使用raw_unicode_escape解析转义过的unicode字符串

tiny_json_decoder.py

# -*- coding:utf-8 -*- #

import sys
reload(sys)
sys.setdefaultencoding("utf-8")

import operator
import urllib
import json

# Copyright: Liao Xuefeng (https://github.com/michaelliao/sinaweibopy/blob/master/weibo.py)
class JsonDict(dict):
  """general json object that allows attributes to be bound to and also behaves like a dict"""

  def __getattr__(self, attr):
    try:
      return self[attr]
    except KeyError:
      raise AttributeError(r"'JsonDict' object has no attribute '%s'" % attr)

  def __setattr__(self, attr, value):
    self[attr] = value

def parse_json(s):
  """parse str into JsonDict"""

  def _obj_hook(pairs):
    """convert json object to python object"""
    o = JsonDict()
    for k, v in pairs.iteritems():
      o[str(k)] = v
    return o

  return json.loads(s, object_hook=_obj_hook)

def main():
  for line in file(sys.argv[1]):
    try:
      json_data = parse_json(line)
      print json_data.text.decode('raw_unicode_escape')
    except:
      print "[ERROR]", line

main()