traverseda
/
cpython
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

Issue #28541: Improve test coverage for encoding detection in json library. 

Original patch by Eric Appelt.
This commit is contained in:
Serhiy Storchaka 2016-10-30 23:00:01 +02:00
parent a0d9c685d0
commit c4a35daa97
2 changed files with 15 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
Lib/json/__init__.py
View File

@ -257,7 +257,8 @@ def detect_encoding(b):
return 'utf-16-be' if b[1] else 'utf-32-be'
if not b[1]:
# XX 00 00 00 - utf-32-le
# XX 00 XX XX - utf-16-le
# XX 00 00 XX - utf-16-le
# XX 00 XX -- - utf-16-le
return 'utf-16-le' if b[2] or b[3] else 'utf-32-le'
elif len(b) == 2:
if not b[0]:

13
Lib/test/test_json/test_unicode.py
View File

@ -65,6 +65,19 @@ class TestUnicode:
self.assertEqual(self.loads(bom + encoded), data)
self.assertEqual(self.loads(encoded), data)
self.assertRaises(UnicodeDecodeError, self.loads, b'["\x80"]')
# RFC-7159 and ECMA-404 extend JSON to allow documents that
# consist of only a string, which can present a special case
# not covered by the encoding detection patterns specified in
# RFC-4627 for utf-16-le (XX 00 XX 00).
self.assertEqual(self.loads('"\u2600"'.encode('utf-16-le')),
'\u2600')
# Encoding detection for small (<4) bytes objects
# is implemented as a special case. RFC-7159 and ECMA-404
# allow single codepoint JSON documents which are only two
# bytes in utf-16 encodings w/o BOM.
self.assertEqual(self.loads(b'5\x00'), 5)
self.assertEqual(self.loads(b'\x007'), 7)
self.assertEqual(self.loads(b'57'), 57)
def test_object_pairs_hook_with_unicode(self):
s = '{"xkd":1, "kcw":2, "art":3, "hxm":4, "qrt":5, "pad":6, "hoy":7}'