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：
importosimporturllibimporturllib.requestimportre#爬行图片download_path = os.path.dirname(os.path.abspath(__file__))classspider(object):def__init__(self, url):
self.url=urldefparse(self,content):
pattern='src="(http://.*\.jpg)\s*"'matchs=re.findall(pattern,content,re.M)returnmatchsdefdownloads(self,urls):
d_path= download_path +"/test"ifnotos.path.exists(d_path):
os.mkdir(d_path)forurlinurls:
filename= url.split("/")[-1]PRint(url)print("Downloads %s"%(filename))
output="%s/%s"%(d_path, filename)
urllib.request.urlretrieve(url,output)defrun(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()