修复 FinNLP 爬虫bug

asmcos

在使用 FinNLP 提供例子去获取

Easymoney

from finnlp.data_sources.news.eastmoney_streaming import Eastmoney_Streaming
东方财富论坛的数据的时候，
由于 xpath发生了变化。也就是网友html内部变动了。导致线上代码，不能获取数据。
现在已经修改，提交了 issues

里面包含了正确的代码：
https://github.com/AI4Finance-Foundation/FinNLP/issues/3

 def _gather_pages(self, stock, page):
     ....
     # gather the comtent of the first page
        page = etree.HTML(response.text)
        trs = page.xpath('//*[@id="mainlist"]/div/ul/li[1]/table/tbody/tr')
        have_one = False
        for item in trs:
            have_one = True
            read_amount = item.xpath("./td[1]//text()")[0]
            comments = item.xpath("./td[2]//text()")[0]
            title = item.xpath("./td[3]/div/a//text()")[0]
            content_link = item.xpath("./td[3]/div/a/@href")[0]
            author = item.xpath("./td[4]//text()")[0]
            time = item.xpath("./td[5]//text()")[0]
            tmp = pd.DataFrame([read_amount, comments, title, content_link, author, time]).T
            columns = [ "read amount", "comments", "title", "content link", "author", "create time" ]
            tmp.columns = columns
            self.dataframe = pd.concat([self.dataframe, tmp])
            #print(title)
        if have_one == False:
            return "break"
   ...