1.1 --- /dev/null Thu Jan 01 00:00:00 1970 +0000
1.2 +++ b/layout/style/test/test_parser_diagnostics_unprintables.html Wed Dec 31 06:09:35 2014 +0100
1.3 @@ -0,0 +1,220 @@
1.4 +<!doctype html>
1.5 +<html>
1.6 +<head>
1.7 + <meta charset="utf-8">
1.8 + <title>Test for CSS parser diagnostics escaping unprintable
1.9 + characters correctly</title>
1.10 + <script src="/tests/SimpleTest/SimpleTest.js"></script>
1.11 + <link rel="stylesheet" href="/tests/SimpleTest/test.css">
1.12 +</head>
1.13 +<body>
1.14 +<a target="_blank"
1.15 + href="https://bugzilla.mozilla.org/show_bug.cgi?id=229827"
1.16 +>Mozilla Bug 229827</a>
1.17 +<style id="testbench"></style>
1.18 +<script type="application/javascript;version=1.8">
1.19 +// This test has intimate knowledge of how to get the CSS parser to
1.20 +// emit diagnostics that contain text under control of the user.
1.21 +// That's not the point of the test, though; the point is only that
1.22 +// *that text* is properly escaped.
1.23 +
1.24 +// There is one "pattern" for each code path through the error reporter
1.25 +// that might need to escape some kind of user-supplied text.
1.26 +// Each "pattern" is tested once with each of the "substitution"s below:
1.27 +// <t>, <i>, and <s> are replaced by the t:, i:, and s: fields of
1.28 +// each substitution object in turn.
1.29 +const patterns = [
1.30 + // REPORT_UNEXPECTED_P (only ever used in contexts where identifier-like
1.31 + // escaping is appropriate)
1.32 + { i: "<t>|x{}", o: "prefix '<i>'" },
1.33 + // REPORT_UNEXPECTED_TOKEN with:
1.34 + // _Ident
1.35 + { i: "@namespace fnord <t>;", o: "within @namespace: '<i>'" },
1.36 + // _Ref
1.37 + { i: "@namespace fnord #<t>;", o: "within @namespace: '#<i>'" },
1.38 + // _Function
1.39 + { i: "@namespace fnord <t>();", o: "within @namespace: '<i>('" },
1.40 + // _Dimension
1.41 + { i: "@namespace fnord 14<t>;", o: "within @namespace: '14<i>'" },
1.42 + // _AtKeyword
1.43 + { i: "x{@<t>: }", o: "declaration but found '@<i>'." },
1.44 + // _String
1.45 + { i: "x{ '<t>'}" , o: "declaration but found ''<s>''." },
1.46 + // _Bad_String
1.47 + { i: "x{ '<t>\n}", o: "declaration but found ''<s>'." },
1.48 + // _URL
1.49 + { i: "x{ url('<t>')}", o: "declaration but found 'url('<s>')'." },
1.50 + // _Bad_URL
1.51 + { i: "x{ url('<t>'.)}" , o: "declaration but found 'url('<s>''." }
1.52 +];
1.53 +
1.54 +// Blocks of characters to test, and how they should be escaped when
1.55 +// they appear in identifiers and string constants.
1.56 +const substitutions = [
1.57 + // ASCII printables that _can_ normally appear in identifiers,
1.58 + // so should of course _not_ be escaped.
1.59 + { t: "-_0123456789", i: "-_0123456789",
1.60 + s: "-_0123456789" },
1.61 + { t: "abcdefghijklmnopqrstuvwxyz", i: "abcdefghijklmnopqrstuvwxyz",
1.62 + s: "abcdefghijklmnopqrstuvwxyz" },
1.63 + { t: "ABCDEFGHIJKLMNOPQRSTUVWXYZ", i: "ABCDEFGHIJKLMNOPQRSTUVWXYZ",
1.64 + s: "ABCDEFGHIJKLMNOPQRSTUVWXYZ" },
1.65 +
1.66 + // ASCII printables that are not normally valid as the first character
1.67 + // of an identifier, or the character immediately after a leading dash,
1.68 + // but can be forced into that position with escapes.
1.69 + { t: "\\-", i: "\\-", s: "-" },
1.70 + { t: "\\30 ", i: "\\30 ", s: "0" },
1.71 + { t: "\\31 ", i: "\\31 ", s: "1" },
1.72 + { t: "\\32 ", i: "\\32 ", s: "2" },
1.73 + { t: "\\33 ", i: "\\33 ", s: "3" },
1.74 + { t: "\\34 ", i: "\\34 ", s: "4" },
1.75 + { t: "\\35 ", i: "\\35 ", s: "5" },
1.76 + { t: "\\36 ", i: "\\36 ", s: "6" },
1.77 + { t: "\\37 ", i: "\\37 ", s: "7" },
1.78 + { t: "\\38 ", i: "\\38 ", s: "8" },
1.79 + { t: "\\39 ", i: "\\39 ", s: "9" },
1.80 + { t: "-\\-", i: "-\\-", s: "--" },
1.81 + { t: "-\\30 ", i: "-\\30 ", s: "-0" },
1.82 + { t: "-\\31 ", i: "-\\31 ", s: "-1" },
1.83 + { t: "-\\32 ", i: "-\\32 ", s: "-2" },
1.84 + { t: "-\\33 ", i: "-\\33 ", s: "-3" },
1.85 + { t: "-\\34 ", i: "-\\34 ", s: "-4" },
1.86 + { t: "-\\35 ", i: "-\\35 ", s: "-5" },
1.87 + { t: "-\\36 ", i: "-\\36 ", s: "-6" },
1.88 + { t: "-\\37 ", i: "-\\37 ", s: "-7" },
1.89 + { t: "-\\38 ", i: "-\\38 ", s: "-8" },
1.90 + { t: "-\\39 ", i: "-\\39 ", s: "-9" },
1.91 +
1.92 + // ASCII printables that must be escaped in identifiers.
1.93 + // Most of these should not be escaped in strings.
1.94 + { t: "\\!\\\"\\#\\$", i: "\\!\\\"\\#\\$", s: "!\\\"#$" },
1.95 + { t: "\\%\\&\\'\\(", i: "\\%\\&\\'\\(", s: "%&\\'(" },
1.96 + { t: "\\)\\*\\+\\,", i: "\\)\\*\\+\\,", s: ")*+," },
1.97 + { t: "\\.\\/\\:\\;", i: "\\.\\/\\:\\;", s: "./:;" },
1.98 + { t: "\\<\\=\\>\\?", i: "\\<\\=\\>\\?", s: "<=>?", },
1.99 + { t: "\\@\\[\\\\\\]", i: "\\@\\[\\\\\\]", s: "@[\\\\]" },
1.100 + { t: "\\^\\`\\{\\}\\~", i: "\\^\\`\\{\\}\\~", s: "^`{}~" },
1.101 +
1.102 + // U+0000 - U+0020 (C0 controls, space)
1.103 + // U+000A LINE FEED, U+000C FORM FEED, and U+000D CARRIAGE RETURN
1.104 + // cannot be put into a CSS token as escaped literal characters, so
1.105 + // we do them with hex escapes instead.
1.106 + // The parser replaces U+0000 with U+FFFD.
1.107 + { t: "\\\x00\\\x01\\\x02\\\x03", i: "�\\1 \\2 \\3 ",
1.108 + s: "�\\1 \\2 \\3 " },
1.109 + { t: "\\\x04\\\x05\\\x06\\\x07", i: "\\4 \\5 \\6 \\7 ",
1.110 + s: "\\4 \\5 \\6 \\7 " },
1.111 + { t: "\\\x08\\\x09\\000A\\\x0B", i: "\\8 \\9 \\A \\B ",
1.112 + s: "\\8 \\9 \\A \\B " },
1.113 + { t: "\\000C\\000D\\\x0E\\\x0F", i: "\\C \\D \\E \\F ",
1.114 + s: "\\C \\D \\E \\F " },
1.115 + { t: "\\\x10\\\x11\\\x12\\\x13", i: "\\10 \\11 \\12 \\13 ",
1.116 + s: "\\10 \\11 \\12 \\13 " },
1.117 + { t: "\\\x14\\\x15\\\x16\\\x17", i: "\\14 \\15 \\16 \\17 ",
1.118 + s: "\\14 \\15 \\16 \\17 " },
1.119 + { t: "\\\x18\\\x19\\\x1A\\\x1B", i: "\\18 \\19 \\1A \\1B ",
1.120 + s: "\\18 \\19 \\1A \\1B " },
1.121 + { t: "\\\x1C\\\x1D\\\x1E\\\x1F\\ ", i: "\\1C \\1D \\1E \\1F \\ ",
1.122 + s: "\\1C \\1D \\1E \\1F " },
1.123 +
1.124 + // U+007F (DELETE) and U+0080 - U+009F (C1 controls)
1.125 + { t: "\\\x7f\\\x80\\\x81\\\x82", i: "\\7F \\80 \\81 \\82 ",
1.126 + s: "\\7F \\80 \\81 \\82 " },
1.127 + { t: "\\\x83\\\x84\\\x85\\\x86", i: "\\83 \\84 \\85 \\86 ",
1.128 + s: "\\83 \\84 \\85 \\86 " },
1.129 + { t: "\\\x87\\\x88\\\x89\\\x8A", i: "\\87 \\88 \\89 \\8A ",
1.130 + s: "\\87 \\88 \\89 \\8A " },
1.131 + { t: "\\\x8B\\\x8C\\\x8D\\\x8E", i: "\\8B \\8C \\8D \\8E ",
1.132 + s: "\\8B \\8C \\8D \\8E " },
1.133 + { t: "\\\x8F\\\x90\\\x91\\\x92", i: "\\8F \\90 \\91 \\92 ",
1.134 + s: "\\8F \\90 \\91 \\92 " },
1.135 + { t: "\\\x93\\\x94\\\x95\\\x96", i: "\\93 \\94 \\95 \\96 ",
1.136 + s: "\\93 \\94 \\95 \\96 " },
1.137 + { t: "\\\x97\\\x98\\\x99\\\x9A", i: "\\97 \\98 \\99 \\9A ",
1.138 + s: "\\97 \\98 \\99 \\9A " },
1.139 + { t: "\\\x9B\\\x9C\\\x9D\\\x9E\\\x9F", i: "\\9B \\9C \\9D \\9E \\9F ",
1.140 + s: "\\9B \\9C \\9D \\9E \\9F " },
1.141 +
1.142 + // CSS doesn't bother with the full Unicode rules for identifiers,
1.143 + // instead declaring that any code point greater than or equal to
1.144 + // U+00A0 is a valid identifier character. Test a small handful
1.145 + // of both basic and astral plane characters.
1.146 +
1.147 + // Arabic (caution to editors: there is a possibly-invisible U+200E
1.148 + // LEFT-TO-RIGHT MARK in each string, just before the close quote)
1.149 + { t: "أبجدهوزحطيكلمنسعفصقرشتثخذضظغ",
1.150 + i: "أبجدهوزحطيكلمنسعفصقرشتثخذضظغ",
1.151 + s: "أبجدهوزحطيكلمنسعفصقرشتثخذضظغ" },
1.152 +
1.153 + // Box drawing
1.154 + { t: "─│┌┐└┘├┤┬┴┼╭╮╯╰╴╵╶╷",
1.155 + i: "─│┌┐└┘├┤┬┴┼╭╮╯╰╴╵╶╷",
1.156 + s: "─│┌┐└┘├┤┬┴┼╭╮╯╰╴╵╶╷" },
1.157 +
1.158 + // CJK Unified Ideographs
1.159 + { t: "一丁丂七丄丅丆万丈三上下丌不与丏",
1.160 + i: "一丁丂七丄丅丆万丈三上下丌不与丏",
1.161 + s: "一丁丂七丄丅丆万丈三上下丌不与丏" },
1.162 +
1.163 + // CJK Unified Ideographs Extension B (astral)
1.164 + { t: "𠀀𠀁𠀂𠀃𠀄𠀅𠀆𠀇𠀈𠀉𠀊𠀋𠀌𠀍𠀎𠀏",
1.165 + i: "𠀀𠀁𠀂𠀃𠀄𠀅𠀆𠀇𠀈𠀉𠀊𠀋𠀌𠀍𠀎𠀏",
1.166 + s: "𠀀𠀁𠀂𠀃𠀄𠀅𠀆𠀇𠀈𠀉𠀊𠀋𠀌𠀍𠀎𠀏" },
1.167 +
1.168 + // Devanagari
1.169 + { t: "कखगघङचछजझञटठडढणतथदधनपफबभमयरलळवशषसह",
1.170 + i: "कखगघङचछजझञटठडढणतथदधनपफबभमयरलळवशषसह",
1.171 + s: "कखगघङचछजझञटठडढणतथदधनपफबभमयरलळवशषसह" },
1.172 +
1.173 + // Emoticons (astral)
1.174 + { t: "😁😂😃😄😅😆😇😈😉😊😋😌😍😎😏😐",
1.175 + i: "😁😂😃😄😅😆😇😈😉😊😋😌😍😎😏😐",
1.176 + s: "😁😂😃😄😅😆😇😈😉😊😋😌😍😎😏😐" },
1.177 +
1.178 + // Greek
1.179 + { t: "αβγδεζηθικλμνξοπρςστυφχψω",
1.180 + i: "αβγδεζηθικλμνξοπρςστυφχψω",
1.181 + s: "αβγδεζηθικλμνξοπρςστυφχψω" }
1.182 +];
1.183 +
1.184 +const npatterns = patterns.length;
1.185 +const nsubstitutions = substitutions.length;
1.186 +
1.187 +function quotemeta(str) {
1.188 + return str.replace(/[-[\]{}()*+?.,\\^$|#\s]/g, "\\$&");
1.189 +}
1.190 +function subst(str, sub) {
1.191 + return str.replace("<t>", sub.t)
1.192 + .replace("<i>", sub.i)
1.193 + .replace("<s>", sub.s);
1.194 +}
1.195 +
1.196 +var curpat = 0;
1.197 +var cursubst = -1;
1.198 +var testbench = document.getElementById("testbench");
1.199 +
1.200 +function nextTest() {
1.201 + cursubst++;
1.202 + if (cursubst == nsubstitutions) {
1.203 + curpat++;
1.204 + cursubst = 0;
1.205 + }
1.206 + if (curpat == npatterns) {
1.207 + SimpleTest.finish();
1.208 + return;
1.209 + }
1.210 +
1.211 + let css = subst(patterns[curpat].i, substitutions[cursubst]);
1.212 + let msg = quotemeta(subst(patterns[curpat].o, substitutions[cursubst]));
1.213 +
1.214 + SimpleTest.expectConsoleMessages(function () { testbench.innerHTML = css },
1.215 + [{ errorMessage: new RegExp(msg) }],
1.216 + nextTest);
1.217 +}
1.218 +
1.219 +SimpleTest.waitForExplicitFinish();
1.220 +nextTest();
1.221 +</script>
1.222 +</body>
1.223 +</html>
