scrapy,我现在把文章列表的文章 url 和 title 爬到了，但是怎么爬到文章内容呢？

[Ixizi]

URL 已经有了，但是 scrapy 里面该怎么操作爬文章内容啊。

class BlogSpider(CrawlSpider):
    name = 'bee'
    allowed_domains = ['xxx.com']
    start_urls = ['https://xxxxxx.com/blogs?']
    rules = (
        Rule(LinkExtractor(allow="page=[0-9]{1,20}"),follow=True,callback='parse_item'),
    )
    def parse_item(self,response):
        item = ScrapyBlogItem()
        title = response.xpath('//h2[@class="title"]/a/text()').extract()
        url = response.xpath('//h2[@class="title"]/a/@href').extract()
        item_dict = dict(zip(url,title))
        item['page'] = item_dict
        # yield item
        return item

[naomhan]

还是一样的啊 通过 css 选择权或者 xpath 取到正文所在区域 再去除干扰数据抽取正文

[icybee]

楼上说的对

[ChiChou]

跟进每一个链接再抓全部的内容。下面是爬乌云漏洞的例子，仅供参考

```python
import re

from scrapy.spiders import CrawlSpider, Rule
from scrapy.linkextractors import LinkExtractor
from scrapy.http import Request

RE_VULID = r'\-(\d+\-\d+)'


class WooyunSpider(CrawlSpider):
name = 'wooyun'
allowed_domains = ['www.wooyun.org']
start_urls = ['http://www.wooyun.org/bugs/new_public/page/1']
rules = (
Rule(LinkExtractor(allow=(r'\/bugs\/new_public\/page\/\d+', ), )),
Rule(LinkExtractor(allow=(r'\/bugs\/wooyun\-\d+\-\d+', )), callback='parse_vul'),
)


def __init__(self, *args, **kwarg):
super(WooyunSpider, self).__init__(*args, **kwarg)
self.finished = set()


def make_requests_from_url(self, url):
match = re.findall(RE_VULID, url)
if match:
vulid, = match
if vulid in self.finished:
return
else:
self.finished.add(vulid)

return Request(url, dont_filter=False)


def parse_vul(self, response):
item = {key: ''.join([text.strip() for text in extracted]) for key, extracted in {
'title': response.css('h3.wybug_title::text').re(ur'\t\t(\S+)'),
'vulid': response.xpath('//h3/a[starts-with(@href,"/bugs/wooyun")]/@href').re(RE_VULID),
'vendor': response.css('h3.wybug_corp a::text').extract(),
'author': response.css('h3.wybug_author a::text').extract(),
'submitted': response.css('h3.wybug_date::text').re('\t\t(\d+\-\d+\-\d+\s+\d+:\d+)'),
'published': response.css('h3.wybug_open_date::text').re('\t\t(\d+\-\d+\-\d+\s+\d+:\d+)'),
'detail': response.css('.wybug_detail').xpath('./node()').extract(),
'patch': response.css('.wybug_patch .detail').xpath('./node()').extract(),
'rank': response.css('.bug_result .detail').re(r'Rank[\s\S]?(\d*)'),
'description': response.css('p.detail.wybug_description::text').extract(),
'vultype': response.css('h3.wybug_type::text').re('\t\t(\S+)'),
'level': response.css('.bug_result .detailTitle + p.detail::text').re(ur'\uff1a(\S+)'),
}.iteritems()}

yield item
```

[ChiChou]

所以特么怎么在回复里贴代码？

[v2014]

楼上都没理解楼主的意思，他的意思是取到了 url 和 title ，但是内容不在这个页面，怎么去爬内容。
只要在 return 一个 Request(url, callback=self.parse)，后面带个 callback 处理你的内容页就可以了，
什么时候 return ，就看你是 bfs 还是 dfs 了

[bdbai]

@ChiChou 贴 GitHub Gist 地址就可以，或者把空格转成全角空格。

[ChiChou]

@bdbai 并不想发 gist

[ChiChou]

@v2014 不用扯上我啊，我可是跟题主一模一样的场景

[Ixizi]

@ChiChou 淡定....

[ChiChou]

@Ixizi pastebin: http://pastebin.com/stRTs7Yj

[v2014]

@ChiChou 谁叫你打字比我快

[Ixizi]

@v2014
@ChiChou

我我我我我我我写出来了。。。代码应该还能优化，求指教

[ChiChou]

@v2014 怪我咯