序言

2017，定下一個小目標(biāo)，深度學(xué)Python~

爬取對象

抓取豆瓣電影Top250

使用包

import codecs  #非必須，用于指定創(chuàng)建記錄爬取文件字符集
import requests #數(shù)據(jù)抓取
from bs4 import BeautifulSoup #抓取內(nèi)容解析

實現(xiàn)步驟

1、編寫抓取偽代碼

url = 'https://www.douban.com/'
webPage = requests.get(url).text
soup = BeautifulSoup(webPage,"html.parser")
print(soup.title)  # <title>豆瓣</title>

2、抓取對象分析

• 通過F12捕獲頁面內(nèi)容得知待抓取內(nèi)容以列表形式顯示在grid_view中，每“行”主要分為左側(cè)圖片、右側(cè)電影名稱等文本介紹
  
  肖申克的救贖
• 分析翻頁機制，重點關(guān)注不同頁間URL區(qū)別
  
  翻頁

3、電影名稱抓取實現(xiàn)

import codecs  #非必須，用于指定創(chuàng)建記錄爬取文件字符集
import requests #數(shù)據(jù)抓取
from bs4 import BeautifulSoup #抓取內(nèi)容解析

DOWNLOAD_URL = 'http://movie.douban.com/top250/'

def download_page(url):
    return requests.get(url, headers={
        'User-Agent': 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_11_2) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/47.0.2526.80 Safari/537.36'
    }).content

def parse_html(html):
    soup = BeautifulSoup(html, "html.parser")
    movie_list_soup = soup.find('ol', attrs={'class': 'grid_view'})

    movie_name_list = []

    print("當(dāng)前準(zhǔn)備解析頁碼：" + soup.find('span', attrs={'class': 'thispage'}).getText())

    top_num = 1 + (int(soup.find('span', attrs={'class': 'thispage'}).getText())-1) * 25

    for movie_li in movie_list_soup.find_all('li'):
        detail = movie_li.find('div', attrs = {'class': 'hd'})

        movie_name_list.append("## Top " + str(top_num))

        #名稱
        movie_name = ""

        for sp in detail.find_all('span', attrs={'class': 'title'}):
            movie_name += sp.text

        img = movie_li.find('div', attrs={'class': 'pic'}).find('a').find('img')

        movie_name_list.append("### "+ movie_name )

        top_num +=1

    next_page = soup.find('span', attrs={'class': 'next'}).find('a')

    if next_page:
        return movie_name_list, DOWNLOAD_URL + next_page['href']
    return movie_name_list, None

def main():
    url = DOWNLOAD_URL

    with codecs.open('douban_moviesList_top250.md', 'wb', encoding='utf-8') as fp:
        while url:
            html = download_page(url)
            movies, url = parse_html(html)
            fp.write(u'{movies}\n'.format(movies='\n'.join(movies)))

if __name__ == '__main__':
    main()

電影名稱

4、基于名稱，補充剩余信息解析

• 封面解析示意

        img = movie_li.find('div', attrs={'class': 'pic'}).find('a').find('img')
        
        try:
            img_req = requests.get(img["src"], timeout=20)
            img_localhost = 'douban_moviesList_top250\\'+str(top_num)+ '.jpg'
            with open(img_localhost, 'wb') as f:
                f.write(img_req.content)

            movie_name_list.append('![](douban_moviesList_top250/'+str(top_num)+'.jpg "douban_moviesList_top250")')
            
        except requests.exceptions.ConnectionError:
            print('【錯誤】當(dāng)前圖片無法下載，失效地址為：' + img["src"])

總結(jié)

因豆瓣大部分均為靜態(tài)頁面，相對較為簡單，主要涉及翻頁循環(huán)、圖片抓取異常處理、BeautifulSoup 中 find、find_all使用......
好了，到此結(jié)束，我是不會承認我會抓取豆瓣妹子什么的......
完整代碼

豆瓣Top250