1、默认请求头

USER_AGENT = "Scrapy/1.8.0 (+http://scrapy.org)"

2、修改请求头

  • 全局设置

所有爬虫的所有连接生效。

settings.py

USER_AGENT = "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/100.0.0.0 Safari/537.36"

  • 爬虫设置

单个爬虫所有连接生效。

spider.py

class MySpider(scrapy.Spider):
    name = "myspider"
    custom_settings = {
            "USER_AGENT": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/100.0.0.0 Safari/537.36",
    }

  • 链接设置

单个请求的单个链接生效。

spider.py

USER_AGENT = "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/100.0.0.0 Safari/537.36"


class MySpider(scrapy.Spider):
    name = "myspider"
    start_urls = ("https://baidu.com",)

    def start_requests(self):
        for url in self.start_urls:
            yield scrapy.Request(url, headers={"User-Agent": USER_AGENT})

  • 中间件设置

从整个项目中去修改请求头的设置规则，变化多端，不同的写法，可以配置出不同的设置方式。

settings.py

"DOWNLOADER_MIDDLEWARES": {
        "myspider.middlewares.UserAgentMiddleware": 544
}

middlewares.py

class UserAgentMiddleware(object):
    def process_request(self, request, spider):
        request.headers["User-Agent"] = ""

3、优先级

中间件headers > 爬虫的custom_settings > 项目的settings.py > Scrapy默认