scrapy的真实使用

一、伪造头部和cookie
1.HEADER 和 COOKIES 在 settings.py 中定义如下：

HEADER={
    "Host": "www.zhihu.com",
    "Connection": "keep-alive",
    "Cache-Control": "max-age=0",
    "Accept": "text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,*/*;q=0.8",
    "User-Agent": "Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/34.0.1847.131 Safari/537.36",
    "Referer": "http://www.zhihu.com/people/raymond-wang",
    "Accept-Encoding": "gzip,deflate,sdch",
    "Accept-Language": "zh-CN,zh;q=0.8,en-US;q=0.6,en;q=0.4,zh-TW;q=0.2",
    }

COOKIES={
    'checkcode':r'"$2a$10$9FVE.1nXJKq/F.nH62OhCevrCqs4skby2bC4IO6VPJITlc7Sh.NZa"',
    'c_c':r'a153f80493f411e3801452540a3121f7',
    '_ga':r'GA1.2.1063404131.1384259893',
    'zata':r'zhihu.com.021715f934634a988abbd3f1f7f31f37.470330',
    'q_c1':r'59c45c60a48d4a5f9a12a52028a9aee7|1400081868000|1400081868000',
    '_xsrf':r'2a7cf7208bf24dbda3f70d953e948135',
    'q_c0':r'"NmE0NzBjZTdmZGI4Yzg3ZWE0NjhkNjkwZGNiZTNiN2F8V2FhRTQ1QklrRjNjNGhMdQ==|1400082425|a801fc83ab07cb92236a75c87de58dcf3fa15cff"',
    '__utma':r'51854390.1063404131.1384259893.1400518549.1400522270.5',
    '__utmb':r'51854390.4.10.1400522270',
    '__utmc':r'51854390',
    '__utmz':r'51854390.1400513283.3.3.utmcsr=zhihu.com|utmccn=(referral)|utmcmd=referral|utmcct=/people/hallson',
    '__utmv':r'51854390.100-1|2=registration_date=20121016=1^3=entry_date=20121016=1'
}


2.调用
# -*- coding:utf-8 -*-

from scrapy.selector import Selector
from scrapy.contrib.linkextractors.sgml import SgmlLinkExtractor
from scrapy.contrib.spiders import CrawlSpider, Rule
from scrapy.http import Request,FormRequest

from zhihu.settings import *

class ZhihuLoginSpider(CrawlSpider):
    name = 'zhihulogin1'
    allowed_domains = ['zhihu.com']
    start_urls = ['http://www.zhihu.com/lookup/class/']

    rules = (
        Rule(SgmlLinkExtractor(allow=r'search/')),
        Rule(SgmlLinkExtractor(allow=r'')),
    )

    def __init__(self):
        self.headers =HEADER
        self.cookies =COOKIES

    def start_requests(self):
        for i, url in enumerate(self.start_urls):
            yield FormRequest(url, meta = {'cookiejar': i}, \
                              headers = self.headers, \
                              cookies =self.cookies,
                              callback = self.parse_item)#jump to login page

    def parse_item(self, response):
        selector = Selector(response)

        urls = []
        for ele in selector.xpath('//ul/li[@class="suggest-item"]/div/a/@href').extract():
           urls.append(ele)
        print urls


二、自动登录并采集某网站中的所有用户信息
1）.重写 start_requests 方法，通过设置 FormRequest 的 formdata 参数，这里是 email 和 password，然后提交请求到 http://www.zhihu.com/login进行登陆，如果登陆成功之后，调用 after_login 回调方法。
2）.在 after_login 方法中，逐个访问 start_urls 中的 url
3）rules 中定义了一些正则匹配的 url 所对应的回调函数

使用账户和密码进行登陆代码如下：

# -*- coding:utf-8 -*-
from scrapy.selector import Selector
from scrapy.contrib.linkextractors.sgml import SgmlLinkExtractor
from scrapy.contrib.spiders import CrawlSpider, Rule
from scrapy.http import Request,FormRequest

import sys

reload(sys)
sys.setdefaultencoding('utf-8')

host='http://www.zhihu.com'

class ZhihuUserSpider(CrawlSpider):
    name = 'zhihu_user'
    allowed_domains = ['zhihu.com']
    start_urls = ["http://www.zhihu.com/lookup/people",]

    #使用rule时候，不要定义parse方法
    rules = (
        Rule(SgmlLinkExtractor(allow=("/lookup/class/[^/]+/?$", )), follow=True,callback='parse_item'),
        Rule(SgmlLinkExtractor(allow=("/lookup/class/$", )), follow=True,callback='parse_item'),
        Rule(SgmlLinkExtractor(allow=("/lookup/people", )),  callback='parse_item'),
   )

    def __init__(self,  *a,  **kwargs):
        super(ZhihuLoginSpider, self).__init__(*a, **kwargs)

    #设置账号密码进行自动登录
    def start_requests(self):
        return [FormRequest(
            "http://www.zhihu.com/login",
            formdata = {'email':'XXXXXX',
                        'password':'XXXXXX'
            },
            callback = self.after_login
        )]

    #逐个访问start_urls 中的 url
    def after_login(self, response):
        for url in self.start_urls:
            yield self.make_requests_from_url(url)

    def parse_item(self, response):
        selector = Selector(response)
        for link in selector.xpath('//div[@id="suggest-list-wrap"]/ul/li/div/a/@href').extract():
            #link  ===> /people/javachen
            yield Request(host+link+"/about", callback=self.parse_user)

    #提取用户信息，通过提取关注数，从而提取关注列表并从中使用parse_follow_url分析处更多关注列表
    def parse_user(self, response): 
        selector = Selector(response)
        user = ZhihuUserItem()
        user['_id']=user['username']=response.url.split('/')[-2]
        user['url']= response.url
        user['nickname'] = ''.join(selector.xpath("//div[@class='title-section ellipsis']/a[@class='name']/text()").extract())
        user['location'] = ''.join(selector.xpath("//span[@class='location item']/@title").extract())
        user['industry'] = ''.join(selector.xpath("//span[@class='business item']/@title").extract())
        user['sex'] = ''.join(selector.xpath('//div[@class="item editable-group"]/span/span[@class="item"]/i/@class').extract()).replace("zg-icon gender ","")
        user['description'] = ''.join(selector.xpath("//span[@class='description unfold-item']/span/text()").extract()).strip().replace("\n",'')
        user['view_num'] = ''.join(selector.xpath("//span[@class='zg-gray-normal']/strong/text()").extract())
        user['update_time'] = str(datetime.now())
        #抓取用户信息，此处省略更多代码

    #获取到用户的关注数 followee_num后，继续采集该用户所有的关注列表
        _xsrf = ''.join(selector.xpath('//input[@name="_xsrf"]/@value').extract())
        hash_id = ''.join(selector.xpath('//div[@class="zm-profile-header-op-btns clearfix"]/button/@data-id').extract())

        num = int(followee_num) if followee_num else 0
    page_num = num/20
    page_num += 1 if num%20 else 0
    for i in xrange(page_num):
            params = json.dumps({"hash_id":hash_id,"order_by":"created","offset":i*20})
            payload = {"method":"next", "params": params, "_xsrf":_xsrf}
            yield Request("http://www.zhihu.com/node/ProfileFolloweesListV2?"+urlencode(payload), callback=self.parse_follow_url)

    #将用户关注列表中的链接转化为用户信息采集链接，并继续采集，并回调parse_user提取用户信息
    def parse_follow_url(self, response):
        selector = Selector(response)

        for link in selector.xpath('//div[@class="zm-list-content-medium"]/h2/a/@href').extract():
            #link  ===> http://www.zhihu.com/people/peng-leslie-97
            username_tmp = link.split('/')[-1]
            if username_tmp in self.user_names:
                print 'GET:' + '%s' % username_tmp
                continue

            yield Request(link+"/about", callback=self.parse_user)