使用python-scrapy爬取网页中带有地图展示的数据

34次阅读

共计 5361 个字符,预计需要花费 14 分钟才能阅读完成。

最近有个需求,是要爬取某个物流公司的官网信息,我看了下官网,基本上都是静态页面比较好抓取,不像那种资讯类,电子商务类型的网站结果复杂,反爬严格,AJAX 众多,还内心暗自庆幸,当我进一步分析时候发现并非普通的静态页面。
例如这个 URL 界面,我要获取全中国各大城市的物流园区分布信息,并且要获取详情信息,
这个页面里面是有个地图镶嵌,每个城市物流信息你要单独点击地图上的信息才能显示。
https://www.glprop.com.cn/our…

我刚开始想,这种会不会是 ajax 请求呢,通过 chrmoe 抓包并没有发现,然后我查看网页源代码
发现所有城市信息在一个 scripts 里面
如图:

然后各个园区的信息在一个叫 park={xx}里面存着

原来都在这里面,直接获取源代码,正则匹配,开干。
item:

# 普洛斯
class PuluosiNewsItem(scrapy.Item):
    newstitle=scrapy.Field()
    newtiems=scrapy.Field()
    newslink=scrapy.Field()
class PuluosiItem(scrapy.Item):
    assetstitle = scrapy.Field()
    assetaddress=scrapy.Field()
    assetgaikuang=scrapy.Field()
    assetpeople=scrapy.Field()
    asseturl = scrapy.Field()

pipelines:

class PuluosiNewsPipeline(object):
    def __init__(self):
        self.wb=Workbook()
        self.ws=self.wb.active
        #设置表头
        self.ws.append(['普洛斯新闻标题','新闻发布时间','新闻 URL'])
        self.wb2 = Workbook()
        self.ws2 = self.wb2.active
        self.ws2.append(['资产标题', '资产地址', '资产概况','其他信息','URL'])
    def process_item(self,item,spider):
        if isinstance(item, PuluosiNewsItem):
            line = [item['newstitle'], item['newtiems'], item['newslink']]  # 把数据中每一项整理出来
            self.ws.append(line)
            self.wb.save('PuluosiNews.xlsx')  # 保存 xlsx 文件
        elif isinstance(item,PuluosiItem):
            line = [item['assetstitle'], item['assetaddress'], item['assetgaikuang'],item['assetpeople'],item['asseturl']]
            self.ws2.append(line)
            self.wb2.save('PuluosiAsset.xlsx')  # 保存 xlsx 文件
        return item

spider:

# -*- coding: utf-8 -*-
import scrapy,re,json
from news.items import PuluosiNewsItem,PuluosiItem
from scrapy.linkextractors import LinkExtractor

class PuluosiSpider(scrapy.Spider):
    name = 'puluosi'
    allowed_domains = ['glprop.com.cn']
    # start_urls = ['https://www.glprop.com.cn/press-releases.html']

    def start_requests(self):
        yield scrapy.Request('https://www.glprop.com.cn/press-releases.html', self.parse1)
        yield scrapy.Request('https://www.glprop.com.cn/in-the-news.html', self.parse2)
        yield scrapy.Request('https://www.glprop.com.cn/proposed-privatization.html', self.parse3)
        yield scrapy.Request('https://www.glprop.com.cn/our-network/network-detail.html', self.parse4)

    def parse1(self, response):
        print('此时启动的爬虫为:puluosi')
        item=PuluosiNewsItem()
        web=response.xpath('//tbody/tr')
        web.pop(0)
        for node in  web:
            item['newstitle'] = node.xpath('.//a/text()').extract()[0].strip()
            print(item['newstitle'])
            item['newtiems'] = node.xpath('.//td/text()').extract()[0].strip()
            print(item['newtiems'])
            # urljoin 创建绝对的 links 路径,始用于网页中的 href 值为相对路径的连接
            item['newslink'] = response.urljoin(web.xpath('.//a/@href').extract()[0])
            # print(item['newslink'])
            yield item
        #加入 try 来判断当前年份的新闻是否有下一页出现
        try:
            next_url_tmp = response.xpath('//div[@class="page"]/a[contains(text()," 下一页 ")]/@href').extract()[0]
            if next_url_tmp:
                next_url = "https://www.glprop.com.cn" + next_url_tmp
                yield scrapy.Request(next_url,callback=self.parse1)
        except Exception as e:
            print("当前页面没有下一页")
        href=response.xpath('//ul[@class="timeList"]/li/a/@href')
        for nexturl in href:
            url1 =nexturl.extract()
            if url1:
                url="https://www.glprop.com.cn"+url1
                yield scrapy.Request(url,callback=self.parse1)

    def parse2(self,response):
        item = PuluosiNewsItem()
        web = response.xpath('//tbody/tr')
        web.pop(0)
        for node in  web:
            item['newstitle'] = node.xpath('.//a/text()').extract()[0].strip()
            print(item['newstitle'])
            item['newtiems'] = node.xpath('.//td/text()').extract()[0].strip()
            print(item['newtiems'])
            # urljoin 创建绝对的 links 路径,始用于网页中的 href 值为相对路径的连接
            item['newslink'] = response.urljoin(web.xpath('.//a/@href').extract()[0])
            print(item['newslink'])
            yield item
        #加入 try 来判断当前年份的新闻是否有下一页出现
        try:
            next_url_tmp = response.xpath('//div[@class="page"]/a[contains(text()," 下一页 ")]/@href').extract()[0]
            if next_url_tmp:
                next_url = "https://www.glprop.com.cn" + next_url_tmp
                yield scrapy.Request(next_url,callback=self.parse2)
        except Exception as e:
            print("当前页面没有下一页")
        href=response.xpath('//ul[@class="timeList"]/li/a/@href')
        for nexturl in href:
            url1 =nexturl.extract()
            if url1:
                url="https://www.glprop.com.cn"+url1
                yield scrapy.Request(url,callback=self.parse2)

    def parse3(self,response):
        item=PuluosiNewsItem()
        web=response.xpath('//tbody/tr')
        web.pop()
        for node in  web:
            item['newstitle'] = node.xpath('.//a/text()').extract()[0].strip()
            print(item['newstitle'])
            item['newtiems'] = node.xpath('.//td/text()').extract()[0].strip()
            print(item['newtiems'])
            # urljoin 创建绝对的 links 路径,始用于网页中的 href 值为相对路径的连接
            item['newslink'] = response.urljoin(web.xpath('.//a/@href').extract()[0])
            print(item['newslink'])
            yield item

    def parse4(self,response):
        link=LinkExtractor(restrict_xpaths='//div[@class="net_pop1"]//div[@class="city"]')
        links=link.extract_links(response)
        #获取所有城市的 links
        for i in links:
            detailurl=i.url
            yield scrapy.Request(url=detailurl,callback=self.parse5)

    def parse4(self, response):
        item = PuluosiItem()
        citycode=re.findall('var cities =(.*);',response.text )
        citycodejson=json.loads(("".join(citycode)))
        #把每个城市的 id 和 name 取出来放到一个字典
        dictcity={}
        for i in citycodejson:
            citycodename=i['name']
            citycodenm=i['id']
            dictcity[citycodenm]=citycodename
        detail=re.findall('var parks =(.*);',response.text )
        jsonBody = json.loads(("".join(detail)))
        list = []
        for key1 in jsonBody:
            for key2  in jsonBody[key1]:
                tmp=jsonBody[key1][key2]
                list.append(jsonBody[key1][key2])
        for node in list:
            assetaddress = node['city_id']
            item['assetaddress'] = dictcity[assetaddress]
            # print(item['assetaddress'])
            item['assetstitle'] = node['name']
            # print(item['assetstitle'])
            item['assetgaikuang'] = node['detail_single'].strip().replace(' ', '').replace(' ','')
            # print(item['assetgaikuang'])
            assetpeople = node['description']
            item['assetpeople'] = re.sub(r'<.*?>', '', (assetpeople.strip())).replace('&nbsp;','')
            item['asseturl']='https://www.glprop.com.cn/network-city-detail.html?city='+item['assetaddress']
            # print(item['assetpeople'])
            yield item

然后我顺便把页面的新闻信息也爬取了。

正文完
 0