python爬虫之图片下载

最新推荐文章于 2024-05-08 15:02:16 发布

「已注销」

最新推荐文章于 2024-05-08 15:02:16 发布

阅读量844

点赞数

文章标签： python

本文链接：https://blog.csdn.net/qq_36316736/article/details/54693424

版权

import urllib.request
import re

url = 'http://www.3lian.com/gif/more/05/'

def open_url(url):
    req = urllib.request.Request(url)
    req.add_header('User-Agent','Mozilla/5.0 (Windows NT 10.0; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/49.0.2623.22 Safari/537.36 SE 2.X MetaSr 1.0')
    page = urllib.request.urlopen(req)
    html = page.read().decode('GBK')

    return html

def get_img(html):
    p = r'src="(.*?\.jpg)" width'
    img = re.findall(p,html)

    for img_url in img:
        filename = img_url.split("/")[-1]
        print(filename)
        urllib.request.urlretrieve(img_url,filename)

def get_imgs(html):
    p = r'<a href="(.*?.html)" target'
    imgs = re.findall(p,html)
    for each in imgs:
        imgsurl = 'http://www.3lian.com/' + each
        imgshtml = open_url(imgsurl)
        get_img(imgshtml)

html = open_url(url)
get_imgs(html)