python - 使用 Scrapy 抓取没有“下一步按钮”的表数据

Question

我是新手Scrapy，我尝试从该网站的每个页面获取表格数据。

这是我的代码：

import scrapy

class UAESpider(scrapy.Spider):
    name = 'uae_free'

    allowed_domains = ['https://www.uaeonlinedirectory.com']

    start_urls = [
        'https://www.uaeonlinedirectory.com/UFZOnlineDirectory.aspx?item=A'
    ]
    
    def parse(self, response):
        pages = response.xpath('//table[@class="GridViewStyle"]//tr[12]')

        for page in pages[1:11]:
            rows = page.xpath('//table[@class="GridViewStyle"]//tr')
            for row in rows[1:11]:
                yield {
                    'company_name': row.xpath('.//td[2]//text()').get(),
                    'company_name_link': row.xpath('.//td[2]//a/@href').get(),
                    'zone': row.xpath('.//td[4]//text()').get(),
                    'category': row.xpath('.//td[6]//text()').get(),
                    'category_link': row.xpath('.//td[6]//a/@href').get()
                }

        next_page = response.xpath('//table[@class="GridViewStyle"]//tr[12]//td[11]//a/@href').get()

        if next_page:
            yield scrapy.Request(url=next_page, callback=self.parse)

但它不起作用，我收到这个错误，下面的 URL 是链接page 11：

ValueError: Missing scheme in request url: javascript:__doPostBack('ctl00$ContentPlaceHolder2$grdDirectory','Page$11')

大佬们知道怎么修复bug吗？

更新：

按照@zmike建议的this answer中的说明进行操作，这是我到目前为止所做的：

import scrapy
from scrapy.http import FormRequest

URL = 'https://www.uaeonlinedirectory.com/UFZOnlineDirectory.aspx?item=A'

class UAESpider(scrapy.Spider):
    name = 'uae_free'

    allowed_domains = ['https://www.uaeonlinedirectory.com/UFZOnlineDirectory.aspx?item=A']

    start_urls = [
        'https://www.uaeonlinedirectory.com/UFZOnlineDirectory.aspx?item=A'
    ]

    def parse(self, response):
        self.data = {}

        for form_input in response.css('form#aspnetForm input'):
            name = form_input.xpath('@name').extract()[0]
            try:
                value = form_input.xpath('@value').extract()[0]
            except IndexError:
                value = ""
            self.data[name] = value

        self.data['ctl00_ContentPlaceHolder2_panelGrid'] = 'ctl00$ContentPlaceHolder2$grdDirectory'
        self.data['__EVENTTARGET'] = 'ctl00$ContentPlaceHolder2$grdDirectory'
        self.data['__EVENTARGUMENT'] = 'Page$1'

        return FormRequest(url=URL,
                            method='POST',
                            callback=self.parse_page,
                            formdata=self.data,
                            meta={'page':1},
                            dont_filter=True)

    def parse_page(self, response):
        current_page = response.meta['page'] + 1
        rows = response.xpath('//table[@class="GridViewStyle"]//tr')
        for row in rows[1:11]:
            yield {
                'company_name': row.xpath('.//td[2]//text()').get(),
                'company_name_link': row.xpath('.//td[2]//a/@href').get(),
                'zone': row.xpath('.//td[4]//text()').get(),
                'category': row.xpath('.//td[6]//text()').get(),
                'category_link': row.xpath('.//td[6]//a/@href').get()
            }

        return FormRequest(url=URL,
                            method='POST',
                            formdata={
                                '__EVENTARGUMENT': 'Page$%d' % current_page,
                                '__EVENTTARGET': 'ctl00$ContentPlaceHolder2$grdDirectory',
                                'ctl00_ContentPlaceHolder2_panelGrid':'ctl00$ContentPlaceHolder2$grdDirectory',
                                '':''},
                            meta={'page': current_page},
                           dont_filter=True)

并且此代码仅从第一页获取表格数据，不会移动到剩余页面。你知道我哪里做错了吗？

标签： pythonweb-scrapingscrapy

Answer 1

这是遍历所有页面的爬虫的工作（尽管很粗糙）实现。一些注意事项：

• 表单数据需要不同的参数，例如__EVENTTARGET、__EVENTVALIDATION、__VIEWSTATEGENERATOR等。
  • 我使用 XPath 而不是正则表达式来获取它们
• 以下是不必要的：self.data['ctl00_ContentPlaceHolder2_panelGrid'] = 'ctl00$ContentPlaceHolder2$grdDirectory'
• 为了简单起见，我组合了这些功能。回调允许它遍历所有页面。

import scrapy
from scrapy.http import FormRequest


class UAESpider(scrapy.Spider):
    name = 'uae_free'
    headers = {
        'X-MicrosoftAjax': 'Delta=true',
        'User-Agent': 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_10_2) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/41.0.2272.76 Safari/537.36'
    }

    allowed_domains = ['www.uaeonlinedirectory.com']
    # TODO: Include the urls for all other items (e.g. A-Z)
    start_urls = ['https://www.uaeonlinedirectory.com/UFZOnlineDirectory.aspx?item=A']
    current_page = 0

    def parse(self, response):
        # request the next page
        self.current_page = self.current_page + 1

        if self.current_page == 1:
            # submit a form (first page)
            data = {}
            for form_input in response.css('form#aspnetForm input'):
                name = form_input.xpath('@name').extract()[0]
                try:
                    value = form_input.xpath('@value').extract()[0]
                except IndexError:
                    value = ""
                data[name] = value
            data['__EVENTTARGET'] = 'ctl00$MainContent$List'
            data['__EVENTARGUMENT'] = 'Page$1'
        else:
            # Extract the form fields and arguments using XPATH
            event_validation = response.xpath('//input[@id="__EVENTVALIDATION"]/@value').extract()
            view_state = response.xpath('//input[@id="__VIEWSTATE"]/@value').extract()
            view_state_generator = response.xpath('//input[@id="__VIEWSTATEGENERATOR"]/@value').extract()
            view_state_encrypted = response.xpath('//input[@id="__VIEWSTATEENCRYPTED"]/@value').extract()

            data = {
                '__EVENTTARGET': 'ctl00$ContentPlaceHolder2$grdDirectory',
                '__EVENTARGUMENT': 'Page$%d' % self.current_page,
                '__EVENTVALIDATION': event_validation,
                '__VIEWSTATE': view_state,
                '__VIEWSTATEGENERATOR': view_state_generator,
                '__VIEWSTATEENCRYPTED': view_state_encrypted,
                '__ASYNCPOST': 'true',
                '': ''
            }

        # Yield the companies
        # TODO: move this to a different function
        rows = response.xpath('//table[@class="GridViewStyle"]//tr')
        for row in rows[1:11]:
            result = {
                'company_name': row.xpath('.//td[2]//text()').get(),
                'company_name_link': row.xpath('.//td[2]//a/@href').get(),
                'zone': row.xpath('.//td[4]//text()').get(),
                'category': row.xpath('.//td[6]//text()').get(),
                'category_link': row.xpath('.//td[6]//a/@href').get()
            }
            print(result)
            yield result

        # TODO: check if there is a next page, and only yield if there is one
        yield FormRequest(url=self.start_urls[0],  # TODO: change this so that index is not hardcoded
                          method='POST',
                          formdata=data,
                          callback=self.parse,
                          meta={'page': self.current_page},
                          dont_filter=True,
                          headers=self.headers)