def _extractAnswer(self, block):
# aid
aid = block['data-aid']
# 回答人
responderBlock = block.find('a', class_='zm-item-link-avatar')
if responderBlock is None:
responder = -1 # 匿名用户
else:
responder = responderBlock['href'][8:] # /people/<responder>
# 日期
date = block['data-created']
# 内容
content = block.find('div', class_='zm-editable-content').text.strip()
# 赞同数
upvote = parseNum(block.find('span', class_='count').text)
# 评论数目
comments = block.find('a', class_='toggle-comment').text.strip()
p = comments.find('条评论')
if p > 0:
commentsCount = int(comments[:p])
else:
commentsCount = 0
answer = dict()
answer['aid'] = aid
answer['responder'] = responder
answer['date'] = date
answer['content'] = content
answer['upvote'] = upvote
answer['commentsCount'] = commentsCount
return answer
def update(self):
'''
更新Question，并获取Answers
'''
self.lastModified = str(datetime.datetime.now())
qurl = 'http://www.zhihu.com/question/%d' % (self.qid)
r = get(qurl)
if r.status_code != 200:
return False
soup = BeautifulSoup(r.text)
# 标题
self.title = soup.find('h2', class_='zm-item-title').text.strip()
# 内容
self.detail = soup.find('div', id='zh-question-detail').div.text.strip()
# 所属的话题标签
self.tags = [a.string.strip() for a in soup.find_all("a", class_='zm-item-tag')]
# 关注人数
followersCountBlock = soup.find('div', class_='zg-gray-normal')
if followersCountBlock is None or followersCountBlock.strong is None:
# 当”还没有人关注该问题” followersCountBlock.strong is None
self.followersCount = 0
else:
self.followersCount = parseNum(followersCountBlock.strong.text)
self.answers = []
# 回答数目
answersCountBlock = soup.find('h3', id='zh-question-answer-num')
if answersCountBlock is None:
if soup.find('span', class_='count') is not None:
answersCount = 1
else:
answersCount = 0
else:
answersCount = int(answersCountBlock['data-num'])
# 答案部分 每次50个
for block in soup.find_all('div', class_='zm-item-answer'):
if block.find('div', class_='answer-status') is not None:
continue # 忽略建议修改的答案
self.answers.append(self._extractAnswer(block))
if answersCount > 50:
_xsrf = soup.find('input', attrs={'name': '_xsrf'})['value']
otherHeaders = {'Referer': qurl}
for i in range(1, math.ceil(answersCount/50)): # more answers
data = {"_xsrf": _xsrf, "method": 'next', 'params': '{"url_token": %d, "pagesize": 50, "offset": %d}' % (self.qid, i*50)}
r = post('http://www.zhihu.com/node/QuestionAnswerListV2', otherHeaders, data)
for block in r.json()['msg']:
div = BeautifulSoup(block).div
if div.find('div', class_='answer-status') is not None:
continue # 忽略建议修改的答案
self.answers.append(self._extractAnswer(div))
return True