char-decoding.html revision fd5c6425ce58eb75211be7718d5dee960842a37e 
1<html>
2<head>
3<link rel="stylesheet" href="/js/resources/js-test-style.css">
4<script src="/js/resources/js-test-pre.js"></script>
5<script src="resources/char-decoding-utils.js"></script>
6</head>
7<body>
8<p id="description"></p>
9<div id="console"></div>
10<script>
11
12description("This tests decoding characters in various character sets.");
13
14testDecode('UTF-8', '%E2%88%9A', 'U+221A');
15
16// <http://bugs.webkit.org/show_bug.cgi?id=17014> EUC-CN code A3A0 is mapped to U+E5E5 instead of U+3000
17testDecode('gb2312', '%A3%A0', 'U+3000');
18testDecode('gb_2312-80', '%A3%A0', 'U+3000');
19testDecode('chinese', '%A3%A0', 'U+3000');
20testDecode('gbk', '%A3%A0', 'U+3000');
21testDecode('gb18030', '%A3%A0', 'U+3000');
22testDecode('EUC-CN', '%A3%A0', 'U+3000');
23
24// Test Shift_JIS aliases.
25testDecode('Shift_JIS', '%82%d0', 'U+3072');
26testDecode('shift-jis', '%82%d0', 'U+3072');
27
28// Test that all Korean encodings of EUC-KR family are treated as windows-949.
29var korean = {
30    encodings: ['korean', 'EUC-KR', 'windows-949', 'x-windows-949', 'x-uhc',
31                'iso-ir-149', 'KS_C_5601-1987', 'KS_C_5601-1989' ],
32    encoded: ['%A2%E6', '%A1%A4', '%A1%A9', '%A1%AA', '%A1%AD', '%A2%A6',
33              '%A2%C1', '%1A',    '%1C',    '%8F%A1', '%B4%D3', '%A2%41'],
34    unicode: ['U+20AC', 'U+00B7', 'U+00AD', 'U+2015', 'U+223C', 'U+FF5E', 
35              'U+2299', 'U+001A', 'U+001C', 'U+B8EA', 'U+B2D2', 'U+C910'] 
36};
37
38batchTestDecode(korean);
39   
40// Test that ISO-8859-9 (Turkish) is upgraded to windows-1254 with Euro symbol.
41var turkish = {
42  encodings: ['iso-8859-9', 'latin5', 'windows-1254'],
43  encoded: ['%80', '%9F', '%FD'],
44  unicode: ['U+20AC', 'U+0178', 'U+0131']
45};
46
47batchTestDecode(turkish);
48
49// FIXME: Have to add tests for Euro and a few new characters added to ISO-8859-x 
50// that are NOT subsets of the corresponding Windows codepages. For instance,
51// ISO-8859-7:2003 has Euro at 0xA4 and a couple of other new characters.
52// ICU 3.8.x or later has them. Perhaps, we need to have a separate test that
53// can be enabled only with modern ICU.
54
55// Baltic encodings fine points.
56testDecode('ISO-8859-13', '%A1', 'U+201D');
57testDecode('ISO-8859-13', '%A5', 'U+201E');
58testDecode('ISO-8859-13', '%B4', 'U+201C');
59testDecode('ISO-8859-13', '%FF', 'U+2019');
60testDecode('windows-1257', '%80', 'U+20AC');
61testDecode('windows-1257', '%B4', 'U+00B4');
62testDecode('windows-1257', '%FF', 'U+02D9');
63
64// Greek encodings fine points.
65testDecode('iso-8859-7', '%A1', 'U+2018');
66testDecode('iso-8859-7', '%B5', 'U+0385');
67testDecode('iso-8859-7', '%B6', 'U+0386');
68testDecode('windows-1253', '%80', 'U+20AC');
69testDecode('windows-1253', '%A1', 'U+0385');
70testDecode('windows-1253', '%B5', 'U+00B5');
71testDecode('windows-1253', '%B6', 'U+00B6');
72
73// KOI-8 variants
74testDecode('KOI8-R', '%A4', 'U+2553');
75testDecode('KOI8-R', '%AD', 'U+255C');
76testDecode('KOI8-U', '%A4', 'U+0454');
77testDecode('KOI8-U', '%AD', 'U+0491');
78
79// Test that TIS-620 and ISO-8859-11 (Thai) are upgraded to windows-874.
80// "0xDB => U+F8C1" is a weird PUA mapping that doesn't seem to be of
81// any use, even on Windows.
82var thai = {
83  encodings: ['TIS-620', 'ISO-8859-11', 'windows-874', 'dos-874'],
84  encoded: ['%80', '%96', '%A0', '%A1', '%DB'],
85  unicode: ['U+20AC', 'U+2013', 'U+00A0', 'U+0E01', 'U+F8C1']
86};
87
88batchTestDecode(thai);
89
90// UTF-7 is expressly forbidden, so decoding it should not work correctly.
91// This attempts to decode '<' as UTF-7 (+AD4) but it ends up being decoded
92// as a '+'.
93testDecode('UTF-7', '+AD4', 'U+002B');
94testDecode('utf-7', '+AD4', 'U+002B');
95
96successfullyParsed = true;
97
98</script>
99<script src="/js/resources/js-test-post.js"></script>
100</body>
101</html>
102