Splash对接Scrapy

Posted 2021-11-26 lokvahkoor

1. 安装Splash：

1. 安装docker
2. docker拉取Splash镜像：docker pull scrapinghub/splash
3. 启动Splash：docker run -p 8050:8050 scrapinghub/splash

2. 安装并配置scrapy-splash：https://github.com/scrapy-plugins/scrapy-splash

在settings.py中增加：

SPLASH_URL = 'http://Splash的部署地址:8050/'
DOWNLOADER_MIDDLEWARES = {
    'scrapy_splash.SplashCookiesMiddleware': 723,
    'scrapy_splash.SplashMiddleware': 725,
    'scrapy.downloadermiddlewares.httpcompression.HttpCompressionMiddleware': 810,
}
SPIDER_MIDDLEWARES = {
    'scrapy_splash.SplashDeduplicateArgsMiddleware': 100,
}
DUPEFILTER_CLASS = 'scrapy_splash.SplashAwareDupeFilter'
HTTPCACHE_STORAGE = 'scrapy_splash.SplashAwareFSCacheStorage'

3. Spider代码

用SplashRequest代替scrapy.Request：

yield SplashRequest(url, self.parse_result,  # 第一个参数是要请求的url，第二个参数是回调函数
    args={  # 常用于传递Lua脚本
        "lua_source": """
                    splash:set_user_agent("...")
                    assert(splash:go(args.url))
                    assert(splash:wait(args.time))  --注意脚本中可以用args.*来接收外界参数
                    return {html = splash:html()}  --返回html
                    """
        “time”: time  # lua脚本参数从这里传入
    },
    endpoint='run', # 默认为render.html，常用的是’execute’ 和 ‘run’（一般用run），用来执行脚本(run和execute的差别在于：run的脚本只需要包含main函数里的内容就可以了，就像上面的示例代码一样)
)

完整Spider示例：

import scrapy
from scrapy_splash import SplashRequest


class ExampleSpider(scrapy.Spider):
    name = 'connect_splash'
    
    def start_requests(self):
        url = 'http://www.baidu.com'
        script = """
        assert(splash:go(args.url))
        assert(splash:wait(args.wait))
        return {html = splash:html()}
        """
        yield SplashRequest(url, self.parse, endpoint='run', args={'lua_source':script,'wait':3})
        
    def parse(self, response):
        from scrapy.shell import inspect_response
        inspect_response(response, self)
        pass