| 1 | n/a | import codecs |
|---|
| 2 | n/a | from collections import OrderedDict |
|---|
| 3 | n/a | from test.test_json import PyTest, CTest |
|---|
| 4 | n/a | |
|---|
| 5 | n/a | |
|---|
| 6 | n/a | class TestUnicode: |
|---|
| 7 | n/a | # test_encoding1 and test_encoding2 from 2.x are irrelevant (only str |
|---|
| 8 | n/a | # is supported as input, not bytes). |
|---|
| 9 | n/a | |
|---|
| 10 | n/a | def test_encoding3(self): |
|---|
| 11 | n/a | u = '\N{GREEK SMALL LETTER ALPHA}\N{GREEK CAPITAL LETTER OMEGA}' |
|---|
| 12 | n/a | j = self.dumps(u) |
|---|
| 13 | n/a | self.assertEqual(j, '"\\u03b1\\u03a9"') |
|---|
| 14 | n/a | |
|---|
| 15 | n/a | def test_encoding4(self): |
|---|
| 16 | n/a | u = '\N{GREEK SMALL LETTER ALPHA}\N{GREEK CAPITAL LETTER OMEGA}' |
|---|
| 17 | n/a | j = self.dumps([u]) |
|---|
| 18 | n/a | self.assertEqual(j, '["\\u03b1\\u03a9"]') |
|---|
| 19 | n/a | |
|---|
| 20 | n/a | def test_encoding5(self): |
|---|
| 21 | n/a | u = '\N{GREEK SMALL LETTER ALPHA}\N{GREEK CAPITAL LETTER OMEGA}' |
|---|
| 22 | n/a | j = self.dumps(u, ensure_ascii=False) |
|---|
| 23 | n/a | self.assertEqual(j, '"{0}"'.format(u)) |
|---|
| 24 | n/a | |
|---|
| 25 | n/a | def test_encoding6(self): |
|---|
| 26 | n/a | u = '\N{GREEK SMALL LETTER ALPHA}\N{GREEK CAPITAL LETTER OMEGA}' |
|---|
| 27 | n/a | j = self.dumps([u], ensure_ascii=False) |
|---|
| 28 | n/a | self.assertEqual(j, '["{0}"]'.format(u)) |
|---|
| 29 | n/a | |
|---|
| 30 | n/a | def test_big_unicode_encode(self): |
|---|
| 31 | n/a | u = '\U0001d120' |
|---|
| 32 | n/a | self.assertEqual(self.dumps(u), '"\\ud834\\udd20"') |
|---|
| 33 | n/a | self.assertEqual(self.dumps(u, ensure_ascii=False), '"\U0001d120"') |
|---|
| 34 | n/a | |
|---|
| 35 | n/a | def test_big_unicode_decode(self): |
|---|
| 36 | n/a | u = 'z\U0001d120x' |
|---|
| 37 | n/a | self.assertEqual(self.loads('"' + u + '"'), u) |
|---|
| 38 | n/a | self.assertEqual(self.loads('"z\\ud834\\udd20x"'), u) |
|---|
| 39 | n/a | |
|---|
| 40 | n/a | def test_unicode_decode(self): |
|---|
| 41 | n/a | for i in range(0, 0xd7ff): |
|---|
| 42 | n/a | u = chr(i) |
|---|
| 43 | n/a | s = '"\\u{0:04x}"'.format(i) |
|---|
| 44 | n/a | self.assertEqual(self.loads(s), u) |
|---|
| 45 | n/a | |
|---|
| 46 | n/a | def test_unicode_preservation(self): |
|---|
| 47 | n/a | self.assertEqual(type(self.loads('""')), str) |
|---|
| 48 | n/a | self.assertEqual(type(self.loads('"a"')), str) |
|---|
| 49 | n/a | self.assertEqual(type(self.loads('["a"]')[0]), str) |
|---|
| 50 | n/a | |
|---|
| 51 | n/a | def test_bytes_encode(self): |
|---|
| 52 | n/a | self.assertRaises(TypeError, self.dumps, b"hi") |
|---|
| 53 | n/a | self.assertRaises(TypeError, self.dumps, [b"hi"]) |
|---|
| 54 | n/a | |
|---|
| 55 | n/a | def test_bytes_decode(self): |
|---|
| 56 | n/a | for encoding, bom in [ |
|---|
| 57 | n/a | ('utf-8', codecs.BOM_UTF8), |
|---|
| 58 | n/a | ('utf-16be', codecs.BOM_UTF16_BE), |
|---|
| 59 | n/a | ('utf-16le', codecs.BOM_UTF16_LE), |
|---|
| 60 | n/a | ('utf-32be', codecs.BOM_UTF32_BE), |
|---|
| 61 | n/a | ('utf-32le', codecs.BOM_UTF32_LE), |
|---|
| 62 | n/a | ]: |
|---|
| 63 | n/a | data = ["a\xb5\u20ac\U0001d120"] |
|---|
| 64 | n/a | encoded = self.dumps(data).encode(encoding) |
|---|
| 65 | n/a | self.assertEqual(self.loads(bom + encoded), data) |
|---|
| 66 | n/a | self.assertEqual(self.loads(encoded), data) |
|---|
| 67 | n/a | self.assertRaises(UnicodeDecodeError, self.loads, b'["\x80"]') |
|---|
| 68 | n/a | # RFC-7159 and ECMA-404 extend JSON to allow documents that |
|---|
| 69 | n/a | # consist of only a string, which can present a special case |
|---|
| 70 | n/a | # not covered by the encoding detection patterns specified in |
|---|
| 71 | n/a | # RFC-4627 for utf-16-le (XX 00 XX 00). |
|---|
| 72 | n/a | self.assertEqual(self.loads('"\u2600"'.encode('utf-16-le')), |
|---|
| 73 | n/a | '\u2600') |
|---|
| 74 | n/a | # Encoding detection for small (<4) bytes objects |
|---|
| 75 | n/a | # is implemented as a special case. RFC-7159 and ECMA-404 |
|---|
| 76 | n/a | # allow single codepoint JSON documents which are only two |
|---|
| 77 | n/a | # bytes in utf-16 encodings w/o BOM. |
|---|
| 78 | n/a | self.assertEqual(self.loads(b'5\x00'), 5) |
|---|
| 79 | n/a | self.assertEqual(self.loads(b'\x007'), 7) |
|---|
| 80 | n/a | self.assertEqual(self.loads(b'57'), 57) |
|---|
| 81 | n/a | |
|---|
| 82 | n/a | def test_object_pairs_hook_with_unicode(self): |
|---|
| 83 | n/a | s = '{"xkd":1, "kcw":2, "art":3, "hxm":4, "qrt":5, "pad":6, "hoy":7}' |
|---|
| 84 | n/a | p = [("xkd", 1), ("kcw", 2), ("art", 3), ("hxm", 4), |
|---|
| 85 | n/a | ("qrt", 5), ("pad", 6), ("hoy", 7)] |
|---|
| 86 | n/a | self.assertEqual(self.loads(s), eval(s)) |
|---|
| 87 | n/a | self.assertEqual(self.loads(s, object_pairs_hook = lambda x: x), p) |
|---|
| 88 | n/a | od = self.loads(s, object_pairs_hook = OrderedDict) |
|---|
| 89 | n/a | self.assertEqual(od, OrderedDict(p)) |
|---|
| 90 | n/a | self.assertEqual(type(od), OrderedDict) |
|---|
| 91 | n/a | # the object_pairs_hook takes priority over the object_hook |
|---|
| 92 | n/a | self.assertEqual(self.loads(s, object_pairs_hook = OrderedDict, |
|---|
| 93 | n/a | object_hook = lambda x: None), |
|---|
| 94 | n/a | OrderedDict(p)) |
|---|
| 95 | n/a | |
|---|
| 96 | n/a | |
|---|
| 97 | n/a | class TestPyUnicode(TestUnicode, PyTest): pass |
|---|
| 98 | n/a | class TestCUnicode(TestUnicode, CTest): pass |
|---|
