layout/style/test/test_parser_diagnostics_unprintables.html

   1 <!doctype html>
   2 <html>
   3 <head>
   4   <meta charset="utf-8">
   5   <title>Test for CSS parser diagnostics escaping unprintable
   6          characters correctly</title>
   7   <script src="/tests/SimpleTest/SimpleTest.js"></script>
   8   <link rel="stylesheet" href="/tests/SimpleTest/test.css">
   9 </head>
  10 <body>
  11 <a target="_blank"
  12    href="https://bugzilla.mozilla.org/show_bug.cgi?id=229827"
  13 >Mozilla Bug 229827</a>
  14 <style id="testbench"></style>
  15 <script type="application/javascript">
  16 // This test has intimate knowledge of how to get the CSS parser to
  17 // emit diagnostics that contain text under control of the user.
  18 // That's not the point of the test, though; the point is only that
  19 // *that text* is properly escaped.
  20
  21 SpecialPowers.wrap(window).docShell.cssErrorReportingEnabled = true;
  22
  23 // There is one "pattern" for each code path through the error reporter
  24 // that might need to escape some kind of user-supplied text.
  25 // Each "pattern" is tested once with each of the "substitution"s below:
  26 // <t>, <i>, and <s> are replaced by the t:, i:, and s: fields of
  27 // each substitution object in turn.
  28 let patterns = [
  29   // REPORT_UNEXPECTED_P (only ever used in contexts where identifier-like
  30   // escaping is appropriate)
  31   { i: "<t>|x{}",                 o: "prefix \u2018<i>\u2019" },
  32   // REPORT_UNEXPECTED_TOKEN with:
  33   // _Ident
  34   { i: "@namespace fnord <t>;",    o: "within @namespace: \u2018<i>\u2019" },
  35   // _Ref
  36   { i: "@namespace fnord #<t>;",   o: "within @namespace: \u2018#<i>\u2019" },
  37   // _Function
  38   { i: "@namespace fnord <t>();",  o: "within @namespace: \u2018<i>(\u2019" },
  39   // _Dimension
  40   { i: "@namespace fnord 14<t>;",  o: "within @namespace: \u201814<i>\u2019" },
  41   // _AtKeyword
  42   { i: "x{@<t>: }",                o: "declaration but found \u2018@<i>\u2019." },
  43   // _String
  44   { i: "x{ color: '<t>'}" ,        o: 'color but found \u2018"<s>"\u2019.' },
  45   // _Bad_String
  46   { i: "x{ color: '<t>\n}",        o: 'color but found \u2018"<s>\u2019.' },
  47 ];
  48
  49 // Blocks of characters to test, and how they should be escaped when
  50 // they appear in identifiers and string constants.
  51 const substitutions = [
  52   // ASCII printables that _can_ normally appear in identifiers,
  53   // so should of course _not_ be escaped.
  54   { t: "-_0123456789",               i: "-_0123456789",
  55                                      s: "-_0123456789" },
  56   { t: "abcdefghijklmnopqrstuvwxyz", i: "abcdefghijklmnopqrstuvwxyz",
  57                                      s: "abcdefghijklmnopqrstuvwxyz" },
  58   { t: "ABCDEFGHIJKLMNOPQRSTUVWXYZ", i: "ABCDEFGHIJKLMNOPQRSTUVWXYZ",
  59                                      s: "ABCDEFGHIJKLMNOPQRSTUVWXYZ" },
  60
  61   // ASCII printables that are not normally valid as the first character
  62   // of an identifier, or the character immediately after a leading dash,
  63   // but can be forced into that position with escapes.
  64   { t: "\\-",    i: "\\-",    s: "-"  },
  65   { t: "\\30 ",  i: "\\30 ",  s: "0"  },
  66   { t: "\\31 ",  i: "\\31 ",  s: "1"  },
  67   { t: "\\32 ",  i: "\\32 ",  s: "2"  },
  68   { t: "\\33 ",  i: "\\33 ",  s: "3"  },
  69   { t: "\\34 ",  i: "\\34 ",  s: "4"  },
  70   { t: "\\35 ",  i: "\\35 ",  s: "5"  },
  71   { t: "\\36 ",  i: "\\36 ",  s: "6"  },
  72   { t: "\\37 ",  i: "\\37 ",  s: "7"  },
  73   { t: "\\38 ",  i: "\\38 ",  s: "8"  },
  74   { t: "\\39 ",  i: "\\39 ",  s: "9"  },
  75   { t: "-\\-",   i: "--",     s: "--" },
  76   { t: "-\\30 ", i: "-\\30 ", s: "-0" },
  77   { t: "-\\31 ", i: "-\\31 ", s: "-1" },
  78   { t: "-\\32 ", i: "-\\32 ", s: "-2" },
  79   { t: "-\\33 ", i: "-\\33 ", s: "-3" },
  80   { t: "-\\34 ", i: "-\\34 ", s: "-4" },
  81   { t: "-\\35 ", i: "-\\35 ", s: "-5" },
  82   { t: "-\\36 ", i: "-\\36 ", s: "-6" },
  83   { t: "-\\37 ", i: "-\\37 ", s: "-7" },
  84   { t: "-\\38 ", i: "-\\38 ", s: "-8" },
  85   { t: "-\\39 ", i: "-\\39 ", s: "-9" },
  86
  87   // ASCII printables that must be escaped in identifiers.
  88   // Most of these should not be escaped in strings.
  89   { t: "\\!\\\"\\#\\$",   i: "\\!\\\"\\#\\$",   s: "!\\\"#$" },
  90   { t: "\\%\\&\\'\\(",    i: "\\%\\&\\'\\(",    s: "%&'(" },
  91   { t: "\\)\\*\\+\\,",    i: "\\)\\*\\+\\,",    s: ")*+," },
  92   { t: "\\.\\/\\:\\;",    i: "\\.\\/\\:\\;",    s: "./:;" },
  93   { t: "\\<\\=\\>\\?",    i: "\\<\\=\\>\\?",    s: "<=>?", },
  94   { t: "\\@\\[\\\\\\]",   i: "\\@\\[\\\\\\]",   s: "@[\\\\]" },
  95   { t: "\\^\\`\\{\\}\\~", i: "\\^\\`\\{\\}\\~", s: "^`{}~" },
  96
  97   // U+0000 - U+0020 (C0 controls, space)
  98   // U+000A LINE FEED, U+000C FORM FEED, and U+000D CARRIAGE RETURN
  99   // cannot be put into a CSS token as escaped literal characters, so
 100   // we do them with hex escapes instead.
 101   // The parser replaces U+0000 with U+FFFD.
 102   { t: "\\\x00\\\x01\\\x02\\\x03",       i: "�\\1 \\2 \\3 ",
 103                                          s: "�\\1 \\2 \\3 " },
 104   { t: "\\\x04\\\x05\\\x06\\\x07",       i: "\\4 \\5 \\6 \\7 ",
 105                                          s: "\\4 \\5 \\6 \\7 " },
 106   { t: "\\\x08\\\x09\\000A\\\x0B",       i: "\\8 \\9 \\a \\b ",
 107                                          s: "\\8 \\9 \\a \\b " },
 108   { t: "\\000C\\000D\\\x0E\\\x0F",       i: "\\c \\d \\e \\f ",
 109                                          s: "\\c \\d \\e \\f " },
 110   { t: "\\\x10\\\x11\\\x12\\\x13",       i: "\\10 \\11 \\12 \\13 ",
 111                                          s: "\\10 \\11 \\12 \\13 " },
 112   { t: "\\\x14\\\x15\\\x16\\\x17",       i: "\\14 \\15 \\16 \\17 ",
 113                                          s: "\\14 \\15 \\16 \\17 " },
 114   { t: "\\\x18\\\x19\\\x1A\\\x1B",       i: "\\18 \\19 \\1a \\1b ",
 115                                          s: "\\18 \\19 \\1a \\1b " },
 116   { t: "\\\x1C\\\x1D\\\x1E\\\x1F\\ ",    i: "\\1c \\1d \\1e \\1f \\ ",
 117                                          s: "\\1c \\1d \\1e \\1f  " },
 118
 119   // U+007F (DELETE) and U+0080 - U+009F (C1 controls)
 120   { t: "\\\x7f\\\x80\\\x81\\\x82",       i: "\\7f \x80\x81\x82",
 121                                          s: "\\7f \x80\x81\x82" },
 122   { t: "\\\x83\\\x84\\\x85\\\x86",       i: "\x83\x84\x85\x86",
 123                                          s: "\x83\x84\x85\x86" },
 124   { t: "\\\x87\\\x88\\\x89\\\x8A",       i: "\x87\x88\x89\x8A",
 125                                          s: "\x87\x88\x89\x8A" },
 126   { t: "\\\x8B\\\x8C\\\x8D\\\x8E",       i: "\x8B\x8C\x8D\x8E",
 127                                          s: "\x8B\x8C\x8D\x8E" },
 128   { t: "\\\x8F\\\x90\\\x91\\\x92",       i: "\x8F\x90\x91\x92",
 129                                          s: "\x8F\x90\x91\x92" },
 130   { t: "\\\x93\\\x94\\\x95\\\x96",       i: "\x93\x94\x95\x96",
 131                                          s: "\x93\x94\x95\x96" },
 132   { t: "\\\x97\\\x98\\\x99\\\x9A",       i: "\x97\x98\x99\x9A",
 133                                          s: "\x97\x98\x99\x9A" },
 134   { t: "\\\x9B\\\x9C\\\x9D\\\x9E\\\x9F", i: "\x9B\x9C\x9D\x9E\x9F",
 135                                          s: "\x9B\x9C\x9D\x9E\x9F" },
 136
 137   // CSS doesn't bother with the full Unicode rules for identifiers,
 138   // instead declaring that any code point greater than or equal to
 139   // U+0080 is a valid identifier character.  Test a small handful
 140   // of both basic and astral plane characters.
 141
 142   // Arabic (caution to editors: there is a possibly-invisible U+200E
 143   // LEFT-TO-RIGHT MARK in each string, just before the close quote)
 144   { t: "أبجدهوزحطيكلمنسعفصقرشتثخذضظغ‎",
 145     i: "أبجدهوزحطيكلمنسعفصقرشتثخذضظغ‎",
 146     s: "أبجدهوزحطيكلمنسعفصقرشتثخذضظغ‎" },
 147
 148   // Box drawing
 149   { t: "─│┌┐└┘├┤┬┴┼╭╮╯╰╴╵╶╷",
 150     i: "─│┌┐└┘├┤┬┴┼╭╮╯╰╴╵╶╷",
 151     s: "─│┌┐└┘├┤┬┴┼╭╮╯╰╴╵╶╷" },
 152
 153   // CJK Unified Ideographs
 154   { t: "一丁丂七丄丅丆万丈三上下丌不与丏",
 155     i: "一丁丂七丄丅丆万丈三上下丌不与丏",
 156     s: "一丁丂七丄丅丆万丈三上下丌不与丏" },
 157
 158   // CJK Unified Ideographs Extension B (astral)
 159   { t: "𠀀𠀁𠀂𠀃𠀄𠀅𠀆𠀇𠀈𠀉𠀊𠀋𠀌𠀍𠀎𠀏",
 160     i: "𠀀𠀁𠀂𠀃𠀄𠀅𠀆𠀇𠀈𠀉𠀊𠀋𠀌𠀍𠀎𠀏",
 161     s: "𠀀𠀁𠀂𠀃𠀄𠀅𠀆𠀇𠀈𠀉𠀊𠀋𠀌𠀍𠀎𠀏" },
 162
 163   // Devanagari
 164   { t: "कखगघङचछजझञटठडढणतथदधनपफबभमयरलळवशषसह",
 165     i: "कखगघङचछजझञटठडढणतथदधनपफबभमयरलळवशषसह",
 166     s: "कखगघङचछजझञटठडढणतथदधनपफबभमयरलळवशषसह" },
 167
 168   // Emoticons (astral)
 169   { t: "😁😂😃😄😅😆😇😈😉😊😋😌😍😎😏😐",
 170     i: "😁😂😃😄😅😆😇😈😉😊😋😌😍😎😏😐",
 171     s: "😁😂😃😄😅😆😇😈😉😊😋😌😍😎😏😐" },
 172
 173   // Greek
 174   { t: "αβγδεζηθικλμνξοπρςστυφχψω",
 175     i: "αβγδεζηθικλμνξοπρςστυφχψω",
 176     s: "αβγδεζηθικλμνξοπρςστυφχψω" }
 177 ];
 178
 179 const npatterns = patterns.length;
 180 const nsubstitutions = substitutions.length;
 181
 182 function quotemeta(str) {
 183   return str.replace(/[-[\]{}()*+?.,\\^$|#\s]/g, "\\$&");
 184 }
 185 function subst(str, sub) {
 186   return str.replace("<t>", sub.t)
 187     .replace("<i>", sub.i)
 188     .replace("<s>", sub.s);
 189 }
 190
 191 var curpat = 0;
 192 var cursubst = -1;
 193 var testbench = document.getElementById("testbench");
 194
 195 function nextTest() {
 196   cursubst++;
 197   if (cursubst == nsubstitutions) {
 198     curpat++;
 199     cursubst = 0;
 200   }
 201   if (curpat == npatterns) {
 202     SimpleTest.finish();
 203     return;
 204   }
 205
 206   let css = subst(patterns[curpat].i, substitutions[cursubst]);
 207   let msg = quotemeta(subst(patterns[curpat].o, substitutions[cursubst]));
 208
 209   info(css);
 210   info(msg);
 211   SimpleTest.expectConsoleMessages(function () { testbench.innerHTML = css },
 212                                    [{ errorMessage: new RegExp(msg) }],
 213                                    nextTest);
 214 }
 215
 216 SimpleTest.waitForExplicitFinish();
 217 nextTest();
 218 </script>
 219 </body>
 220 </html>