pconline.py 文件源码

def gen_item_comment(self, response, is_first=False):
        comment = []
        new_comment = {}
        comments_data = []
        rep_time_list = response.xpath('//span[@class="date"]/text()').extract()
        for indexi, content in enumerate(response.xpath('//div[@class="replycontent"]').extract()):
            soup = BeautifulSoup(content, 'lxml')
            [s.extract() for s in soup('script')]  # remove script tag
            c = StrClean.clean_comment(soup.get_text())
            time_index = indexi
            if is_first:
                time_index += 1
            if time_index >= len(rep_time_list):
                rep_time = self.format_rep_date(rep_time_list[-1])
            else:
                rep_time = self.format_rep_date(rep_time_list[time_index])
            comments_data.append({'content': c, 'reply_time': rep_time})
        new_comment['url'] = response.url
        new_comment['comments_data'] = comments_data
        comment.append(new_comment)
        return comment