mirror of
https://github.com/kennethreitz/requests.git
synced 2026-06-05 22:50:18 +00:00
Martijn Pieters
a4be9a2578
Needed to appease Travis; it's python 2.6 and 2.7 builds are weird and the `__builtins__` dict is not following CPython conventions.
57 lines
1.6 KiB
Python
57 lines
1.6 KiB
Python
#!/usr/bin/env python
|
|
# -*- coding: utf-8 -*-
|
|
|
|
import os
|
|
import sys
|
|
import unittest
|
|
import random
|
|
|
|
# Path hack.
|
|
sys.path.insert(0, os.path.abspath('..'))
|
|
import requests.utils
|
|
from requests.compat import is_py3, bytes
|
|
|
|
|
|
if is_py3:
|
|
unichr = chr
|
|
chr = lambda c: bytes([c])
|
|
|
|
|
|
class GuessJSONUTFTests(unittest.TestCase):
|
|
"""Tests for the JSON UTF encoding guessing code."""
|
|
|
|
codecs = (
|
|
'utf-8', 'utf-8-sig',
|
|
'utf-16', 'utf-16-le', 'utf-16-be',
|
|
'utf-32', 'utf-32-le', 'utf-32-be'
|
|
)
|
|
|
|
def test_guess_encoding(self):
|
|
# Throw 4-character ASCII strings (encoded to a UTF encoding)
|
|
# at the guess routine; it should correctly guess all codecs.
|
|
guess = requests.utils.guess_json_utf
|
|
for c in range(33, 127): # printable only
|
|
sample = unichr(c) * 4
|
|
for codec in self.codecs:
|
|
res = guess(sample.encode(codec))
|
|
self.assertEqual(res, codec)
|
|
|
|
def test_smoke_encoding(self):
|
|
# Throw random 4-byte strings at the guess function.
|
|
# Any guess for a UTF encoding is verified, a decode exception
|
|
# is a test failure.
|
|
guess = requests.utils.guess_json_utf
|
|
for i in range(1000):
|
|
sample = bytes().join(
|
|
[chr(random.randrange(256)) for _ in range(4)])
|
|
res = guess(sample)
|
|
if res is not None and res != 'utf-8':
|
|
# This should decode without errors if this is *really*
|
|
# something in this encoding. Skip UTF-8, it is more
|
|
# picky about valid data.
|
|
sample.decode(res)
|
|
|
|
|
|
if __name__ == '__main__':
|
|
unittest.main()
|