如何使用 Scrapy 从 Google 新闻网页中获取标题

Question

我保存了一个离线文件 https://news.google.com/search?q=amazon&hl=en-US&gl=US&ceid=US%3Aen

难以确定如何抓取所列文章的标题。

import scrapy
class newsSpider(scrapy.Spider):
    name = "news"
    start_urls = ['file:///127.0.0.1/home/toni/Desktop/crawldeez/googlenewsoffline.html/'
                  ]
    def parse(self, response):
        for xrnccd in response.css('a.MQsxIb.xTewfe.R7GTQ.keNKEd.j7vNaf.Cc0Z5d.EjqUne'):
            yield {
                'ipQwMb.ekueJc.RD0gLb': xrnccd.css('h3.ipQwMb.ekueJc.RD0gLb::ipQwMb.ekueJc.RD0gLb').get(),
            }

Answer 1

问题似乎在于页面内容是使用 JavaScript 动态呈现的，因此无法使用css或xpath方法从 HTML 中提取。但是，它存在于响应正文中，因此您可以使用正则表达式提取它。以下是 Scrapy shell 会话，用于演示如何：

$ scrapy shell "https://news.google.com/search?q=amazon&hl=en-US&gl=US&ceid=US%3Aen"
...
>>> import re
>>> from pprint import pprint
>>>
>>> titles = re.findall(r'<h3 class="[^"]+?"><a[^>]+?>(.+?)</a>', response.text)
>>> pprint(titles)
['Amazon will no longer sell Chinese goods in China',
 'YouTube is finally coming back to Amazon’s Fire TV devices',
 'Amazon Plans to Use Digital Media to Expand Its Advertising Business',
 'Amazon flooded with fake reviews; Learn how to spot them',
 'How To Win in Today&#39;s Amazon World',
 'Amazon Day: How to schedule Amazon deliveries',
 'Bezos Disputes Amazon’s Market Power. But His Merchants Feel the Pinch',
 '20 Best Action Movies to Stream on Amazon Prime',
 ...]