1import codecs 2from collections import OrderedDict 3from test.test_json import PyTest, CTest 4 5 6class TestUnicode: 7 # test_encoding1 and test_encoding2 from 2.x are irrelevant (only str 8 # is supported as input, not bytes). 9 10 def test_encoding3(self): 11 u = '\N{GREEK SMALL LETTER ALPHA}\N{GREEK CAPITAL LETTER OMEGA}' 12 j = self.dumps(u) 13 self.assertEqual(j, '"\\u03b1\\u03a9"') 14 15 def test_encoding4(self): 16 u = '\N{GREEK SMALL LETTER ALPHA}\N{GREEK CAPITAL LETTER OMEGA}' 17 j = self.dumps([u]) 18 self.assertEqual(j, '["\\u03b1\\u03a9"]') 19 20 def test_encoding5(self): 21 u = '\N{GREEK SMALL LETTER ALPHA}\N{GREEK CAPITAL LETTER OMEGA}' 22 j = self.dumps(u, ensure_ascii=False) 23 self.assertEqual(j, '"{0}"'.format(u)) 24 25 def test_encoding6(self): 26 u = '\N{GREEK SMALL LETTER ALPHA}\N{GREEK CAPITAL LETTER OMEGA}' 27 j = self.dumps([u], ensure_ascii=False) 28 self.assertEqual(j, '["{0}"]'.format(u)) 29 30 def test_big_unicode_encode(self): 31 u = '\U0001d120' 32 self.assertEqual(self.dumps(u), '"\\ud834\\udd20"') 33 self.assertEqual(self.dumps(u, ensure_ascii=False), '"\U0001d120"') 34 35 def test_big_unicode_decode(self): 36 u = 'z\U0001d120x' 37 self.assertEqual(self.loads('"' + u + '"'), u) 38 self.assertEqual(self.loads('"z\\ud834\\udd20x"'), u) 39 40 def test_unicode_decode(self): 41 for i in range(0, 0xd7ff): 42 u = chr(i) 43 s = '"\\u{0:04x}"'.format(i) 44 self.assertEqual(self.loads(s), u) 45 46 def test_unicode_preservation(self): 47 self.assertEqual(type(self.loads('""')), str) 48 self.assertEqual(type(self.loads('"a"')), str) 49 self.assertEqual(type(self.loads('["a"]')[0]), str) 50 51 def test_bytes_encode(self): 52 self.assertRaises(TypeError, self.dumps, b"hi") 53 self.assertRaises(TypeError, self.dumps, [b"hi"]) 54 55 def test_bytes_decode(self): 56 for encoding, bom in [ 57 ('utf-8', codecs.BOM_UTF8), 58 ('utf-16be', codecs.BOM_UTF16_BE), 59 ('utf-16le', codecs.BOM_UTF16_LE), 60 ('utf-32be', codecs.BOM_UTF32_BE), 61 ('utf-32le', codecs.BOM_UTF32_LE), 62 ]: 63 data = ["a\xb5\u20ac\U0001d120"] 64 encoded = self.dumps(data).encode(encoding) 65 self.assertEqual(self.loads(bom + encoded), data) 66 self.assertEqual(self.loads(encoded), data) 67 self.assertRaises(UnicodeDecodeError, self.loads, b'["\x80"]') 68 # RFC-7159 and ECMA-404 extend JSON to allow documents that 69 # consist of only a string, which can present a special case 70 # not covered by the encoding detection patterns specified in 71 # RFC-4627 for utf-16-le (XX 00 XX 00). 72 self.assertEqual(self.loads('"\u2600"'.encode('utf-16-le')), 73 '\u2600') 74 # Encoding detection for small (<4) bytes objects 75 # is implemented as a special case. RFC-7159 and ECMA-404 76 # allow single codepoint JSON documents which are only two 77 # bytes in utf-16 encodings w/o BOM. 78 self.assertEqual(self.loads(b'5\x00'), 5) 79 self.assertEqual(self.loads(b'\x007'), 7) 80 self.assertEqual(self.loads(b'57'), 57) 81 82 def test_object_pairs_hook_with_unicode(self): 83 s = '{"xkd":1, "kcw":2, "art":3, "hxm":4, "qrt":5, "pad":6, "hoy":7}' 84 p = [("xkd", 1), ("kcw", 2), ("art", 3), ("hxm", 4), 85 ("qrt", 5), ("pad", 6), ("hoy", 7)] 86 self.assertEqual(self.loads(s), eval(s)) 87 self.assertEqual(self.loads(s, object_pairs_hook = lambda x: x), p) 88 od = self.loads(s, object_pairs_hook = OrderedDict) 89 self.assertEqual(od, OrderedDict(p)) 90 self.assertEqual(type(od), OrderedDict) 91 # the object_pairs_hook takes priority over the object_hook 92 self.assertEqual(self.loads(s, object_pairs_hook = OrderedDict, 93 object_hook = lambda x: None), 94 OrderedDict(p)) 95 96 97class TestPyUnicode(TestUnicode, PyTest): pass 98class TestCUnicode(TestUnicode, CTest): pass 99