使用requests+BeaBeautiful Soup爬取妹子图图片

1. Requests：让 HTTP 服务人类

Requests 继承了urllib2的所有特性。Requests支持HTTP连接保持和连接池，支持使用cookie保持会话，支持文件上传，支持自动确定响应内容的编码，支持国际化的 URL 和 POST 数据自动编码。

中文文档API：http://docs.python-requests.org/zh_CN/latest/index.html

安装方式

两种方式任选其一

pip install requests

easy_install requests

2. BeautifulSoup

Beautiful Soup 也是一个HTML/XML的解析器，主要的功能也是如何解析和提取 HTML/XML 数据。

2.1 安装

pip install beautifulsoup4

2.2 导入

from bs4 import BeautifulSoup

爬取代码

import requests
from bs4 import BeautifulSoup
import os
import re

Hostreferer = {
    'User-Agent': 'Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.1)',
    'Referer': 'http://www.mzitu.com'
}
Picreferer = {
    'User-Agent': 'Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.1)',
    'Referer': 'http://i.meizitu.net'
}

def get_page_name(url):  # 获得图集最大页数和名称
    html = get_html(url)
    # 创建Beautiful Soup对象，指定lxml解析器
    soup = BeautifulSoup(html, 'lxml')
    span = soup.findAll('span')
    title = soup.find('h2', class_="main-title")
    return span[9].text, title.text

def get_html(url):  # 获得页面html代码
    req = requests.get(url, headers=Hostreferer)
    html = req.text
    return html

def get_img_url(url, name):  # 获取图集里面的每张图片
    html = get_html(url)
    soup = BeautifulSoup(html, 'lxml')
    img_url = soup.find('img', alt=name)
    return img_url['src']

def save_img(img_url, count, name): # 保存图集里面的每张图片
    req = requests.get(img_url, headers=Picreferer)
    new_name = rename(name)
    with open(new_name + '/' + str(count) + '.jpg', 'wb') as f:
        f.write(req.content)

def rename(name):   # 正则匹配
    rstr = r'[\/\\\:\*\?\<\>\|]'
    new_name = re.sub(rstr, "", name)
    return new_name

def save_one_atlas(old_url):  # 保存图集
    page, name = get_page_name(old_url)
    new_name = rename(name)
    os.mkdir(new_name)

    print("图集--" + name + "--开始保存")
    for i in range(1, int(page) + 1):
        url = old_url + "/" + str(i)
        img_url = get_img_url(url, name)
        # print(img_url)
        save_img(img_url, i, name)
        print('正在保存第' + str(i) + '张图片')
    print("图集--" + name + "保存成功")

def get_atlas_list(url):   # 获取网站里面的所有图集
    req = requests.get(url, headers=Hostreferer)
    soup = BeautifulSoup(req.text, 'lxml')
    atlas = soup.find_all(attrs={'class': 'lazy'})
    atlas_list = []
    for atla in atlas:
        atlas_list.append(atla.parent['href'])
    return atlas_list

def save_one_page(start_url): # 调用方法save_one_atlas保存
    atlas_url = get_atlas_list(start_url)
    for url in atlas_url:
        save_one_atlas(url)

if __name__ == '__main__':
    start_url = "http://www.mzitu.com/"
    for count in range(1, 3):
        url = start_url + "page/" + str(count) + "/"
        save_one_page(url)
    print("爬取完成")