Question

使用XPath从表中选择使用Scrapy的HTML元素时遇到问题。我使用的例子是Scrapy网站的基本示例：http://doc.scrapy.org/en/latest/intro/tutorial.html我要解析的网站是http://www.euroleague.net/main/results/showgame?gamecode=5&gamenumber=1&phasetypecode=RS&seasoncode=E2013#!playbyplay

起初我使用了这段代码：

from basketbase.items import BasketbaseItem
from scrapy.contrib.spiders import CrawlSpider, Rule
from scrapy.contrib.linkextractors.sgml import SgmlLinkExtractor
from scrapy.selector import HtmlXPathSelector
from scrapy.http import HtmlResponse


class Basketspider(CrawlSpider):
    name = "playbyplay"
    download_delay = 0.5

    allowed_domains = ["www.euroleague.net"]
    start_urls = ["http://www.euroleague.net/main/results/showgame?gamenumber=1&phasetypecode=RS&gamecode=4&seasoncode=E2013"]
    rules = (
        Rule(SgmlLinkExtractor(allow=(),),callback='parse_item',),        
    )  


    def parse(self,response):
        response = HtmlResponse(url=response.url, status=response.status, headers=response.headers, body=response.body)
        return super(Basketspider,self).parse(response)

    def parse_item(self, response):
        response = HtmlResponse(url=response.url, status=response.status, headers=response.headers, body=response.body)
        sel = HtmlXPathSelector(response)

        items=[]
        item = BasketbaseItem()         
        item['game_time'] = sel.select('//div[@style="overflow: auto; height: 250px; width: 800px;"]/table/tbody/tr/td[1]/text()').extract() #
        item['game_event'] = sel.select('//div[@style="overflow: auto; height: 250px; width: 800px;"]/table/tbody/tr/td[2]/text()').extract() #
        item['game_event_res_home'] = sel.select('//div[@style="overflow: auto; height: 250px; width: 800px;"]/table/tbody/tr/td[3]/text()').extract() #
        item['game_event_res_visitor'] = sel.select('//div[@style="overflow: auto; height: 250px; width: 800px;"]/table/tbody/tr/td[3]/text()').extract() #
        item['game_event_team'] = sel.select('//div[@style="overflow: auto; height: 250px; width: 800px;"]/table/tbody/tr/td[4]/text()').extract() #
        item['game_event_player'] = sel.select('//div[@style="overflow: auto; height: 250px; width: 800px;"]/table/tbody/tr/td[5]/text()').extract() #          
        items.append(item)



        return items

这是基本的，规则在这个时候不是很正确，但这个例子的主要关注点是XPath。

它有效，但不是我想要的方式。 我希望每个项目只提取一个td / tr的值，但是使用此代码，它会立即将所有td元素提取到项目中。项目game_event_res_visitor：

'game_event_res_visitor': [u'0-0',
                           u'0-0',
                           u'0-0',.......(list goes on and on)

为了获得我想要的结果，我决定使用循环（就像在Scrapy教程（http://doc.scrapy.org/en/latest/intro/tutorial.html）中），但它根本不返回任何值。这是代码：

def parse(self,response):
    response = HtmlResponse(url=response.url, status=response.status, headers=response.headers, body=response.body)
    return super(Basketspider,self).parse(response)

def parse_item(self, response):
    response = HtmlResponse(url=response.url, status=response.status, headers=response.headers, body=response.body)
    sel = HtmlXPathSelector(response)
    sites = sel.xpath('//div[@style="overflow: auto; height: 250px; width: 800px;"]/table/tbody/tr')        
    items=[]
    item = BasketbaseItem()
    for site in sites:

        item = BasketbaseItem()
        item['game_time'] = sel.select('td[1]/text()').extract() #
        item['game_event'] = sel.select('td[2]/text()').extract() #
        item['game_event_res_home'] = sel.select('td[3]/text()').extract() #
        item['game_event_res_visitor'] = sel.select('td[3]/text()').extract() #
        item['game_event_team'] = sel.select('td[4]/text()').extract() #
        item['game_event_player'] = sel.select('td[5]/text()').extract() #          
        items.append(item)



    return items

和终端输出：

2014-03-07 16:57:45+0200 [playbyplay] DEBUG: Scraped from <200 http://www.euroleague.net/main/results/showgame?gamecode=9&gamenumber=1&phasetypecode=RS&seasoncode=E2013>
    {'game_event': [],
     'game_event_player': [],
     'game_event_res_home': [],
     'game_event_res_visitor': [],
     'game_event_team': [],
     'game_time': []}
2014-03-07 16:57:45+0200 [playbyplay] DEBUG: Scraped from <200 http://www.euroleague.net/main/results/showgame?gamecode=9&gamenumber=1&phasetypecode=RS&seasoncode=E2013>
    {'game_event': [],
     'game_event_player': [],
     'game_event_res_home': [],
     'game_event_res_visitor': [],
     'game_event_team': [],
     'game_time': []}

我明白我的XPath有问题，但我不明白。如果我在item元素中使用相对XPath，它给出的结果与第一个示例中的结果相同。所以它就在那里，但我无法用我拥有的代码达到它。我甚至试过“外卡”。

    item['game_time'] = sel.select('*/text()').extract() #
    item['game_event'] = sel.select('*/text()').extract() #
    item['game_event_res_home'] = sel.select('*/text()').extract() #
    item['game_event_res_visitor'] = sel.select('*/text()').extract() #
    item['game_event_team'] = sel.select('*/text()').extract() #
    item['game_event_player'] = sel.select('*/text()').extract() #

未能获得任何文字结果。

2014-03-07 19:11:14+0200 [playbyplay] DEBUG: Scraped from <200 http://www.euroleague.net/main/results/showgame?gamecode=7&gamenumber=1&phasetypecode=RS&seasoncode=E2013>
    {'game_event': [u' \r\n', u'\r\n'],
     'game_event_player': [u' \r\n', u'\r\n'],
     'game_event_res_home': [u' \r\n', u'\r\n'],
     'game_event_res_visitor': [u' \r\n', u'\r\n'],
     'game_event_team': [u' \r\n', u'\r\n'],
     'game_time': [u' \r\n', u'\r\n']}

我很困惑，我不明白我的XPath或我的代码有什么问题。

Answer 1

这对我有用：

def parse_item(self, response):
    response = HtmlResponse(url=response.url, status=response.status, headers=response.headers, body=response.body)
    sel = HtmlXPathSelector(response)

    rows = sel.select('//div[@style="overflow: auto; height: 250px; width: 800px;"]/table/tbody/tr')
    for row in rows:
        item = BasketbaseItem()
        item['game_time'] = row.select("td[1]/text()").extract()[0]
        item['game_event'] = row.select("td[2]/text()").extract()[0]
        result = row.select("td[3]/text()").extract()[0]
        item['game_event_res_home'], item['game_event_res_visitor'] = result.split('-')
        item['game_event_team'] = row.select("td[4]/text()").extract()[0]
        item['game_event_player'] = row.select("td[5]/text()").extract()[0]
        yield item

这是我得到的一个示例项目：

{'game_event': u'Steal',
 'game_event_player': u'DJEDOVIC, NIHAD',
 'game_event_res_home': u'0 ',
 'game_event_res_visitor': u' 0',
 'game_event_team': u'FC Bayern Munich',
 'game_time': u'2'}

对于你来说，这只是一个开始 - 有时由于IndexError异常而没有产生项目 - 正确处理它。

希望有所帮助。

XPath没有指向带有迭代器的正确HTML表格元素（Scrapy）

1 个答案: