yarr/YarrCanonicalizeUCS2.js

   1 /*
   2  * Copyright (C) 2012 Apple Inc. All rights reserved.
   3  *
   4  * Redistribution and use in source and binary forms, with or without
   5  * modification, are permitted provided that the following conditions
   6  * are met:
   7  * 1. Redistributions of source code must retain the above copyright
   8  *    notice, this list of conditions and the following disclaimer.
   9  * 2. Redistributions in binary form must reproduce the above copyright
  10  *    notice, this list of conditions and the following disclaimer in the
  11  *    documentation and/or other materials provided with the distribution.
  12  *
  13  * THIS SOFTWARE IS PROVIDED BY APPLE INC. ``AS IS'' AND ANY
  14  * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
  15  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
  16  * PURPOSE ARE DISCLAIMED.  IN NO EVENT SHALL APPLE INC. OR
  17  * CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
  18  * EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
  19  * PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
  20  * PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY
  21  * OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
  22  * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
  23  * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
  24  */
  25
  26 // See ES 5.1, 15.10.2.8
  27 function canonicalize(ch)
  28 {
  29     var u = String.fromCharCode(ch).toUpperCase();
  30     if (u.length > 1)
  31         return ch;
  32     var cu = u.charCodeAt(0);
  33     if (ch >= 128 && cu < 128)
  34         return ch;
  35     return cu;
  36 }
  37
  38 var MAX_UCS2 = 0xFFFF;
  39 var MAX_LATIN = 0xFF;
  40
  41 var groupedCanonically = [];
  42 // Pass 1: populate groupedCanonically - this is mapping from canonicalized
  43 // values back to the set of character code that canonicalize to them.
  44 for (var i = 0; i <= MAX_UCS2; ++i) {
  45     var ch = canonicalize(i);
  46     if (!groupedCanonically[ch])
  47         groupedCanonically[ch] = [];
  48     groupedCanonically[ch].push(i);
  49 }
  50
  51 var typeInfo = [];
  52 var latinTypeInfo = [];
  53 var characterSetInfo = [];
  54 // Pass 2: populate typeInfo & characterSetInfo. For every character calculate
  55 // a typeInfo value, described by the types above, and a value payload.
  56 for (cu in groupedCanonically) {
  57     // The set of characters that canonicalize to cu
  58     var characters = groupedCanonically[cu];
  59
  60     // If there is only one, it is unique.
  61     if (characters.length == 1) {
  62         typeInfo[characters[0]] = "CanonicalizeUnique:0";
  63         latinTypeInfo[characters[0]] = characters[0] <= MAX_LATIN ? "CanonicalizeLatinSelf:0" : "CanonicalizeLatinInvalid:0";
  64         continue;
  65     }
  66
  67     // Sort the array.
  68     characters.sort(function(x,y){return x-y;});
  69
  70     // If there are more than two characters, create an entry in characterSetInfo.
  71     if (characters.length > 2) {
  72         for (i in characters)
  73             typeInfo[characters[i]] = "CanonicalizeSet:" + characterSetInfo.length;
  74         characterSetInfo.push(characters);
  75
  76         if (characters[1] <= MAX_LATIN)
  77             throw new Error("sets with more than one latin character not supported!");
  78         if (characters[0] <= MAX_LATIN) {
  79             for (i in characters)
  80                 latinTypeInfo[characters[i]] = "CanonicalizeLatinOther:" + characters[0];
  81             latinTypeInfo[characters[0]] = "CanonicalizeLatinSelf:0";
  82         } else {
  83             for (i in characters)
  84                 latinTypeInfo[characters[i]] = "CanonicalizeLatinInvalid:0";
  85         }
  86
  87         continue;
  88     }
  89
  90     // We have a pair, mark alternating ranges, otherwise track whether this is the low or high partner.
  91     var lo = characters[0];
  92     var hi = characters[1];
  93     var delta = hi - lo;
  94     if (delta == 1) {
  95         var type = lo & 1 ? "CanonicalizeAlternatingUnaligned:0" : "CanonicalizeAlternatingAligned:0";
  96         typeInfo[lo] = type;
  97         typeInfo[hi] = type;
  98     } else {
  99         typeInfo[lo] = "CanonicalizeRangeLo:" + delta;
 100         typeInfo[hi] = "CanonicalizeRangeHi:" + delta;
 101     }
 102
 103     if (lo > MAX_LATIN) {
 104         latinTypeInfo[lo] = "CanonicalizeLatinInvalid:0";
 105         latinTypeInfo[hi] = "CanonicalizeLatinInvalid:0";
 106     } else if (hi > MAX_LATIN) {
 107         latinTypeInfo[lo] = "CanonicalizeLatinSelf:0";
 108         latinTypeInfo[hi] = "CanonicalizeLatinOther:" + lo;
 109     } else {
 110         if (delta != 0x20 || lo & 0x20)
 111             throw new Error("pairs of latin characters that don't mask with 0x20 not supported!");
 112         latinTypeInfo[lo] = "CanonicalizeLatinMask0x20:0";
 113         latinTypeInfo[hi] = "CanonicalizeLatinMask0x20:0";
 114     }
 115 }
 116
 117 var rangeInfo = [];
 118 // Pass 3: coallesce types into ranges.
 119 for (var end = 0; end <= MAX_UCS2; ++end) {
 120     var begin = end;
 121     var type = typeInfo[end];
 122     while (end < MAX_UCS2 && typeInfo[end + 1] == type)
 123         ++end;
 124     rangeInfo.push({begin:begin, end:end, type:type});
 125 }
 126
 127 var latinRangeInfo = [];
 128 // Pass 4: coallesce latin-1 types into ranges.
 129 for (var end = 0; end <= MAX_UCS2; ++end) {
 130     var begin = end;
 131     var type = latinTypeInfo[end];
 132     while (end < MAX_UCS2 && latinTypeInfo[end + 1] == type)
 133         ++end;
 134     latinRangeInfo.push({begin:begin, end:end, type:type});
 135 }
 136
 137
 138 // Helper function to convert a number to a fixed width hex representation of a C uint16_t.
 139 function hex(x)
 140 {
 141     var s = Number(x).toString(16);
 142     while (s.length < 4)
 143         s = 0 + s;
 144     return "0x" + s + "u";
 145 }
 146
 147 var copyright = (
 148     "/*"                                                                            + "\n" +
 149     " * Copyright (C) 2012 Apple Inc. All rights reserved."                         + "\n" +
 150     " *"                                                                            + "\n" +
 151     " * Redistribution and use in source and binary forms, with or without"         + "\n" +
 152     " * modification, are permitted provided that the following conditions"         + "\n" +
 153     " * are met:"                                                                   + "\n" +
 154     " * 1. Redistributions of source code must retain the above copyright"          + "\n" +
 155     " *    notice, this list of conditions and the following disclaimer."           + "\n" +
 156     " * 2. Redistributions in binary form must reproduce the above copyright"       + "\n" +
 157     " *    notice, this list of conditions and the following disclaimer in the"     + "\n" +
 158     " *    documentation and/or other materials provided with the distribution."    + "\n" +
 159     " *"                                                                            + "\n" +
 160     " * THIS SOFTWARE IS PROVIDED BY APPLE INC. ``AS IS'' AND ANY"                  + "\n" +
 161     " * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE"          + "\n" +
 162     " * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR"         + "\n" +
 163     " * PURPOSE ARE DISCLAIMED.  IN NO EVENT SHALL APPLE INC. OR"                   + "\n" +
 164     " * CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,"      + "\n" +
 165     " * EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,"        + "\n" +
 166     " * PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR"         + "\n" +
 167     " * PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY"        + "\n" +
 168     " * OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT"               + "\n" +
 169     " * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE"      + "\n" +
 170     " * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. "      + "\n" +
 171     " */");
 172
 173 print(copyright);
 174 print();
 175 print("// DO NOT EDIT! - this file autogenerated by YarrCanonicalizeUCS2.js");
 176 print();
 177 print('#include "config.h"');
 178 print('#include "YarrCanonicalizeUCS2.h"');
 179 print();
 180 print("namespace JSC { namespace Yarr {");
 181 print();
 182 print("#include <stdint.h>");
 183 print();
 184
 185 for (i in characterSetInfo) {
 186     var characters = ""
 187     var set = characterSetInfo[i];
 188     for (var j in set)
 189         characters += hex(set[j]) + ", ";
 190     print("uint16_t ucs2CharacterSet" + i + "[] = { " + characters + "0 };");
 191 }
 192 print();
 193 print("static const size_t UCS2_CANONICALIZATION_SETS = " + characterSetInfo.length + ";");
 194 print("uint16_t* characterSetInfo[UCS2_CANONICALIZATION_SETS] = {");
 195 for (i in characterSetInfo)
 196 print("    ucs2CharacterSet" + i + ",");
 197 print("};");
 198 print();
 199 print("const size_t UCS2_CANONICALIZATION_RANGES = " + rangeInfo.length + ";");
 200 print("UCS2CanonicalizationRange rangeInfo[UCS2_CANONICALIZATION_RANGES] = {");
 201 for (i in rangeInfo) {
 202     var info = rangeInfo[i];
 203     var typeAndValue = info.type.split(':');
 204     print("    { " + hex(info.begin) + ", " + hex(info.end) + ", " + hex(typeAndValue[1]) + ", " + typeAndValue[0] + " },");
 205 }
 206 print("};");
 207 print();
 208 print("const size_t LATIN_CANONICALIZATION_RANGES = " + latinRangeInfo.length + ";");
 209 print("LatinCanonicalizationRange latinRangeInfo[LATIN_CANONICALIZATION_RANGES] = {");
 210 for (i in latinRangeInfo) {
 211     var info = latinRangeInfo[i];
 212     var typeAndValue = info.type.split(':');
 213     print("    { " + hex(info.begin) + ", " + hex(info.end) + ", " + hex(typeAndValue[1]) + ", " + typeAndValue[0] + " },");
 214 }
 215 print("};");
 216 print();
 217 print("} } // JSC::Yarr");
 218 print();
 219