您所在的位置:小祥子 » 编程 » Python » 正文

python爬图

时间:2015-08-14 编辑:学苒 来源:CnBlogs

闲的无事,看着知乎里种种python优点,按捺不住,装起python3.4。

网上找了点爬行图片的代码,修改至兼容3.4,成功爬行指定url所有jpg图片,代码段如下:

import os
import urllib
import urllib.request
import re
#爬行图片
download_path = os.path.dirname(os.path.abspath(__file__))
class spider(object):
    def __init__(self, url):
        self.url = url
    def parse(self,content):
        pattern = 'src="(http://.*/.jpg)\s*"'
        matchs = re.findall(pattern,content,re.M)
        return matchs    
    def downloads(self,urls):
        d_path = download_path + "/test"
        if not os.path.exists(d_path):
            os.mkdir(d_path)
        for url in urls:
            filename = url.split("/")[-1]
            print (url)
            print ("Downloads %s" % (filename))
            output = "%s/%s" % (d_path, filename)
            urllib.request.urlretrieve(url,output)
           
    def run(self):
        d_url = self.url
        fd = urllib.request.urlopen(d_url)
        try:
            content = fd.read()
            content = content.decode("UTF-8")
            urls = self.parse(content)
            self.downloads(urls)
        finally:
            fd.close()
if __name__ == "__main__":
    sp = spider("http://news.cnfol.com/img/20150814/17638.shtml")
    sp.run()
关键词: