当当图书信息爬取

效果：

分析：

>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>

version_0

声明：未经许可，不能作为商业用途

总结：通过//div[@class="xxx"]可能取到的数据是不全面的,这时候不妨考虑使用//div[contains(@calss,'xxx')]的方式来提取

　　　如果通过re模块去提取数据，在首页(book.dangdang.com/index)取获取分类信息的时候，会提示errordecode，

　　　这是因为当当图书在网页中插入了别国字符导致编码不统一的问题。

　　　当当网在获取图书信息，翻页时，未采用任何动态技术，通过价格也是直接嵌入在网页上的，这个就比较容易获取到。

<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<

源码

# -*- coding: utf-8 -*-
import scrapy
import re
from copy import deepcopy
from pprint import pprint
from urllib import parse
 
 
class DdtsSpider(scrapy.Spider):
    name = 'ddts'
    allowed_domains = ['dangdang.com']
    start_urls = ['http://book.dangdang.com/index']
 
    def process_info(self,con_list):
        """传入一个列表，处理空字符串并将字段拼接在一起"""
        con_list = [re.sub(r"\s|\n", '', i).strip() for i in con_list if i]
        s = str()
        for a_ in con_list:
            s += a_
        return s
 
    def parse(self, response):
        div_cate_list = response.xpath("//div[@class='con flq_body']//div[contains(@class,'level_one')]")
        # 去掉空字符串，去掉当当出版
        div_cate_list = div_cate_list[2:13]+div_cate_list[14:-4]
        for div_cate in div_cate_list:
            item = dict()
            # 获取大分类标题
            #   提取标题部分
            item["b_cate"] = div_cate.xpath(".//dl[contains(@class,'primary_dl')]/dt//text()").extract()
            item["b_cate"] = self.process_info(item["b_cate"])
            # 拿到所有弹出层列表
            t_list = div_cate.xpath(".//dl[contains(@class,'inner_dl')]")
            for t in t_list:
                # 获取中级标题
                item["m_cate"] = t.xpath(".//dt//text()").extract()
                item["m_cate"] = self.process_info(item["m_cate"])
                # 获取小分类及地址
                a_list = t.xpath(".//dd/a")
                for a in a_list:
                    item["s_cate"] = a.xpath("./text()").extract()
                    item["s_cate"] = self.process_info(item["s_cate"])
                    s_href = a.xpath("./@href").extract_first()
                    # 请求小分类的地址
                    yield scrapy.Request(
                        url = s_href,
                        callback=self.parse_s_cate,
                        meta={"item":deepcopy(item)}
                    )
 
    def parse_s_cate(self,response):
        item = deepcopy(response.meta["item"])
        # 选取图书列表
        book_li_list = response.xpath("//div[contains(@id,'search_nature_rg')]/ul[contains(@class,'bigimg')]/li")
        # 当前请求的url包含该页面下所有的请求，无任何动态加载
        for book_li in book_li_list:
            book_info = dict()
            book_info["title"] = book_li.xpath(".//p[contains(@class,'name')]//a/@title").extract()
            book_info["title"] = self.process_info(book_info["title"])
            book_info["href"] = book_li.xpath(".//p[contains(@class,'name')]//a/@href").extract_first()
            book_info["price"] = book_li.xpath(".//p[contains(@class,'price')]//span[contains(@class,'earch_now_price')]/text()").extract_first()
            book_info["price"] = book_info["price"].split(r";",1)[-1]
            book_info["author"] = book_li.xpath(".//a[contains(@name,'itemlist-author')]/text()").extract_first()
            book_info["press"] = book_li.xpath(".//a[contains(@name,'P_cbs')]/text()").extract_first()
            book_info["description"] = book_li.xpath(".//p[contains(@class,'detail')]//text()").extract_first()
            item["book_info"] = book_info
            pprint(item)
        url = response.xpath("//li[@class='next']/a/@href").extract_first()
        if url is not None:
            next_url = parse.urljoin(response.url,url)
            yield scrapy.Request(
                url=next_url,
                callback=self.parse_s_cate,
                meta={"item":response.meta["item"]}
            )