Python之简单抓取豆瓣读书信息

最近出差学习，闲来撸一把 Python。看语法书这些，真是看完就忘，还不如来写点小程序，有实践性又有趣。

我的环境是Ubuntu 17，开始之前先装几个依赖包，用于解析 html 文件。

        
            sudo apt install python-lxml,python-requests
    

小程序实现从豆瓣读书上抓取评分8以上，且评分人数不低于800人的书籍。这里取了一个种子，是刘震云老师的《一句顶一万句》。

        
            from lxml import html
import requests

urlPrefix = ‘https://book.douban.com/subject/‘
candidateBookNums = []
candidateBookNums.append(‘3633461‘)

selectedBooks = {}

# 控制循环次数# i = 1while candidateBookNums:

    bookNum = candidateBookNums.pop(0)
    bookUrl = urlPrefix + str(bookNum)

    # 获取网页
    page = requests.get(bookUrl)

    # 将网页格式化为树型
    tree = html.fromstring(page.text)

    # 书籍名称
    bookName = tree.xpath(‘//title/text()‘)
    # 平均分
    rating_num = tree.xpath(‘//strong[@property="v:average"]/text()‘)[0]
    # 评分人数
    rating_people = tree.xpath(‘//a/span[@property="v:votes"]/text()‘)[0]

    if rating_num < 8or rating_people < 800:
        continue

    stars = tree.xpath(‘//span[@]/text()‘)
    # 5星评价比例
    stars5 = stars[0]
    # 4星评价比例
    stars4 = stars[1]
    # 3星评价比例
    stars3 = stars[2]
    # 2星评价比例
    stars2 = stars[3]
    # 1星评价比例
    stars1 = stars[4]
    # 豆瓣读书中指向其他书的链接
    links = tree.xpath(‘//div[@]/dl/dd/a/@href‘)

    # 去掉空白符，如回车、换行、空格、缩进
    bookName = bookName[0].strip()

    # 整理豆瓣上书籍的评分信息
    book = {
        ‘name‘:bookName,
        ‘score‘:rating_num,
        ‘rating_people‘:rating_people,
        ‘stars5‘:stars5,
        ‘stars4‘:stars4,
        ‘stars3‘:stars3,
        ‘stars2‘:stars2,
        ‘stars1‘:stars1,   
    }
    selectedBooks[bookNum] = book
    print bookName,book

    for j in links:
        bookNum = j.split(‘/‘)[-2]
        if bookNum notin selectedBooks.keys() and bookNum notin candidateBookNums:
            candidateBookNums.append(bookNum)

    # i += 1# if i > 100:#   breakprint selectedBooks
    

OK，这样就完成了一个简单的从豆瓣抓取符合要求的书籍的程序。其实实现倒是次要的，主要是从豆瓣读书的页面代码中找到相应信息的位置，提取之。

原文：https://www.cnblogs.com/read-the-spring-and-autumn-annals-in-night/p/12041922.html