felinx

# based on http://uberblo.gs/2011/06/high-performance-url-shortening-with-redis-backed-nginx
# using code from http://stackoverflow.com/questions/3554315/lua-base-converter

# "database scheme"
# database 0: id ~> url
# database 1: id ~> hits
# database 2: id ~> [{referer|user_agent}]
# database 3: id ~> hits (when id is not found)
# database 4: id ~> [{referer|user_agent}] (when id is not found)
# database 5: key "count" storing the number of shortened urls; the id is generated by (this number + 1) converted to base 62

felinx

#!/usr/bin/env python
# -*- coding: utf-8 -*-
#
#       zipdb.py
#       Use a zipfile store a dict like k-v database.
#       Known bug: duplicate key(filenames) allowed
#       
#       Copyright 2012 mayli <mayli.he@gmail.com>
#