1 /*
   2  * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
   3  *
   4  * This code is free software; you can redistribute it and/or modify it
   5  * under the terms of the GNU General Public License version 2 only, as
   6  * published by the Free Software Foundation.  Oracle designates this
   7  * particular file as subject to the "Classpath" exception as provided
   8  * by Oracle in the LICENSE file that accompanied this code.
   9  *
  10  * This code is distributed in the hope that it will be useful, but WITHOUT
  11  * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
  12  * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
  13  * version 2 for more details (a copy is included in the LICENSE file that
  14  * accompanied this code).
  15  *
  16  * You should have received a copy of the GNU General Public License version
  17  * 2 along with this work; if not, write to the Free Software Foundation,
  18  * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.
  19  *
  20  * Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA
  21  * or visit www.oracle.com if you need additional information or have any
  22  * questions.
  23  *
  24  */
  25 
  26 /*
  27  *
  28  * (C) Copyright IBM Corp. 1998-2004 - All Rights Reserved
  29  *
  30  */
  31 
  32 #include "LETypes.h"
  33 #include "LEGlyphFilter.h"
  34 #include "OpenTypeTables.h"
  35 #include "LEGlyphStorage.h"
  36 #include "ThaiShaping.h"
  37 
  38 U_NAMESPACE_BEGIN
  39 
  40 enum {
  41     CH_SPACE        = 0x0020,
  42     CH_YAMAKKAN     = 0x0E4E,
  43     CH_MAI_HANAKAT  = 0x0E31,
  44     CH_SARA_AA      = 0x0E32,
  45     CH_SARA_AM      = 0x0E33,
  46     CH_SARA_UEE     = 0x0E37,
  47     CH_MAITAIKHU    = 0x0E47,
  48     CH_NIKHAHIT     = 0x0E4D,
  49     CH_SARA_U       = 0x0E38,
  50     CH_PHINTHU      = 0x0E3A,
  51     CH_YO_YING      = 0x0E0D,
  52     CH_THO_THAN     = 0x0E10,
  53     CH_DOTTED_CIRCLE = 0x25CC
  54 };
  55 
  56     le_uint8 ThaiShaping::getCharClass(LEUnicode ch)
  57 {
  58     le_uint8 charClass = NON;
  59 
  60     if (ch >= 0x0E00 && ch <= 0x0E5B) {
  61         charClass = classTable[ch - 0x0E00];
  62     }
  63 
  64     return charClass;
  65 }
  66 
  67 
  68 LEUnicode ThaiShaping::leftAboveVowel(LEUnicode vowel, le_uint8 glyphSet)
  69 {
  70     static const LEUnicode leftAboveVowels[][7] = {
  71         {0x0E61, 0x0E32, 0x0E33, 0x0E64, 0x0E65, 0x0E66, 0x0E67},
  72         {0xF710, 0x0E32, 0x0E33, 0xF701, 0xF702, 0xF703, 0xF704},
  73         {0xF884, 0x0E32, 0x0E33, 0xF885, 0xF886, 0xF887, 0xF788},
  74         {0x0E31, 0x0E32, 0x0E33, 0x0E34, 0x0E35, 0x0E36, 0x0E37}
  75     };
  76 
  77     if (vowel >= CH_MAI_HANAKAT && vowel <= CH_SARA_UEE) {
  78         return leftAboveVowels[glyphSet][vowel - CH_MAI_HANAKAT];
  79     }
  80 
  81     if (vowel == CH_YAMAKKAN && glyphSet == 0) {
  82         return 0x0E7E;
  83     }
  84 
  85     return vowel;
  86 }
  87 
  88 LEUnicode ThaiShaping::lowerRightTone(LEUnicode tone, le_uint8 glyphSet)
  89 {
  90     static const LEUnicode lowerRightTones[][7] = {
  91         {0x0E68, 0x0E69, 0x0E6A, 0x0E6B, 0x0E6C, 0x0E6D, 0x0E6E},
  92         {0x0E47, 0xF70A, 0xF70B, 0xF70C, 0xF70D, 0xF70E, 0x0E4D},
  93         {0x0E47, 0xF88B, 0xF88E, 0xF891, 0xF894, 0xF897, 0x0E4D},
  94         {0x0E47, 0x0E48, 0x0E49, 0x0E4A, 0x0E4B, 0x0E4C, 0x0E4D}
  95     };
  96 
  97     if (tone >= CH_MAITAIKHU && tone <= CH_NIKHAHIT) {
  98         return lowerRightTones[glyphSet][tone - CH_MAITAIKHU];
  99     }
 100 
 101     return tone;
 102 }
 103 
 104 LEUnicode ThaiShaping::lowerLeftTone(LEUnicode tone, le_uint8 glyphSet)
 105 {
 106     static const LEUnicode lowerLeftTones[][7] = {
 107         {0x0E76, 0x0E77, 0x0E78, 0x0E79, 0x0E7A, 0x0E7B, 0x0E7C},
 108         {0xF712, 0xF705, 0xF706, 0xF707, 0xF708, 0xF709, 0xF711},
 109         {0xF889, 0xF88C, 0xF88F, 0xF892, 0xF895, 0xF898, 0xF899},
 110         {0x0E47, 0x0E48, 0x0E49, 0x0E4A, 0x0E4B, 0x0E4C, 0x0E4D}
 111     };
 112 
 113     if (tone >= CH_MAITAIKHU && tone <= CH_NIKHAHIT) {
 114         return lowerLeftTones[glyphSet][tone - CH_MAITAIKHU];
 115     }
 116 
 117     return tone;
 118 }
 119 
 120 LEUnicode ThaiShaping::upperLeftTone(LEUnicode tone, le_uint8 glyphSet)
 121 {
 122     static const LEUnicode upperLeftTones[][7] = {
 123         {0x0E6F, 0x0E70, 0x0E71, 0x0E72, 0x0E73, 0x0E74, 0x0E75},
 124         {0xF712, 0xF713, 0xF714, 0xF715, 0xF716, 0xF717, 0xF711},
 125         {0xF889, 0xF88A, 0xF88D, 0xF890, 0xF893, 0xF896, 0xF899},
 126         {0x0E47, 0x0E48, 0x0E49, 0x0E4A, 0x0E4B, 0x0E4C, 0x0E4D}
 127     };
 128 
 129     if (tone >= CH_MAITAIKHU && tone <= CH_NIKHAHIT) {
 130         return upperLeftTones[glyphSet][tone - CH_MAITAIKHU];
 131     }
 132 
 133     return tone;
 134 }
 135 
 136 LEUnicode ThaiShaping::lowerBelowVowel(LEUnicode vowel, le_uint8 glyphSet)
 137 {
 138     static const LEUnicode lowerBelowVowels[][3] = {
 139         {0x0E3C, 0x0E3D, 0x0E3E},
 140         {0xF718, 0xF719, 0xF71A},
 141         {0x0E38, 0x0E39, 0x0E3A},
 142         {0x0E38, 0x0E39, 0x0E3A}
 143 
 144     };
 145 
 146     if (vowel >= CH_SARA_U && vowel <= CH_PHINTHU) {
 147         return lowerBelowVowels[glyphSet][vowel - CH_SARA_U];
 148     }
 149 
 150     return vowel;
 151 }
 152 
 153 LEUnicode ThaiShaping::noDescenderCOD(LEUnicode cod, le_uint8 glyphSet)
 154 {
 155     static const LEUnicode noDescenderCODs[][4] = {
 156         {0x0E60, 0x0E0E, 0x0E0F, 0x0E63},
 157         {0xF70F, 0x0E0E, 0x0E0F, 0xF700},
 158         {0x0E0D, 0x0E0E, 0x0E0F, 0x0E10},
 159         {0x0E0D, 0x0E0E, 0x0E0F, 0x0E10}
 160 
 161     };
 162 
 163     if (cod >= CH_YO_YING && cod <= CH_THO_THAN) {
 164         return noDescenderCODs[glyphSet][cod - CH_YO_YING];
 165     }
 166 
 167     return cod;
 168 }
 169 
 170 le_uint8 ThaiShaping::doTransition (StateTransition transition, LEUnicode currChar, le_int32 inputIndex, le_uint8 glyphSet,
 171         LEUnicode errorChar, LEUnicode *outputBuffer, LEGlyphStorage &glyphStorage, le_int32 &outputIndex)
 172 {
 173     LEErrorCode success = LE_NO_ERROR;
 174 
 175     switch (transition.action) {
 176     case tA:
 177         glyphStorage.setCharIndex(outputIndex, inputIndex, success);
 178         outputBuffer[outputIndex++] = currChar;
 179         break;
 180 
 181     case tC:
 182         glyphStorage.setCharIndex(outputIndex, inputIndex, success);
 183         outputBuffer[outputIndex++] = currChar;
 184         break;
 185 
 186     case tD:
 187         glyphStorage.setCharIndex(outputIndex, inputIndex, success);
 188         outputBuffer[outputIndex++] = leftAboveVowel(currChar, glyphSet);
 189         break;
 190 
 191     case tE:
 192         glyphStorage.setCharIndex(outputIndex, inputIndex, success);
 193         outputBuffer[outputIndex++] = lowerRightTone(currChar, glyphSet);
 194         break;
 195 
 196     case tF:
 197         glyphStorage.setCharIndex(outputIndex, inputIndex, success);
 198         outputBuffer[outputIndex++] = lowerLeftTone(currChar, glyphSet);
 199         break;
 200 
 201     case tG:
 202         glyphStorage.setCharIndex(outputIndex, inputIndex, success);
 203         outputBuffer[outputIndex++] = upperLeftTone(currChar, glyphSet);
 204         break;
 205 
 206     case tH:
 207     {
 208         LEUnicode cod = outputBuffer[outputIndex - 1];
 209         LEUnicode coa = noDescenderCOD(cod, glyphSet);
 210 
 211         if (cod != coa) {
 212             outputBuffer[outputIndex - 1] = coa;
 213 
 214             glyphStorage.setCharIndex(outputIndex, inputIndex, success);
 215             outputBuffer[outputIndex++] = currChar;
 216             break;
 217         }
 218 
 219         glyphStorage.setCharIndex(outputIndex, inputIndex, success);
 220         outputBuffer[outputIndex++] = lowerBelowVowel(currChar, glyphSet);
 221         break;
 222     }
 223 
 224     case tR:
 225         glyphStorage.setCharIndex(outputIndex, inputIndex, success);
 226         outputBuffer[outputIndex++] = errorChar;
 227 
 228         glyphStorage.setCharIndex(outputIndex, inputIndex, success);
 229         outputBuffer[outputIndex++] = currChar;
 230         break;
 231 
 232     case tS:
 233         if (currChar == CH_SARA_AM) {
 234             glyphStorage.setCharIndex(outputIndex, inputIndex, success);
 235             outputBuffer[outputIndex++] = errorChar;
 236         }
 237 
 238         glyphStorage.setCharIndex(outputIndex, inputIndex, success);
 239         outputBuffer[outputIndex++] = currChar;
 240         break;
 241 
 242     default:
 243         // FIXME: if we get here, there's an error
 244         // in the state table!
 245         glyphStorage.setCharIndex(outputIndex, inputIndex, success);
 246         outputBuffer[outputIndex++] = currChar;
 247         break;
 248      }
 249 
 250      return transition.nextState;
 251 }
 252 
 253 le_uint8 ThaiShaping::getNextState(LEUnicode ch, le_uint8 prevState, le_int32 inputIndex, le_uint8 glyphSet, LEUnicode errorChar,
 254                               le_uint8 &charClass, LEUnicode *output, LEGlyphStorage &glyphStorage, le_int32 &outputIndex)
 255 {
 256     StateTransition transition;
 257 
 258     charClass = getCharClass(ch);
 259     transition = getTransition(prevState, charClass);
 260 
 261     return doTransition(transition, ch, inputIndex, glyphSet, errorChar, output, glyphStorage, outputIndex);
 262 }
 263 
 264 le_bool ThaiShaping::isLegalHere(LEUnicode ch, le_uint8 prevState)
 265 {
 266     le_uint8 charClass = getCharClass(ch);
 267     StateTransition transition = getTransition(prevState, charClass);
 268 
 269     switch (transition.action) {
 270     case tA:
 271     case tC:
 272     case tD:
 273     case tE:
 274     case tF:
 275     case tG:
 276     case tH:
 277         return TRUE;
 278 
 279     case tR:
 280     case tS:
 281         return FALSE;
 282 
 283     default:
 284         // FIXME: if we get here, there's an error
 285         // in the state table!
 286         return FALSE;
 287     }
 288 }
 289 
 290 le_int32 ThaiShaping::compose(const LEUnicode *input, le_int32 offset, le_int32 charCount, le_uint8 glyphSet,
 291                           LEUnicode errorChar, LEUnicode *output, LEGlyphStorage &glyphStorage)
 292 {
 293     le_uint8 state = 0;
 294     le_int32 inputIndex;
 295     le_int32 outputIndex = 0;
 296     le_uint8 conState = 0xFF;
 297     le_int32 conInput = -1;
 298     le_int32 conOutput = -1;
 299 
 300     for (inputIndex = 0; inputIndex < charCount; inputIndex += 1) {
 301         LEUnicode ch = input[inputIndex + offset];
 302         le_uint8 charClass;
 303 
 304         // Decompose SARA AM into NIKHAHIT + SARA AA
 305         if (ch == CH_SARA_AM && isLegalHere(ch, state) && conState < stateCount) {
 306             outputIndex = conOutput;
 307             state = getNextState(CH_NIKHAHIT, conState, inputIndex, glyphSet, errorChar, charClass,
 308                 output, glyphStorage, outputIndex);
 309 
 310             for (int j = conInput + 1; j < inputIndex; j += 1) {
 311                 ch = input[j + offset];
 312                 state = getNextState(ch, state, j, glyphSet, errorChar, charClass,
 313                     output, glyphStorage, outputIndex);
 314             }
 315 
 316             ch = CH_SARA_AA;
 317         }
 318 
 319         state = getNextState(ch, state, inputIndex, glyphSet, errorChar, charClass,
 320             output, glyphStorage, outputIndex);
 321 
 322         if (charClass >= CON && charClass <= COD) {
 323             conState = state;
 324             conInput = inputIndex;
 325             conOutput = outputIndex;
 326         }
 327     }
 328 
 329     return outputIndex;
 330 }
 331 
 332 U_NAMESPACE_END