2
|
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. Sun designates this
|
|
7 |
* particular file as subject to the "Classpath" exception as provided
|
|
8 |
* by Sun 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 Sun Microsystems, Inc., 4150 Network Circle, Santa Clara,
|
|
21 |
* CA 95054 USA or visit www.sun.com if you need additional information or
|
|
22 |
* have any questions.
|
|
23 |
*
|
|
24 |
*/
|
|
25 |
|
|
26 |
/*
|
|
27 |
*
|
|
28 |
* (C) Copyright IBM Corp. 1998-2005 - All Rights Reserved
|
|
29 |
*
|
|
30 |
*/
|
|
31 |
|
|
32 |
#ifndef __INDICREORDERING_H
|
|
33 |
#define __INDICREORDERING_H
|
|
34 |
|
|
35 |
#include "LETypes.h"
|
|
36 |
#include "OpenTypeTables.h"
|
|
37 |
|
|
38 |
// Characters that get refered to by name...
|
|
39 |
#define C_SIGN_ZWNJ 0x200C
|
|
40 |
#define C_SIGN_ZWJ 0x200D
|
|
41 |
|
|
42 |
// Character class values
|
|
43 |
#define CC_RESERVED 0U
|
|
44 |
#define CC_VOWEL_MODIFIER 1U
|
|
45 |
#define CC_STRESS_MARK 2U
|
|
46 |
#define CC_INDEPENDENT_VOWEL 3U
|
|
47 |
#define CC_INDEPENDENT_VOWEL_2 4U
|
|
48 |
#define CC_INDEPENDENT_VOWEL_3 5U
|
|
49 |
#define CC_CONSONANT 6U
|
|
50 |
#define CC_CONSONANT_WITH_NUKTA 7U
|
|
51 |
#define CC_NUKTA 8U
|
|
52 |
#define CC_DEPENDENT_VOWEL 9U
|
|
53 |
#define CC_SPLIT_VOWEL_PIECE_1 10U
|
|
54 |
#define CC_SPLIT_VOWEL_PIECE_2 11U
|
|
55 |
#define CC_SPLIT_VOWEL_PIECE_3 12U
|
|
56 |
#define CC_VIRAMA 13U
|
|
57 |
#define CC_ZERO_WIDTH_MARK 14U
|
|
58 |
#define CC_COUNT 15U
|
|
59 |
|
|
60 |
// Character class flags
|
|
61 |
#define CF_CLASS_MASK 0x0000FFFFU
|
|
62 |
|
|
63 |
#define CF_CONSONANT 0x80000000U
|
|
64 |
|
|
65 |
#define CF_REPH 0x40000000U
|
|
66 |
#define CF_VATTU 0x20000000U
|
|
67 |
#define CF_BELOW_BASE 0x10000000U
|
|
68 |
#define CF_POST_BASE 0x08000000U
|
|
69 |
#define CF_LENGTH_MARK 0x04000000U
|
|
70 |
|
|
71 |
#define CF_POS_BEFORE 0x00300000U
|
|
72 |
#define CF_POS_BELOW 0x00200000U
|
|
73 |
#define CF_POS_ABOVE 0x00100000U
|
|
74 |
#define CF_POS_AFTER 0x00000000U
|
|
75 |
#define CF_POS_MASK 0x00300000U
|
|
76 |
|
|
77 |
#define CF_INDEX_MASK 0x000F0000U
|
|
78 |
#define CF_INDEX_SHIFT 16
|
|
79 |
|
|
80 |
// Script flag bits
|
|
81 |
#define SF_MATRAS_AFTER_BASE 0x80000000U
|
|
82 |
#define SF_REPH_AFTER_BELOW 0x40000000U
|
|
83 |
#define SF_EYELASH_RA 0x20000000U
|
|
84 |
#define SF_MPRE_FIXUP 0x10000000U
|
|
85 |
|
|
86 |
#define SF_POST_BASE_LIMIT_MASK 0x0000FFFFU
|
|
87 |
#define SF_NO_POST_BASE_LIMIT 0x00007FFFU
|
|
88 |
|
|
89 |
typedef LEUnicode SplitMatra[3];
|
|
90 |
|
|
91 |
class MPreFixups;
|
|
92 |
class LEGlyphStorage;
|
|
93 |
|
|
94 |
struct IndicClassTable
|
|
95 |
{
|
|
96 |
typedef le_uint32 CharClass;
|
|
97 |
typedef le_uint32 ScriptFlags;
|
|
98 |
|
|
99 |
LEUnicode firstChar;
|
|
100 |
LEUnicode lastChar;
|
|
101 |
le_int32 worstCaseExpansion;
|
|
102 |
ScriptFlags scriptFlags;
|
|
103 |
const CharClass *classTable;
|
|
104 |
const SplitMatra *splitMatraTable;
|
|
105 |
|
|
106 |
inline le_int32 getWorstCaseExpansion() const;
|
|
107 |
|
|
108 |
CharClass getCharClass(LEUnicode ch) const;
|
|
109 |
|
|
110 |
inline const SplitMatra *getSplitMatra(CharClass charClass) const;
|
|
111 |
|
|
112 |
inline le_bool isVowelModifier(LEUnicode ch) const;
|
|
113 |
inline le_bool isStressMark(LEUnicode ch) const;
|
|
114 |
inline le_bool isConsonant(LEUnicode ch) const;
|
|
115 |
inline le_bool isReph(LEUnicode ch) const;
|
|
116 |
inline le_bool isVirama(LEUnicode ch) const;
|
|
117 |
inline le_bool isNukta(LEUnicode ch) const;
|
|
118 |
inline le_bool isVattu(LEUnicode ch) const;
|
|
119 |
inline le_bool isMatra(LEUnicode ch) const;
|
|
120 |
inline le_bool isSplitMatra(LEUnicode ch) const;
|
|
121 |
inline le_bool isLengthMark(LEUnicode ch) const;
|
|
122 |
inline le_bool hasPostOrBelowBaseForm(LEUnicode ch) const;
|
|
123 |
inline le_bool hasPostBaseForm(LEUnicode ch) const;
|
|
124 |
inline le_bool hasBelowBaseForm(LEUnicode ch) const;
|
|
125 |
|
|
126 |
inline static le_bool isVowelModifier(CharClass charClass);
|
|
127 |
inline static le_bool isStressMark(CharClass charClass);
|
|
128 |
inline static le_bool isConsonant(CharClass charClass);
|
|
129 |
inline static le_bool isReph(CharClass charClass);
|
|
130 |
inline static le_bool isVirama(CharClass charClass);
|
|
131 |
inline static le_bool isNukta(CharClass charClass);
|
|
132 |
inline static le_bool isVattu(CharClass charClass);
|
|
133 |
inline static le_bool isMatra(CharClass charClass);
|
|
134 |
inline static le_bool isSplitMatra(CharClass charClass);
|
|
135 |
inline static le_bool isLengthMark(CharClass charClass);
|
|
136 |
inline static le_bool hasPostOrBelowBaseForm(CharClass charClass);
|
|
137 |
inline static le_bool hasPostBaseForm(CharClass charClass);
|
|
138 |
inline static le_bool hasBelowBaseForm(CharClass charClass);
|
|
139 |
|
|
140 |
static const IndicClassTable *getScriptClassTable(le_int32 scriptCode);
|
|
141 |
};
|
|
142 |
|
|
143 |
class IndicReordering {
|
|
144 |
public:
|
|
145 |
static le_int32 getWorstCaseExpansion(le_int32 scriptCode);
|
|
146 |
|
|
147 |
static le_int32 reorder(const LEUnicode *theChars, le_int32 charCount, le_int32 scriptCode,
|
|
148 |
LEUnicode *outChars, LEGlyphStorage &glyphStorage,
|
|
149 |
MPreFixups **outMPreFixups);
|
|
150 |
|
|
151 |
static void adjustMPres(MPreFixups *mpreFixups, LEGlyphStorage &glyphStorage);
|
|
152 |
|
|
153 |
static const FeatureMap *getFeatureMap(le_int32 &count);
|
|
154 |
|
|
155 |
private:
|
|
156 |
// do not instantiate
|
|
157 |
IndicReordering();
|
|
158 |
|
|
159 |
static le_int32 findSyllable(const IndicClassTable *classTable, const LEUnicode *chars,
|
|
160 |
le_int32 prev, le_int32 charCount);
|
|
161 |
|
|
162 |
};
|
|
163 |
|
|
164 |
inline le_int32 IndicClassTable::getWorstCaseExpansion() const
|
|
165 |
{
|
|
166 |
return worstCaseExpansion;
|
|
167 |
}
|
|
168 |
|
|
169 |
inline const SplitMatra *IndicClassTable::getSplitMatra(CharClass charClass) const
|
|
170 |
{
|
|
171 |
le_int32 index = (charClass & CF_INDEX_MASK) >> CF_INDEX_SHIFT;
|
|
172 |
|
|
173 |
return &splitMatraTable[index - 1];
|
|
174 |
}
|
|
175 |
|
|
176 |
inline le_bool IndicClassTable::isVowelModifier(CharClass charClass)
|
|
177 |
{
|
|
178 |
return (charClass & CF_CLASS_MASK) == CC_VOWEL_MODIFIER;
|
|
179 |
}
|
|
180 |
|
|
181 |
inline le_bool IndicClassTable::isStressMark(CharClass charClass)
|
|
182 |
{
|
|
183 |
return (charClass & CF_CLASS_MASK) == CC_STRESS_MARK;
|
|
184 |
}
|
|
185 |
|
|
186 |
inline le_bool IndicClassTable::isConsonant(CharClass charClass)
|
|
187 |
{
|
|
188 |
return (charClass & CF_CONSONANT) != 0;
|
|
189 |
}
|
|
190 |
|
|
191 |
inline le_bool IndicClassTable::isReph(CharClass charClass)
|
|
192 |
{
|
|
193 |
return (charClass & CF_REPH) != 0;
|
|
194 |
}
|
|
195 |
|
|
196 |
inline le_bool IndicClassTable::isNukta(CharClass charClass)
|
|
197 |
{
|
|
198 |
return (charClass & CF_CLASS_MASK) == CC_NUKTA;
|
|
199 |
}
|
|
200 |
|
|
201 |
inline le_bool IndicClassTable::isVirama(CharClass charClass)
|
|
202 |
{
|
|
203 |
return (charClass & CF_CLASS_MASK) == CC_VIRAMA;
|
|
204 |
}
|
|
205 |
|
|
206 |
inline le_bool IndicClassTable::isVattu(CharClass charClass)
|
|
207 |
{
|
|
208 |
return (charClass & CF_VATTU) != 0;
|
|
209 |
}
|
|
210 |
|
|
211 |
inline le_bool IndicClassTable::isMatra(CharClass charClass)
|
|
212 |
{
|
|
213 |
charClass &= CF_CLASS_MASK;
|
|
214 |
|
|
215 |
return charClass >= CC_DEPENDENT_VOWEL && charClass <= CC_SPLIT_VOWEL_PIECE_3;
|
|
216 |
}
|
|
217 |
|
|
218 |
inline le_bool IndicClassTable::isSplitMatra(CharClass charClass)
|
|
219 |
{
|
|
220 |
return (charClass & CF_INDEX_MASK) != 0;
|
|
221 |
}
|
|
222 |
|
|
223 |
inline le_bool IndicClassTable::isLengthMark(CharClass charClass)
|
|
224 |
{
|
|
225 |
return (charClass & CF_LENGTH_MARK) != 0;
|
|
226 |
}
|
|
227 |
|
|
228 |
inline le_bool IndicClassTable::hasPostOrBelowBaseForm(CharClass charClass)
|
|
229 |
{
|
|
230 |
return (charClass & (CF_POST_BASE | CF_BELOW_BASE)) != 0;
|
|
231 |
}
|
|
232 |
|
|
233 |
inline le_bool IndicClassTable::hasPostBaseForm(CharClass charClass)
|
|
234 |
{
|
|
235 |
return (charClass & CF_POST_BASE) != 0;
|
|
236 |
}
|
|
237 |
|
|
238 |
inline le_bool IndicClassTable::hasBelowBaseForm(CharClass charClass)
|
|
239 |
{
|
|
240 |
return (charClass & CF_BELOW_BASE) != 0;
|
|
241 |
}
|
|
242 |
|
|
243 |
inline le_bool IndicClassTable::isVowelModifier(LEUnicode ch) const
|
|
244 |
{
|
|
245 |
return isVowelModifier(getCharClass(ch));
|
|
246 |
}
|
|
247 |
|
|
248 |
inline le_bool IndicClassTable::isStressMark(LEUnicode ch) const
|
|
249 |
{
|
|
250 |
return isStressMark(getCharClass(ch));
|
|
251 |
}
|
|
252 |
|
|
253 |
inline le_bool IndicClassTable::isConsonant(LEUnicode ch) const
|
|
254 |
{
|
|
255 |
return isConsonant(getCharClass(ch));
|
|
256 |
}
|
|
257 |
|
|
258 |
inline le_bool IndicClassTable::isReph(LEUnicode ch) const
|
|
259 |
{
|
|
260 |
return isReph(getCharClass(ch));
|
|
261 |
}
|
|
262 |
|
|
263 |
inline le_bool IndicClassTable::isVirama(LEUnicode ch) const
|
|
264 |
{
|
|
265 |
return isVirama(getCharClass(ch));
|
|
266 |
}
|
|
267 |
|
|
268 |
inline le_bool IndicClassTable::isNukta(LEUnicode ch) const
|
|
269 |
{
|
|
270 |
return isNukta(getCharClass(ch));
|
|
271 |
}
|
|
272 |
|
|
273 |
inline le_bool IndicClassTable::isVattu(LEUnicode ch) const
|
|
274 |
{
|
|
275 |
return isVattu(getCharClass(ch));
|
|
276 |
}
|
|
277 |
|
|
278 |
inline le_bool IndicClassTable::isMatra(LEUnicode ch) const
|
|
279 |
{
|
|
280 |
return isMatra(getCharClass(ch));
|
|
281 |
}
|
|
282 |
|
|
283 |
inline le_bool IndicClassTable::isSplitMatra(LEUnicode ch) const
|
|
284 |
{
|
|
285 |
return isSplitMatra(getCharClass(ch));
|
|
286 |
}
|
|
287 |
|
|
288 |
inline le_bool IndicClassTable::isLengthMark(LEUnicode ch) const
|
|
289 |
{
|
|
290 |
return isLengthMark(getCharClass(ch));
|
|
291 |
}
|
|
292 |
|
|
293 |
inline le_bool IndicClassTable::hasPostOrBelowBaseForm(LEUnicode ch) const
|
|
294 |
{
|
|
295 |
return hasPostOrBelowBaseForm(getCharClass(ch));
|
|
296 |
}
|
|
297 |
|
|
298 |
inline le_bool IndicClassTable::hasPostBaseForm(LEUnicode ch) const
|
|
299 |
{
|
|
300 |
return hasPostBaseForm(getCharClass(ch));
|
|
301 |
}
|
|
302 |
|
|
303 |
inline le_bool IndicClassTable::hasBelowBaseForm(LEUnicode ch) const
|
|
304 |
{
|
|
305 |
return hasBelowBaseForm(getCharClass(ch));
|
|
306 |
}
|
|
307 |
|
|
308 |
#endif
|