下面代码将可以做到避免302重定向到新的网页
#!/usr/bin/python# -*- coding: utf-8 -*-#encoding=utf-8#filename:states_code.py import urllib2 class redircthandler(urllib2.httpredirecthandler): docstring for redircthandler def http_error_301(self, req, fp, code, msg, headers): pass def http_error_302(self, req, fp, code, msg, headers): pass def getunredirecturl(url,timeout=10): req = urllib2.request(url) debug_handler = urllib2.httphandler(debuglevel = 1) opener = urllib2.build_opener(debug_handler, redircthandler) html = none response = none try: response = opener.open(url,timeout=timeout) html = response.read() except urllib2.urlerror as e: if hasattr(e, 'code'): error_info = e.code elif hasattr(e, 'reason'): error_info = e.reason finally: if response: response.close() if html: return html else: return error_info html = getunredirecturl('http://jb51.net')print html
