key = urlsplit[-2]
field = params.customized['field']
if field == 'manhua':
comments_url = self.MANHUA_COMMENTS_URL.format(key=key,
pg=1)
hxpath = XPathUtility(params.content)
pubTime = hxpath.getstring(
'//*[@class="synopsises_font"]/li[2]/text()', ' ')
if pubTime:
pubTime = pubTime[0]
pubTime = re.findall('\d+/\d+/\d+', params.content)[0]
info = BaseInfoStorage.getbasicinfo(params.originalurl)
info.pubtime = pubTime
BaseInfoStorage.store(params.originalurl, info)
elif field == 'book':
comments_url = self.BOOK_COMMENTS_URL.format(key=key, pg=1)
else:
return
self.storeurl(comments_url, params.originalurl, self.STEP_2, {
'key': key,
'field': field
})
elif params.step == BookComments.STEP_2:
html = etree.HTML(params.content)
comments_total_xpath = html.xpath(
'//*[@class="content_title"]/span/a')
if comments_total_xpath:
comments_total_str = self.r.parse(
u'(\d+)',
comments_total_xpath[0].text.replace(',', ''))
if not comments_total_str:
return
comments_total = int(comments_total_str[0])
cmtnum = URLStorage.getcmtnum(params.originalurl)
if cmtnum >= comments_total:
return
URLStorage.setcmtnum(params.originalurl, comments_total)
# 获取首页评论
self.geturlcomments(params)
if comments_total > self.limit:
page_max = int(
math.ceil(float(comments_total) / self.limit))
# 拼出首页之外的所有评论url
key = params.customized['key']
field = params.customized['field']
if field == 'manhua':
for page in range(2, page_max + 1, 1):
comments_url = self.MANHUA_COMMENTS_URL.format(
key=key, pg=page)
self.storeurl(comments_url, params.originalurl,
self.STEP_3)
elif field == 'book':
for page in range(2, page_max + 1, 1):
comments_url = self.BOOK_COMMENTS_URL.format(
key=key, pg=page)
self.storeurl(comments_url, params.originalurl,
self.STEP_3)
else:
return
elif params.step == BookComments.STEP_3:
# 获取评论
self.geturlcomments(params)
else:
pass
except:
Logger.printexception()
def process(self, params):
try:
if params.step is NewsComments.STEP_1:
#Step1: 通过原始url得到Key，得到获取评论的首页url
urlsplit = params.originalurl.split('/')
field = params.customized['field']
if field == 'news':
fnamesplit = urlsplit[-1].split('.')
key = fnamesplit[0]
comments_url = self.NEWS_COMMENTS_URL.format(key=key, pg=1)
elif field == 'comic':
if len(urlsplit[-1].strip()) > 0:
key = urlsplit[-1]
else:
key = urlsplit[-2]
comments_url = self.COMIC_COMMENTS_UTL.format(key=key, pg=1)
else:
return
self.storeurl(comments_url, params.originalurl, self.STEP_2, {'key': key, 'field': field})
elif params.step == NewsComments.STEP_2:
html = etree.HTML(params.content)
comments_total_xpath = html.xpath('//*[contains(@class,"li_more")]/strong[1]')
if comments_total_xpath:
comments_total = int(comments_total_xpath[0].text)
cmtnum = URLStorage.getcmtnum(params.originalurl)
if cmtnum >= comments_total:
return
URLStorage.setcmtnum(params.originalurl, comments_total)
# 获取首页评论
self.geturlcomments(params)
key = params.customized['key']
field = params.customized['field']
if field == 'news' and comments_total > self.NEWS_LIMIT:
page_max = int(math.ceil(float(comments_total)/self.NEWS_LIMIT))
# 拼出首页之外的所有评论url
for page in range(2, page_max+1, 1):
comments_url = self.NEWS_COMMENTS_URL.format(key=key, pg=page)
self.storeurl(comments_url, params.originalurl, self.STEP_3)
elif field == 'comic' and comments_total > self.COMMIC_LIMIT:
page_max = int(math.ceil(float(comments_total) / self.COMMIC_LIMIT))
# 拼出首页之外的所有评论url
for page in range(2, page_max + 1, 1):
comments_url = self.COMIC_COMMENTS_UTL.format(key=key, pg=page)
self.storeurl(comments_url, params.originalurl, self.STEP_3)
else:
return
elif params.step == NewsComments.STEP_3:
# 获取评论
self.geturlcomments(params)
else:
pass
except:
Logger.printexception()
def process(self, params):
Logger.getlogging().info(params.url)
try:
if params.step is Xie17NewsComments.STEP_1:
#Step1: 通过得到docurl，得到获取评论的首页url参数。
articleId = self.r.parse('^http://xiaoshuo\.17xie\.com/book/(\d+)/', params.originalurl)[0]
# 取得评论的url列表
comments_url = Xie17NewsComments.COMMENT_URL % (articleId, 1)
self.storeurl(comments_url, params.originalurl, Xie17NewsComments.STEP_2, {'articleId': articleId})
elif params.step == Xie17NewsComments.STEP_2:
# 获得评论参数
articleId = params.customized['articleId']
# 取得总件数
comment_count = float(self.r.parse(ur'共(\d+)人说过', params.content)[0])
if comment_count == 0:
return
# 判断增量
cmtnum = URLStorage.getcmtnum(params.originalurl)
if cmtnum >= comment_count:
return
URLStorage.setcmtnum(params.originalurl, comment_count)
# 获取页数
page = int(math.ceil(comment_count / Xie17NewsComments.PAGE_SIZE))
# 获得url列表
for page in range(1, page + 1, 1):
url = Xie17NewsComments.COMMENT_URL % (articleId, page)
self.storeurl(url, params.originalurl, Xie17NewsComments.STEP_3)
elif params.step == Xie17NewsComments.STEP_3:
# Step3: 通过Step2设置的url，得到所有评论，抽取评论
Logger.getlogging().info("params.step == 3")
xparser = XPathUtility(params.content)
# 取得所有评论
comments = xparser.getcomments('/html/body/ul/li[2]/dl/dd')
# 取得所有评论时间
commenttimes = xparser.xpath('/html/body/ul/li[2]/dl/dt/text()')
commentsInfo = []
# 取得所有评论
for index in range(0, int(len(commenttimes)), 1):
# 提取时间
if self.r.search(ur'\d+年\d+月',commenttimes[index].strip()):
tm = TimeUtility.getuniformtime(str(commenttimes[index]).strip(), '%Y年%m月')
else:
tm = getuniformtime(commenttimes[index].strip())
if URLStorage.storeupdatetime(params.originalurl, tm):
cmti = CommentInfo()
comment = comments[index * 3] + comments[index * 3 + 1] + comments[index * 3 + 2]
cmti.content = comment
commentsInfo.append(cmti)
# 保存获取的评论
if len(commentsInfo) > 0:
self.commentstorage.store(params.originalurl, commentsInfo)
else: