It’s a re-post of my comment here: 1

Basically, I think there is a bug in the json.dump() function in Python 2 (only) - It can’t dump a Python (dictionary / list) data containing non-ASCII characters, even you open the file with the encoding = 'utf-8' parameter. (i.e. no matter what you do). However, json.dump() works fine on Python 3 and json.dumps() works on both Python 2 and 3.

The following code breaks in Python 2 with exception TypeError: must be unicode, not str (Python 2.7.6, Debian):

import json
data = {u'\u0430\u0431\u0432\u0433\u0434': 1} #{u'абвгд': 1}
with open('data.txt', 'w') as outfile:
    json.dump(data, outfile)

It, however, works fine in Python 3.

The following stroke-out statements are wrong, so just ignore them.

So in summary: To dump a JSON *correctly* on both Python 2 and 3, you need the following code (`json.dumps()` uses `'utf-8'` as the default encoding, which agrees with the `io.open()` parameter) 
import io, json
with io.open('data.txt', 'w', encoding='utf-8') as f:
    f.write(json.dumps(data))
In Python 3, however, you can simply use the following: 
import io, json
with io.open('data.txt', 'w', encoding='utf-8') as f:
    json.dump(data, f)

References:

  1. http://stackoverflow.com/a/34574105/404271 ↩