scrapy, splash, lua, 按钮点击

Posted 2023-03-28

【中文标题】scrapy, splash, lua, 按钮点击

【英文标题】：scrapy, splash, lua, button click

【发布时间】：2018-04-17 15:18:45

【问题描述】：

我对这里的所有乐器都很陌生。我的目标是从许多通过“Weiter”/“next”按钮连接的页面中提取所有 URL - 对于几个 URL。我决定用scrapy试试。页面是动态生成的。然后我了解到我需要一个进一步的仪器并为此安装了 Splash。安装正在运行。我根据教程设置了安装。然后我设法通过在搜索输入字段中发送“返回”来获取第一页。使用可以为我提供所需结果的浏览器。我的问题是我尝试单击生成页面上的“下一步”按钮，但不知道具体如何。正如我在几页上所读到的，这并不总是那么容易。我尝试了建议的解决方案但没有成功。我想我并不太远，希望能得到一些帮助。谢谢你。

我的设置.py

BOT_NAME = 'gr'
SPIDER_MODULES = ['gr.spiders']
NEWSPIDER_MODULE = 'gr.spiders'
ROBOTSTXT_OBEY = True
DOWNLOADER_MIDDLEWARES = 
    'scrapy_splash.SplashCookiesMiddleware': 723,
    'scrapy_splash.SplashMiddleware': 725,
    'scrapy.downloadermiddlewares.httpcompression.HttpCompressionMiddleware': 810,

SPLASH_URL = 'http://localhost:8050'
SPIDER_MIDDLEWARES = 
    'scrapy_splash.SplashDeduplicateArgsMiddleware': 100,

DUPEFILTER_CLASS = 'scrapy_splash.SplashAwareDupeFilter'
HTTPCACHE_STORAGE = 'scrapy_splash.SplashAwareFSCacheStorage'

我的蜘蛛：

# -*- coding: utf-8 -*-
import scrapy
from scrapy.linkextractors import LinkExtractor
from scrapy_splash import SplashRequest
import json
# import base64

class GrSpider(scrapy.Spider):
    name = 'gr_'
    allowed_domains = ['lawsearch.gr.ch']
    start_urls = ['http://www.lawsearch.gr.ch/le/']

    def start_requests(self):

        script = """
        function main(splash)
            assert(splash:go(splash.args.url))
            splash:set_viewport_full()
            splash:wait(0.3)
            splash:send_keys("<Return>")
            splash:wait(0.3)
            return splash:html()
        end
        """

        for url in self.start_urls:

             yield SplashRequest(url=url,
                    callback=self.parse,
                    endpoint='execute',
                    args='lua_source': script)

    def parse(self, response):

        script3 = """
            function main(splash)
            splash:autoloadurl="https://code.jquery.com/jquery-3.2.1.min.js"
            assert(splash:go(splash.args.url))
            splash:set_viewport_full()

--            splash:wait(2.8)
--            local element = splash:select('.result-pager-next-active .simplebutton')
--            element:mouse_click()

--            local bounds = element:bounds()
--            assert(element:mouse_clickx=bounds.width, y=bounds.height)

--            naechster VERSCUH
-- https://blog.scrapinghub.com/2015/03/02/handling-javascript-in-scrapy-with-splash/

--  https://***.com/questions/35720323/scrapyjs-splash-click-controller-button
--  assert(splash:runjs("$('#date-controller > a:first-child').click()"))

--  https://github.com/scrapy-plugins/scrapy-splash/issues/27
--            assert(splash:runjs("$('#result-pager-next-active .simplebutton').click()"))



-- https://developer.mozilla.org/de/docs/Web/API/Document/querySelectorAll
-- ANSCHAUEN
-- https://***.com/questions/38043672/splash-lua-script-to-do-multiple-clicks-and-visits

-- elementList = baseElement.querySelectorAll(selectors)
-- var domRect = element.getBoundingClientRect();
-- var rect = obj.getBoundingClientRect();
-- https://***.com/questions/34001917/queryselectorall-with-multiple-conditions

            local get_dimensions = splash:jsfunc([[
               function () 
                 var doc1 = document.querySelectorAll("result-pager-next-active.simplebutton")[0];
                 var el = doc1.documentElement;
                 var rect = el.getClientRects()[0];
                 return 'x': rect.left, 'y': rect.top
               
            ]])
--            splash:set_viewport_full()
            splash:wait(0.1)
            local dimensions = get_dimensions()
            splash:mouse_click(dimensions.x, dimensions.y)


--            splash:runjs("document.querySelectorAll('result-pager-next-active ,simplebutton')[1].click()")
--            assert(splash:runjs("$('.result-pager-next-active .simplebutton')[1].click()"))
--            assert(splash:runjs("$('.simplebutton')[12].click()"))

            splash:wait(1.6)
            return splash:html()
        end
        """

        for teil in response.xpath('//div/div/div/div/a'):
            yield 
                'link': teil.xpath('./@href').extract()
            

        next_page = response.xpath('//div[@class="v-label v-widget simplebutton v-label-simplebutton v-label-undef-w"]').extract_first()

#        print response.body

        print '----------------------'        
#        print  response.xpath('//div[@class="v-slot v-slot-simplebutton"]/div[contains(text(), "Weiter")]').extract_first()
#        print  response.xpath('//div[@class="v-slot v-slot-simplebutton"]/div[contains(text(), "Weiter")]').extract()

#       class="v-slot v-slot-simplebutton"
#        nextPage = HtmlXPathSelector(response).select("//div[@class='paginationControl']/a[contains(text(),'Link Text Next')]/@href")

#        neue_seite=response.url

#        print response.url

        if next_page is not None: 
#            yield SplashRequest(url=neue_seite,

            yield SplashRequest(response.url,
                    callback=self.parse,
                    dont_filter=True,
                    endpoint='execute',
                    args='lua_source': script3)

【问题讨论】：

如果您找到了方法，请添加答案。

“页面是动态生成的。然后我了解到我需要更多的仪器并为此安装了 Splash。”这实际上是一个普遍的误解。见docs.scrapy.org/en/latest/topics/dynamic-content.html

基本上我改变了这个概念，现在使用它工作的硒 - 正如之前评论中所建议的那样。

【参考方案1】：

您不必总是使用 Splash。如果下一个按钮是一个链接，您只需获取 href 属性并将请求发送回解析函数。