2 *******************************************************************************
3 * Copyright (C) 2008-2012, International Business Machines Corporation and *
4 * others. All Rights Reserved. *
5 *******************************************************************************
7 package com.ibm.icu.charset;
9 import java.nio.ByteBuffer;
10 import java.nio.CharBuffer;
11 import java.nio.IntBuffer;
12 import java.nio.charset.CharsetDecoder;
13 import java.nio.charset.CharsetEncoder;
14 import java.nio.charset.CoderResult;
16 import com.ibm.icu.text.UTF16;
17 import com.ibm.icu.text.UnicodeSet;
23 class CharsetISCII extends CharsetICU {
24 private static final short UCNV_OPTIONS_VERSION_MASK = 0X0f;
25 //private static final short NUKTA = 0x093c;
26 //private static final short HALANT = 0x094d;
27 private static final short ZWNJ = 0x200c; /* Zero Width Non Joiner */
28 private static final short ZWJ = 0x200d; /* Zero Width Joiner */
29 //private static final int INVALID_CHAR = 0xffff;
30 private static final short ATR = 0xef; /* Attribute code */
31 private static final short EXT = 0xf0; /* Extension code */
32 private static final short DANDA = 0x0964;
33 private static final short DOUBLE_DANDA = 0x0965;
34 private static final short ISCII_NUKTA = 0xe9;
35 private static final short ISCII_HALANT = 0xe8;
36 private static final short ISCII_DANDA = 0xea;
37 private static final short ISCII_VOWEL_SIGN_E = 0xe0;
38 private static final short ISCII_INV = 0xd9;
39 private static final short INDIC_BLOCK_BEGIN = 0x0900;
40 private static final short INDIC_BLOCK_END = 0x0d7f;
41 private static final short INDIC_RANGE = (INDIC_BLOCK_END - INDIC_BLOCK_BEGIN);
42 private static final short VOCALLIC_RR = 0x0931;
43 private static final short LF = 0x0a;
44 private static final short ASCII_END = 0xa0;
45 private static final short TELUGU_DELTA = (UniLang.DELTA * UniLang.TELUGU);
46 private static final short DEV_ABBR_SIGN = 0x0970;
47 private static final short DEV_ANUDATTA = 0x0952;
48 private static final short EXT_RANGE_BEGIN = 0xa1;
49 private static final short EXT_RANGE_END = 0xee;
50 private static final short PNJ_DELTA = 0x100;
51 private static final int NO_CHAR_MARKER = 0xfffe;
53 /* Used for proper conversion to and from Gurmukhi */
54 private static UnicodeSet PNJ_BINDI_TIPPI_SET;
55 private static UnicodeSet PNJ_CONSONANT_SET;
56 private static final short PNJ_BINDI = 0x0a02;
57 private static final short PNJ_TIPPI = 0x0a70;
58 private static final short PNJ_SIGN_VIRAMA = 0x0a4d;
59 private static final short PNJ_ADHAK = 0x0a71;
60 private static final short PNJ_HA = 0x0a39;
61 private static final short PNJ_RRA = 0x0a5c;
63 private static final class UniLang {
64 static final short DEVALANGARI = 0;
65 static final short BENGALI = DEVALANGARI + 1;
66 static final short GURMUKHI = BENGALI + 1;
67 static final short GUJARATI = GURMUKHI + 1;
68 static final short ORIYA = GUJARATI + 1;
69 static final short TAMIL = ORIYA + 1;
70 static final short TELUGU = TAMIL + 1;
71 static final short KANNADA = TELUGU + 1;
72 static final short MALAYALAM = KANNADA + 1;
73 static final short DELTA = 0x80;
75 @SuppressWarnings("unused")
76 private static final class ISCIILang {
77 static final short DEF = 0x40;
78 static final short RMN = 0x41;
79 static final short DEV = 0x42;
80 static final short BNG = 0x43;
81 static final short TML = 0x44;
82 static final short TLG = 0x45;
83 static final short ASM = 0x46;
84 static final short ORI = 0x47;
85 static final short KND = 0x48;
86 static final short MLM = 0x49;
87 static final short GJR = 0x4a;
88 static final short PNJ = 0x4b;
89 static final short ARB = 0x71;
90 static final short PES = 0x72;
91 static final short URD = 0x73;
92 static final short SND = 0x74;
93 static final short KSM = 0x75;
94 static final short PST = 0x76;
97 private static final class MaskEnum {
98 static final short DEV_MASK = 0x80;
99 static final short PNJ_MASK = 0x40;
100 static final short GJR_MASK = 0x20;
101 static final short ORI_MASK = 0x10;
102 static final short BNG_MASK = 0x08;
103 static final short KND_MASK = 0x04;
104 static final short MLM_MASK = 0x02;
105 static final short TML_MASK = 0x01;
106 static final short ZERO = 0x00;
109 private final static String ISCII_CNV_PREFIX = "ISCII,version=";
111 @SuppressWarnings("unused")
112 private static final class UConverterDataISCII {
114 int contextCharToUnicode; /* previous Unicode codepoint for contextual analysis */
115 int contextCharFromUnicode; /* previous Unicode codepoint for contextual analysis */
116 short defDeltaToUnicode; /* delta for switching to default state when DEF is encountered */
117 short currentDeltaFromUnicode; /* current delta in Indic block */
118 short currentDeltaToUnicode; /* current delta in Indic block */
119 short currentMaskFromUnicode; /* mask for current state in fromUnicode */
120 short currentMaskToUnicode; /* mask for current state in toUnicode */
121 short defMaskToUnicode; /* mask for default state in toUnicode */
122 boolean isFirstBuffer; /* boolean for fromUnicode to see if we need to announce the first script */
123 boolean resetToDefaultToUnicode; /* boolean for reseting to default delta and mask when a newline is encountered */
125 int prevToUnicodeStatus; /* Hold the previous toUnicodeStatus. This is necessary because we may need to know the last two code points. */
127 UConverterDataISCII(int option, String name) {
128 this.option = option;
135 this.contextCharToUnicode = NO_CHAR_MARKER; /* contextCharToUnicode */
136 this.currentDeltaFromUnicode = 0x0000; /* contextCharFromUnicode */
137 this.defDeltaToUnicode = (short)(lookupInitialData[option & UCNV_OPTIONS_VERSION_MASK].uniLang * UniLang.DELTA); /* defDeltaToUnicode */
138 this.currentDeltaFromUnicode = (short)(lookupInitialData[option & UCNV_OPTIONS_VERSION_MASK].uniLang * UniLang.DELTA); /* currentDeltaFromUnicode */
139 this.currentDeltaToUnicode = (short)(lookupInitialData[option & UCNV_OPTIONS_VERSION_MASK].uniLang * UniLang.DELTA); /* currentDeltaToUnicode */
140 this.currentMaskToUnicode = lookupInitialData[option & UCNV_OPTIONS_VERSION_MASK].maskEnum; /* currentMaskToUnicode */
141 this.currentMaskFromUnicode = lookupInitialData[option & UCNV_OPTIONS_VERSION_MASK].maskEnum; /* currentMaskFromUnicode */
142 this.defMaskToUnicode = lookupInitialData[option & UCNV_OPTIONS_VERSION_MASK].maskEnum; /* defMaskToUnicode */
143 this.isFirstBuffer = true; /* isFirstBuffer */
144 this.resetToDefaultToUnicode = false; /* resetToDefaultToUnicode */
145 this.prevToUnicodeStatus = 0x0000;
149 private static final class LookupDataStruct {
154 LookupDataStruct(short uniLang, short maskEnum, short isciiLang) {
155 this.uniLang = uniLang;
156 this.maskEnum = maskEnum;
157 this.isciiLang = isciiLang;
161 private static final LookupDataStruct [] lookupInitialData = {
162 new LookupDataStruct(UniLang.DEVALANGARI, MaskEnum.DEV_MASK, ISCIILang.DEV),
163 new LookupDataStruct(UniLang.BENGALI, MaskEnum.BNG_MASK, ISCIILang.BNG),
164 new LookupDataStruct(UniLang.GURMUKHI, MaskEnum.PNJ_MASK, ISCIILang.PNJ),
165 new LookupDataStruct(UniLang.GUJARATI, MaskEnum.GJR_MASK, ISCIILang.GJR),
166 new LookupDataStruct(UniLang.ORIYA, MaskEnum.ORI_MASK, ISCIILang.ORI),
167 new LookupDataStruct(UniLang.TAMIL, MaskEnum.TML_MASK, ISCIILang.TML),
168 new LookupDataStruct(UniLang.TELUGU, MaskEnum.KND_MASK, ISCIILang.TLG),
169 new LookupDataStruct(UniLang.KANNADA, MaskEnum.KND_MASK, ISCIILang.KND),
170 new LookupDataStruct(UniLang.MALAYALAM, MaskEnum.MLM_MASK, ISCIILang.MLM)
174 * The values in validity table are indexed by the lower bits of Unicode
175 * range 0x0900 - 0x09ff. The values have a structure like:
176 * -----------------------------------------------------------------
177 * |DEV | PNJ | GJR | ORI | BNG | TLG | MLM | TML |
178 * | | | | | ASM | KND | | |
179 * -----------------------------------------------------------------
180 * If a code point is valid in a particular script
181 * then that bit is turned on
183 * Unicode does not distinguish between Bengali and Assamese aso we use 1 bit for
184 * to represent these languages
186 * Telugu and Kannda have same codepoints except for Vocallic_RR which we special case
187 * and combine and use 1 bit to represent these languages
189 private static final short validityTable[] = {
190 /* This state table is tool generated so please do not edit unless you know exactly what you are doing */
191 /* Note: This table was edited to mirror the Windows XP implementation */
192 /* ISCII: Valid: Unicode */
193 /* 0xa0: 0x00: 0x900 */ MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO,
194 /* 0xa1: 0xb8: 0x901 */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO,
195 /* 0xa2: 0xfe: 0x902 */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.TML_MASK,
196 /* 0xa3: 0xbf: 0x903 */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.TML_MASK,
197 /* 0x00: 0x00: 0x904 */ MaskEnum.DEV_MASK + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO,
198 /* 0xa4: 0xff: 0x905 */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.TML_MASK,
199 /* 0xa5: 0xff: 0x906 */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.TML_MASK,
200 /* 0xa6: 0xff: 0x907 */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.TML_MASK,
201 /* 0xa7: 0xff: 0x908 */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.TML_MASK,
202 /* 0xa8: 0xff: 0x909 */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.TML_MASK,
203 /* 0xa9: 0xff: 0x90a */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.TML_MASK,
204 /* 0xaa: 0xfe: 0x90b */ MaskEnum.DEV_MASK + MaskEnum.ZERO + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.ZERO,
205 /* 0x00: 0x00: 0x90c */ MaskEnum.DEV_MASK + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.ZERO,
206 /* 0xae: 0x80: 0x90d */ MaskEnum.DEV_MASK + MaskEnum.ZERO + MaskEnum.GJR_MASK + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO,
207 /* 0xab: 0x87: 0x90e */ MaskEnum.DEV_MASK + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.TML_MASK,
208 /* 0xac: 0xff: 0x90f */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.TML_MASK,
209 /* 0xad: 0xff: 0x910 */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.TML_MASK,
210 /* 0xb2: 0x80: 0x911 */ MaskEnum.DEV_MASK + MaskEnum.ZERO + MaskEnum.GJR_MASK + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO,
211 /* 0xaf: 0x87: 0x912 */ MaskEnum.DEV_MASK + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.TML_MASK,
212 /* 0xb0: 0xff: 0x913 */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.TML_MASK,
213 /* 0xb1: 0xff: 0x914 */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.TML_MASK,
214 /* 0xb3: 0xff: 0x915 */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.TML_MASK,
215 /* 0xb4: 0xfe: 0x916 */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.ZERO,
216 /* 0xb5: 0xfe: 0x917 */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.ZERO,
217 /* 0xb6: 0xfe: 0x918 */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.ZERO,
218 /* 0xb7: 0xff: 0x919 */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.TML_MASK,
219 /* 0xb8: 0xff: 0x91a */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.TML_MASK,
220 /* 0xb9: 0xfe: 0x91b */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.ZERO,
221 /* 0xba: 0xff: 0x91c */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.TML_MASK,
222 /* 0xbb: 0xfe: 0x91d */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.ZERO,
223 /* 0xbc: 0xff: 0x91e */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.TML_MASK,
224 /* 0xbd: 0xff: 0x91f */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.TML_MASK,
225 /* 0xbe: 0xfe: 0x920 */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.ZERO,
226 /* 0xbf: 0xfe: 0x921 */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.ZERO,
227 /* 0xc0: 0xfe: 0x922 */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.ZERO,
228 /* 0xc1: 0xff: 0x923 */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.TML_MASK,
229 /* 0xc2: 0xff: 0x924 */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.TML_MASK,
230 /* 0xc3: 0xfe: 0x925 */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.ZERO,
231 /* 0xc4: 0xfe: 0x926 */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.ZERO,
232 /* 0xc5: 0xfe: 0x927 */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.ZERO,
233 /* 0xc6: 0xff: 0x928 */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.TML_MASK,
234 /* 0xc7: 0x81: 0x929 */ MaskEnum.DEV_MASK + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.TML_MASK,
235 /* 0xc8: 0xff: 0x92a */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.TML_MASK,
236 /* 0xc9: 0xfe: 0x92b */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.ZERO,
237 /* 0xca: 0xfe: 0x92c */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.ZERO,
238 /* 0xcb: 0xfe: 0x92d */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.ZERO,
239 /* 0xcc: 0xfe: 0x92e */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.TML_MASK,
240 /* 0xcd: 0xff: 0x92f */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.TML_MASK,
241 /* 0xcf: 0xff: 0x930 */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.TML_MASK,
242 /* 0xd0: 0x87: 0x931 */ MaskEnum.DEV_MASK + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.MLM_MASK + MaskEnum.TML_MASK,
243 /* 0xd1: 0xff: 0x932 */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.TML_MASK,
244 /* 0xd2: 0xb7: 0x933 */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.ZERO + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.TML_MASK,
245 /* 0xd3: 0x83: 0x934 */ MaskEnum.DEV_MASK + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.MLM_MASK + MaskEnum.TML_MASK,
246 /* 0xd4: 0xff: 0x935 */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.ZERO + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.TML_MASK,
247 /* 0xd5: 0xfe: 0x936 */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.ZERO,
248 /* 0xd6: 0xbf: 0x937 */ MaskEnum.DEV_MASK + MaskEnum.ZERO + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.TML_MASK,
249 /* 0xd7: 0xff: 0x938 */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.TML_MASK,
250 /* 0xd8: 0xff: 0x939 */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.TML_MASK,
251 /* 0x00: 0x00: 0x93a */ MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO,
252 /* 0x00: 0x00: 0x93b */ MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO,
253 /* 0xe9: 0xda: 0x93c */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.ZERO + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO,
254 /* 0x00: 0x00: 0x93d */ MaskEnum.DEV_MASK + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO,
255 /* 0xda: 0xff: 0x93e */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.TML_MASK,
256 /* 0xdb: 0xff: 0x93f */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.TML_MASK,
257 /* 0xdc: 0xff: 0x940 */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.TML_MASK,
258 /* 0xdd: 0xff: 0x941 */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.TML_MASK,
259 /* 0xde: 0xff: 0x942 */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.TML_MASK,
260 /* 0xdf: 0xbe: 0x943 */ MaskEnum.DEV_MASK + MaskEnum.ZERO + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.ZERO,
261 /* 0x00: 0x00: 0x944 */ MaskEnum.DEV_MASK + MaskEnum.ZERO + MaskEnum.GJR_MASK + MaskEnum.ZERO + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.ZERO + MaskEnum.ZERO,
262 /* 0xe3: 0x80: 0x945 */ MaskEnum.DEV_MASK + MaskEnum.ZERO + MaskEnum.GJR_MASK + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO,
263 /* 0xe0: 0x87: 0x946 */ MaskEnum.DEV_MASK + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.TML_MASK,
264 /* 0xe1: 0xff: 0x947 */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.TML_MASK,
265 /* 0xe2: 0xff: 0x948 */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.TML_MASK,
266 /* 0xe7: 0x80: 0x949 */ MaskEnum.DEV_MASK + MaskEnum.ZERO + MaskEnum.GJR_MASK + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO,
267 /* 0xe4: 0x87: 0x94a */ MaskEnum.DEV_MASK + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.TML_MASK,
268 /* 0xe5: 0xff: 0x94b */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.TML_MASK,
269 /* 0xe6: 0xff: 0x94c */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.TML_MASK,
270 /* 0xe8: 0xff: 0x94d */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.TML_MASK,
271 /* 0xec: 0x00: 0x94e */ MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO,
272 /* 0xed: 0x00: 0x94f */ MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO,
273 /* 0x00: 0x00: 0x950 */ MaskEnum.DEV_MASK + MaskEnum.ZERO + MaskEnum.GJR_MASK + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO,
274 /* 0x00: 0x00: 0x951 */ MaskEnum.DEV_MASK + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO,
275 /* 0x00: 0x00: 0x952 */ MaskEnum.DEV_MASK + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO,
276 /* 0x00: 0x00: 0x953 */ MaskEnum.DEV_MASK + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO,
277 /* 0x00: 0x00: 0x954 */ MaskEnum.DEV_MASK + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO,
278 /* 0x00: 0x00: 0x955 */ MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.KND_MASK + MaskEnum.ZERO + MaskEnum.ZERO,
279 /* 0x00: 0x00: 0x956 */ MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ORI_MASK + MaskEnum.ZERO + MaskEnum.KND_MASK + MaskEnum.ZERO + MaskEnum.ZERO,
280 /* 0x00: 0x00: 0x957 */ MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.ZERO + MaskEnum.MLM_MASK + MaskEnum.ZERO,
281 /* 0x00: 0x00: 0x958 */ MaskEnum.DEV_MASK + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO,
282 /* 0x00: 0x00: 0x959 */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO,
283 /* 0x00: 0x00: 0x95a */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO,
284 /* 0x00: 0x00: 0x95b */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO,
285 /* 0x00: 0x00: 0x95c */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.BNG_MASK + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO,
286 /* 0x00: 0x00: 0x95d */ MaskEnum.DEV_MASK + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO,
287 /* 0x00: 0x00: 0x95e */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO,
288 /* 0xce: 0x98: 0x95f */ MaskEnum.DEV_MASK + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO,
289 /* 0x00: 0x00: 0x960 */ MaskEnum.DEV_MASK + MaskEnum.ZERO + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.ZERO,
290 /* 0x00: 0x00: 0x961 */ MaskEnum.DEV_MASK + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.ZERO,
291 /* 0x00: 0x00: 0x962 */ MaskEnum.DEV_MASK + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.BNG_MASK + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO,
292 /* 0x00: 0x00: 0x963 */ MaskEnum.DEV_MASK + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.BNG_MASK + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO,
293 /* 0xea: 0xf8: 0x964 */ MaskEnum.DEV_MASK + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO,
294 /* 0xeaea: 0x00: 0x965 */ MaskEnum.DEV_MASK + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO,
295 /* 0xf1: 0xff: 0x966 */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.TML_MASK,
296 /* 0xf2: 0xff: 0x967 */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.TML_MASK,
297 /* 0xf3: 0xff: 0x968 */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.TML_MASK,
298 /* 0xf4: 0xff: 0x969 */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.TML_MASK,
299 /* 0xf5: 0xff: 0x96a */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.TML_MASK,
300 /* 0xf6: 0xff: 0x96b */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.TML_MASK,
301 /* 0xf7: 0xff: 0x96c */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.TML_MASK,
302 /* 0xf8: 0xff: 0x96d */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.TML_MASK,
303 /* 0xf9: 0xff: 0x96e */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.TML_MASK,
304 /* 0xfa: 0xff: 0x96f */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.GJR_MASK + MaskEnum.ORI_MASK + MaskEnum.BNG_MASK + MaskEnum.KND_MASK + MaskEnum.MLM_MASK + MaskEnum.TML_MASK,
305 /* 0x00: 0x80: 0x970 */ MaskEnum.DEV_MASK + MaskEnum.PNJ_MASK + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO + MaskEnum.ZERO,
308 * The length of the array is 128 to provide values for 0x900..0x97f.
309 * The last 15 entries for 0x971..0x97f of the table are all zero
310 * because no Indic script uses such Unicode code points.
313 /* 0x00: 0x00: 0x971 */ MaskEnum.ZERO,
314 /* 0x00: 0x00: 0x972 */ MaskEnum.ZERO,
315 /* 0x00: 0x00: 0x973 */ MaskEnum.ZERO,
316 /* 0x00: 0x00: 0x974 */ MaskEnum.ZERO,
317 /* 0x00: 0x00: 0x975 */ MaskEnum.ZERO,
318 /* 0x00: 0x00: 0x976 */ MaskEnum.ZERO,
319 /* 0x00: 0x00: 0x977 */ MaskEnum.ZERO,
320 /* 0x00: 0x00: 0x978 */ MaskEnum.ZERO,
321 /* 0x00: 0x00: 0x979 */ MaskEnum.ZERO,
322 /* 0x00: 0x00: 0x97A */ MaskEnum.ZERO,
323 /* 0x00: 0x00: 0x97B */ MaskEnum.ZERO,
324 /* 0x00: 0x00: 0x97C */ MaskEnum.ZERO,
325 /* 0x00: 0x00: 0x97D */ MaskEnum.ZERO,
326 /* 0x00: 0x00: 0x97E */ MaskEnum.ZERO,
327 /* 0x00: 0x00: 0x97F */ MaskEnum.ZERO,
330 private static final char fromUnicodeTable[] = {
411 0xA1E9, /* 0x0950 */ /* OM Symbol */
460 private static final char toUnicodeTable[] = {
718 private static final char nuktaSpecialCases[][] = {
719 { 16 /* length of array */ , 0 },
736 private static final char vowelSignESpecialCases[][] = {
737 { 2 /* length of array */ , 0 },
741 private static final short lookupTable[][] = {
742 { MaskEnum.ZERO, MaskEnum.ZERO }, /* DEFAULT */
743 { MaskEnum.ZERO, MaskEnum.ZERO }, /* ROMAN */
744 { UniLang.DEVALANGARI, MaskEnum.DEV_MASK },
745 { UniLang.BENGALI, MaskEnum.BNG_MASK },
746 { UniLang.TAMIL, MaskEnum.TML_MASK },
747 { UniLang.TELUGU, MaskEnum.KND_MASK },
748 { UniLang.BENGALI, MaskEnum.BNG_MASK },
749 { UniLang.ORIYA, MaskEnum.ORI_MASK },
750 { UniLang.KANNADA, MaskEnum.KND_MASK },
751 { UniLang.MALAYALAM, MaskEnum.MLM_MASK },
752 { UniLang.GUJARATI, MaskEnum.GJR_MASK },
753 { UniLang.GURMUKHI, MaskEnum.PNJ_MASK }
756 private UConverterDataISCII extraInfo = null;
757 protected byte[] fromUSubstitution = new byte[]{(byte)0x1A};
759 public CharsetISCII(String icuCanonicalName, String javaCanonicalName, String[] aliases) {
760 super(icuCanonicalName, javaCanonicalName, aliases);
764 //get the version number of the ISCII converter
765 int option = Integer.parseInt(icuCanonicalName.substring(14));
767 extraInfo = new UConverterDataISCII(
769 ISCII_CNV_PREFIX + (option & UCNV_OPTIONS_VERSION_MASK) /* name */
775 /* Initialize the two UnicodeSets use for proper Gurmukhi conversion if they have not already been created. */
776 private void initializePNJSets() {
777 if (PNJ_BINDI_TIPPI_SET != null && PNJ_CONSONANT_SET != null) {
780 PNJ_BINDI_TIPPI_SET = new UnicodeSet();
781 PNJ_CONSONANT_SET = new UnicodeSet();
783 PNJ_CONSONANT_SET.add(0x0a15, 0x0a28);
784 PNJ_CONSONANT_SET.add(0x0a2a, 0x0a30);
785 PNJ_CONSONANT_SET.add(0x0a35, 0x0a36);
786 PNJ_CONSONANT_SET.add(0x0a38, 0x0a39);
788 PNJ_BINDI_TIPPI_SET.addAll(PNJ_CONSONANT_SET);
789 PNJ_BINDI_TIPPI_SET.add(0x0a05);
790 PNJ_BINDI_TIPPI_SET.add(0x0a07);
792 PNJ_BINDI_TIPPI_SET.add(0x0a41, 0x0a42);
793 PNJ_BINDI_TIPPI_SET.add(0x0a3f);
795 PNJ_CONSONANT_SET.compact();
796 PNJ_BINDI_TIPPI_SET.compact();
800 * Rules for ISCII to Unicode converter
801 * ISCII is a stateful encoding. To convert ISCII bytes to Unicode,
802 * which is both precomposed and decomposed from characters
803 * pre-context and post-context need to be considered.
806 * i) ATR : Attribute code is used to declare the font and script switching.
807 * Currently we only switch scripts and font codes consumed without generating an error
808 * ii) EXT : Extention code is used to declare switching to Sanskrit and for obscure,
809 * obsolete characters
811 * i) Halant: if preceeded by a halant then it is a explicit halant
813 * a) if preceeded by a halant then it is a soft halant
814 * b) if preceeded by specific consonants and the ligatures have pre-composed
815 * characters in Unicode then convert to pre-composed characters
816 * iii) Danda: If Danda is preceeded by a Danda then convert to Double Danda
818 class CharsetDecoderISCII extends CharsetDecoderICU {
819 public CharsetDecoderISCII(CharsetICU cs) {
824 protected void implReset() {
826 this.toUnicodeStatus = 0xFFFF;
827 extraInfo.initialize();
830 @SuppressWarnings("fallthrough")
831 protected CoderResult decodeLoop(ByteBuffer source, CharBuffer target, IntBuffer offsets, boolean flush) {
832 CoderResult cr = CoderResult.UNDERFLOW;
833 int targetUniChar = 0x0000;
834 short sourceChar = 0x0000;
835 UConverterDataISCII data;
836 boolean gotoCallBack = false;
840 //data.contextCharToUnicode; /* contains previous ISCII codepoint visited */
841 //this.toUnicodeStatus; /* contains the mapping to Unicode of the above codepoint */
843 while (source.hasRemaining()) {
844 targetUniChar = UConverterConstants.missingCharMarker;
846 if (target.hasRemaining()) {
847 sourceChar = (short)((short)source.get() & UConverterConstants.UNSIGNED_BYTE_MASK);
849 /* look at the post-context perform special processing */
850 if (data.contextCharToUnicode == ATR) {
851 /* If we have ATR in data.contextCharToUnicode then we need to change our
852 * state to Indic Script specified by sourceChar
854 /* check if the sourceChar is supported script range */
855 if (((short)(ISCIILang.PNJ - sourceChar) & UConverterConstants.UNSIGNED_BYTE_MASK) <= (ISCIILang.PNJ - ISCIILang.DEV)) {
856 data.currentDeltaToUnicode = (short)(lookupTable[sourceChar & 0x0F][0] * UniLang.DELTA);
857 data.currentMaskToUnicode = lookupTable[sourceChar & 0x0F][1];
858 } else if (sourceChar == ISCIILang.DEF) {
859 /* switch back to default */
860 data.currentDeltaToUnicode = data.defDeltaToUnicode;
861 data.currentMaskToUnicode = data.defMaskToUnicode;
863 if ((sourceChar >= 0x21 && sourceChar <= 0x3F)) {
864 /* these are display codes consume and continue */
866 cr = CoderResult.malformedForLength(1);
868 data.contextCharToUnicode = NO_CHAR_MARKER;
874 data.contextCharToUnicode = NO_CHAR_MARKER;
877 } else if (data.contextCharToUnicode == EXT) {
878 /* check if sourceChar is in 0xA1 - 0xEE range */
879 if (((short)(EXT_RANGE_END - sourceChar) & UConverterConstants.UNSIGNED_BYTE_MASK) <= (EXT_RANGE_END - EXT_RANGE_BEGIN)) {
880 /* We currently support only Anudatta and Devanagari abbreviation sign */
881 if (sourceChar == 0xBF || sourceChar == 0xB8) {
882 targetUniChar = (sourceChar == 0xBF) ? DEV_ABBR_SIGN : DEV_ANUDATTA;
884 /* find out if the mappling is valid in this state */
885 if ((validityTable[((short)targetUniChar) & UConverterConstants.UNSIGNED_BYTE_MASK] & data.currentMaskToUnicode) > 0) {
886 data.contextCharToUnicode = NO_CHAR_MARKER;
888 /* Write the previous toUnicodeStatus, this was delayed to handle consonant clustering for Gurmukhi script. */
889 if (data.prevToUnicodeStatus != 0) {
890 cr = WriteToTargetToU(offsets, (source.position() - 1), source, target, data.prevToUnicodeStatus, (short)0);
891 data.prevToUnicodeStatus = 0x0000;
893 /* write to target */
894 cr = WriteToTargetToU(offsets, (source.position() - 2), source, target, targetUniChar, data.currentDeltaToUnicode);
899 /* byte unit is unassigned */
900 targetUniChar = UConverterConstants.missingCharMarker;
901 cr = CoderResult.unmappableForLength(1);
903 /* only 0xA1 - 0xEE are legal after EXT char */
904 data.contextCharToUnicode = NO_CHAR_MARKER;
905 cr = CoderResult.malformedForLength(1);
908 } else if (data.contextCharToUnicode == ISCII_INV) {
909 if (sourceChar == ISCII_HALANT) {
910 targetUniChar = 0x0020; /* replace with space according to Indic FAQ */
915 /* Write the previous toUnicodeStatus, this was delayed to handle consonant clustering for Gurmukhi script. */
916 if (data.prevToUnicodeStatus != 0) {
917 cr = WriteToTargetToU(offsets, (source.position() - 1), source, target, data.prevToUnicodeStatus, (short)0);
918 data.prevToUnicodeStatus = 0x0000;
921 /* write to target */
922 cr = WriteToTargetToU(offsets, (source.position() - 2), source, target, targetUniChar, data.currentDeltaToUnicode);
924 data.contextCharToUnicode = NO_CHAR_MARKER;
927 /* look at the pre-context and perform special processing */
929 switch (sourceChar) {
931 case EXT: /* falls through */
933 data.contextCharToUnicode = (char)sourceChar;
935 if (this.toUnicodeStatus != UConverterConstants.missingCharMarker) {
936 /* Write the previous toUnicodeStatus, this was delayed to handle consonant clustering for Gurmukhi script. */
937 if (data.prevToUnicodeStatus != 0) {
938 cr = WriteToTargetToU(offsets, (source.position() - 1), source, target, data.prevToUnicodeStatus, (short)0);
939 data.prevToUnicodeStatus = 0x0000;
941 cr = WriteToTargetToU(offsets, (source.position() - 2), source, target, this.toUnicodeStatus, data.currentDeltaToUnicode);
942 this.toUnicodeStatus = UConverterConstants.missingCharMarker;
946 /* handle double danda */
947 if (data.contextCharToUnicode == ISCII_DANDA) {
948 targetUniChar = DOUBLE_DANDA;
949 /* clear the context */
950 data.contextCharToUnicode = NO_CHAR_MARKER;
951 this.toUnicodeStatus = UConverterConstants.missingCharMarker;
953 targetUniChar = GetMapping(sourceChar, targetUniChar, data);
954 data.contextCharToUnicode = (char)sourceChar;
958 /* handle explicit halant */
959 if (data.contextCharToUnicode == ISCII_HALANT) {
960 targetUniChar = ZWNJ;
962 data.contextCharToUnicode = NO_CHAR_MARKER;
964 targetUniChar = GetMapping(sourceChar, targetUniChar, data);
965 data.contextCharToUnicode = (char)sourceChar;
971 data.resetToDefaultToUnicode = true;
972 targetUniChar = GetMapping(sourceChar, targetUniChar, data);
973 data.contextCharToUnicode = (char)sourceChar;
975 case ISCII_VOWEL_SIGN_E:
976 /* find <CHAR> + SIGN_VOWEL_E special mapping */
978 boolean find = false;
979 for (; n < vowelSignESpecialCases[0][0]; n++) {
980 if (vowelSignESpecialCases[n][0] == ((short)data.contextCharToUnicode & UConverterConstants.UNSIGNED_BYTE_MASK)) {
981 targetUniChar = vowelSignESpecialCases[n][1];
987 /* find out if the mapping is valid in this state */
988 if ((validityTable[(byte)targetUniChar] & data.currentMaskFromUnicode) > 0) {
989 data.contextCharToUnicode = NO_CHAR_MARKER;
990 this.toUnicodeStatus = UConverterConstants.missingCharMarker;
994 targetUniChar = GetMapping(sourceChar, targetUniChar, data);
995 data.contextCharToUnicode = (char)sourceChar;
998 /* handle soft halant */
999 if (data.contextCharToUnicode == ISCII_HALANT) {
1000 targetUniChar = ZWJ;
1001 /* clear the context */
1002 data.contextCharToUnicode = NO_CHAR_MARKER;
1004 } else if (data.currentDeltaToUnicode == PNJ_DELTA && data.contextCharToUnicode == 0xc0) {
1005 /* We got here because ISCII_NUKTA was preceded by 0xc0 and we are converting Gurmukhi.
1006 * In that case we must convert (0xc0 0xe9) to (\u0a5c\u0a4d\u0a39).
1007 * WriteToTargetToU is given 0x095c instead of 0xa5c because that method will automatically
1008 * convert the code point given based on the delta provided.
1010 cr = WriteToTargetToU(offsets, (source.position() - 2), source, target, PNJ_RRA, (short)0);
1011 if (!cr.isOverflow()) {
1012 cr = WriteToTargetToU(offsets, (source.position() - 2), source, target, PNJ_SIGN_VIRAMA, (short)0);
1013 if (!cr.isOverflow()) {
1014 cr = WriteToTargetToU(offsets, (source.position() - 2), source, target, PNJ_HA, (short)0);
1016 this.charErrorBufferArray[this.charErrorBufferLength++] = PNJ_HA;
1019 this.charErrorBufferArray[this.charErrorBufferLength++] = PNJ_SIGN_VIRAMA;
1020 this.charErrorBufferArray[this.charErrorBufferLength++] = PNJ_HA;
1022 this.toUnicodeStatus = UConverterConstants.missingCharMarker;
1023 data.contextCharToUnicode = NO_CHAR_MARKER;
1024 if (!cr.isError()) {
1029 /* try to handle <CHAR> + ISCII_NUKTA special mappings */
1031 boolean found = false;
1032 for (; i < nuktaSpecialCases[0][0]; i++) {
1033 if (nuktaSpecialCases[i][0] == ((short)data.contextCharToUnicode & UConverterConstants.UNSIGNED_BYTE_MASK)) {
1034 targetUniChar = nuktaSpecialCases[i][1];
1040 /* find out if the mapping is valid in this state */
1041 if ((validityTable[(byte)targetUniChar] & data.currentMaskToUnicode) > 0) {
1042 data.contextCharToUnicode = NO_CHAR_MARKER;
1043 this.toUnicodeStatus = UConverterConstants.missingCharMarker;
1044 if (data.currentDeltaToUnicode == PNJ_DELTA) {
1045 /* Write the previous toUnicodeStatus, this was delayed to handle consonant clustering for Gurmukhi script. */
1046 if (data.prevToUnicodeStatus != 0) {
1047 cr = WriteToTargetToU(offsets, (source.position() - 1), source, target, data.prevToUnicodeStatus, (short)0);
1048 data.prevToUnicodeStatus = 0x0000;
1050 cr = WriteToTargetToU(offsets, (source.position() - 2), source, target, targetUniChar, data.currentDeltaToUnicode);
1055 /* else fall through to default */
1057 /* else fall through to default */
1061 targetUniChar = GetMapping(sourceChar, targetUniChar, data);
1062 data.contextCharToUnicode = (char)sourceChar;
1065 }//end of CallBack if statement
1067 if (!gotoCallBack && this.toUnicodeStatus != UConverterConstants.missingCharMarker) {
1068 /* Check to make sure that consonant clusters are handled correctly for Gurmukhi script. */
1069 if (data.currentDeltaToUnicode == PNJ_DELTA && data.prevToUnicodeStatus != 0 && PNJ_CONSONANT_SET.contains(data.prevToUnicodeStatus) &&
1070 (this.toUnicodeStatus + PNJ_DELTA) == PNJ_SIGN_VIRAMA && (targetUniChar + PNJ_DELTA) == data.prevToUnicodeStatus) {
1071 if (offsets != null) {
1072 offset = source.position() - 3;
1074 cr = WriteToTargetToU(offsets, offset, source, target, PNJ_ADHAK, (short)0);
1075 cr = WriteToTargetToU(offsets, offset, source, target, data.prevToUnicodeStatus, (short)0);
1076 data.prevToUnicodeStatus = 0x0000; /* reset the previous unicode code point */
1077 toUnicodeStatus = UConverterConstants.missingCharMarker;
1080 /* Write the previous toUnicodeStatus, this was delayed to handle consonant clustering for Gurmukhi script. */
1081 if (data.prevToUnicodeStatus != 0) {
1082 cr = WriteToTargetToU(offsets, (source.position() - 1), source, target, data.prevToUnicodeStatus, (short)0);
1083 data.prevToUnicodeStatus = 0x0000;
1085 /* Check to make sure that Bindi and Tippi are handled correctly for Gurmukhi script.
1086 * If 0xA2 is preceded by a codepoint in the PNJ_BINDI_TIPPI_SET then the target codepoint should be Tippi instead of Bindi.
1088 if (data.currentDeltaToUnicode == PNJ_DELTA && (targetUniChar + PNJ_DELTA) == PNJ_BINDI && PNJ_BINDI_TIPPI_SET.contains(this.toUnicodeStatus + PNJ_DELTA)) {
1089 targetUniChar = PNJ_TIPPI - PNJ_DELTA;
1090 cr = WriteToTargetToU(offsets, (source.position() - 2), source, target, this.toUnicodeStatus, PNJ_DELTA);
1091 } else if (data.currentDeltaToUnicode == PNJ_DELTA && (targetUniChar + PNJ_DELTA) == PNJ_SIGN_VIRAMA && PNJ_CONSONANT_SET.contains(this.toUnicodeStatus + PNJ_DELTA)) {
1092 /* Store the current toUnicodeStatus code point for later handling of consonant cluster in Gurmukhi. */
1093 data.prevToUnicodeStatus = this.toUnicodeStatus + PNJ_DELTA;
1095 /* write the previously mapped codepoint */
1096 cr = WriteToTargetToU(offsets, (source.position() - 2), source, target, this.toUnicodeStatus, data.currentDeltaToUnicode);
1099 this.toUnicodeStatus = UConverterConstants.missingCharMarker;
1102 if (!gotoCallBack && targetUniChar != UConverterConstants.missingCharMarker) {
1103 /* now save the targetUniChar for delayed write */
1104 this.toUnicodeStatus = (char)targetUniChar;
1105 if (data.resetToDefaultToUnicode) {
1106 data.currentDeltaToUnicode = data.defDeltaToUnicode;
1107 data.currentMaskToUnicode = data.defMaskToUnicode;
1108 data.resetToDefaultToUnicode = false;
1111 /* we reach here only if targetUniChar == missingCharMarker
1112 * so assign codes to reason and err
1114 if (!gotoCallBack) {
1115 cr = CoderResult.unmappableForLength(1);
1118 toUBytesArray[0] = (byte)sourceChar;
1120 gotoCallBack = false;
1124 cr = CoderResult.OVERFLOW;
1130 if (cr.isUnderflow() && flush && !source.hasRemaining()) {
1131 /*end of the input stream */
1132 if (data.contextCharToUnicode == ATR || data.contextCharToUnicode == EXT || data.contextCharToUnicode == ISCII_INV) {
1133 /* set toUBytes[] */
1134 toUBytesArray[0] = (byte)data.contextCharToUnicode;
1137 /* avoid looping on truncated sequences */
1138 data.contextCharToUnicode = NO_CHAR_MARKER;
1143 if (this.toUnicodeStatus != UConverterConstants.missingCharMarker) {
1144 /* output a remaining target character */
1145 WriteToTargetToU(offsets, (source.position() - 2), source, target, this.toUnicodeStatus, data.currentDeltaToUnicode);
1146 this.toUnicodeStatus = UConverterConstants.missingCharMarker;
1152 private CoderResult WriteToTargetToU(IntBuffer offsets, int offset, ByteBuffer source, CharBuffer target, int targetUniChar, short delta) {
1153 CoderResult cr = CoderResult.UNDERFLOW;
1154 /* add offset to current Indic Block */
1155 if (targetUniChar > ASCII_END &&
1156 targetUniChar != ZWJ &&
1157 targetUniChar != ZWNJ &&
1158 targetUniChar != DANDA &&
1159 targetUniChar != DOUBLE_DANDA) {
1160 targetUniChar += delta;
1163 /* now write the targetUniChar */
1164 if (target.hasRemaining()) {
1165 target.put((char)targetUniChar);
1166 if (offsets != null) {
1167 offsets.put(offset);
1170 charErrorBufferArray[charErrorBufferLength++] = (char)targetUniChar;
1171 cr = CoderResult.OVERFLOW;
1176 private int GetMapping(short sourceChar, int targetUniChar, UConverterDataISCII data) {
1177 targetUniChar = toUnicodeTable[sourceChar];
1178 /* is the code point valid in current script? */
1179 if (sourceChar > ASCII_END &&
1180 (validityTable[targetUniChar & 0x7F] & data.currentMaskToUnicode) == 0) {
1181 /* Vocallic RR is assigne in ISCII Telugu and Unicode */
1182 if (data.currentDeltaToUnicode != (TELUGU_DELTA) || targetUniChar != VOCALLIC_RR) {
1183 targetUniChar = UConverterConstants.missingCharMarker;
1186 return targetUniChar;
1197 class CharsetEncoderISCII extends CharsetEncoderICU {
1198 public CharsetEncoderISCII(CharsetICU cs) {
1199 super(cs, fromUSubstitution);
1203 protected void implReset() {
1205 extraInfo.initialize();
1208 protected CoderResult encodeLoop(CharBuffer source, ByteBuffer target, IntBuffer offsets, boolean flush) {
1209 int targetByteUnit = 0x0000;
1210 int sourceChar = 0x0000;
1211 UConverterDataISCII converterData;
1214 boolean deltaChanged = false;
1215 int tempContextFromUnicode = 0x0000; /* For special handling of the Gurmukhi script. */
1216 CoderResult cr = CoderResult.UNDERFLOW;
1218 /* initialize data */
1219 converterData = extraInfo;
1220 newDelta = converterData.currentDeltaFromUnicode;
1221 range = (short)(newDelta / UniLang.DELTA);
1223 if ((sourceChar = fromUChar32) != 0) {
1224 cr = handleSurrogates(source, (char) sourceChar);
1225 return (cr != null) ? cr : CoderResult.unmappableForLength(2);
1228 /* writing the char to the output stream */
1229 while (source.hasRemaining()) {
1230 if (!target.hasRemaining()) {
1231 return CoderResult.OVERFLOW;
1234 /* Write the language code following LF only if LF is not the last character. */
1235 if (fromUnicodeStatus == LF) {
1236 targetByteUnit = ATR << 8;
1237 targetByteUnit += 0xff & (byte)lookupInitialData[range].isciiLang;
1238 fromUnicodeStatus = 0x0000;
1239 /* now append ATR and language code */
1240 cr = WriteToTargetFromU(offsets, source, target, targetByteUnit);
1241 if (cr.isOverflow()) {
1246 sourceChar = source.get();
1247 tempContextFromUnicode = converterData.contextCharFromUnicode;
1249 targetByteUnit = UConverterConstants.missingCharMarker;
1251 /* check if input is in ASCII and C0 control codes range */
1252 if (sourceChar <= ASCII_END) {
1253 fromUnicodeStatus = sourceChar;
1254 cr = WriteToTargetFromU(offsets, source, target, sourceChar);
1255 if (cr.isOverflow()) {
1261 switch (sourceChar) {
1263 /* contextChar has HALANT */
1264 if (converterData.contextCharFromUnicode != 0) {
1265 converterData.contextCharFromUnicode = 0x00;
1266 targetByteUnit = ISCII_HALANT;
1268 /* consume ZWNJ and continue */
1269 converterData.contextCharFromUnicode = 0x00;
1274 /* contextChar has HALANT */
1275 if (converterData.contextCharFromUnicode != 0) {
1276 targetByteUnit = ISCII_NUKTA;
1278 targetByteUnit = ISCII_INV;
1280 converterData.contextCharFromUnicode = 0x00;
1283 /* is the sourceChar in the INDIC_RANGE? */
1284 if((char)(INDIC_BLOCK_END - sourceChar) <= INDIC_RANGE) {
1285 /* Danda and Doube Danda are valid in Northern scripts.. since Unicode
1286 * does not include these codepoints in all Northern scripts we need to
1289 if (sourceChar != DANDA && sourceChar != DOUBLE_DANDA) {
1290 /* find out to which block the sourceChar belongs */
1291 range = (short)((sourceChar - INDIC_BLOCK_BEGIN) / UniLang.DELTA);
1292 newDelta = (short)(range * UniLang.DELTA);
1294 /* Now are we in the same block as previous? */
1295 if (newDelta != converterData.currentDeltaFromUnicode || converterData.isFirstBuffer) {
1296 converterData.currentDeltaFromUnicode = newDelta;
1297 converterData.currentMaskFromUnicode = lookupInitialData[range].maskEnum;
1298 deltaChanged = true;
1299 converterData.isFirstBuffer = false;
1301 if (converterData.currentDeltaFromUnicode == PNJ_DELTA) {
1302 if (sourceChar == PNJ_TIPPI) {
1303 /* Make sure Tippi is converterd to Bindi. */
1304 sourceChar = PNJ_BINDI;
1305 } else if (sourceChar == PNJ_ADHAK) {
1306 /* This is for consonant cluster handling. */
1307 converterData.contextCharFromUnicode = PNJ_ADHAK;
1310 /* Normalize all Indic codepoints to Devanagari and map them to ISCII */
1311 /* now subtract the new delta from sourceChar */
1312 sourceChar -= converterData.currentDeltaFromUnicode;
1314 /* get the target byte unit */
1315 targetByteUnit = fromUnicodeTable[(short)sourceChar & UConverterConstants.UNSIGNED_BYTE_MASK];
1317 /* is the code point valid in current script? */
1318 if ((validityTable[(short)sourceChar & UConverterConstants.UNSIGNED_BYTE_MASK] & converterData.currentMaskFromUnicode) == 0) {
1319 /* Vocallic RR is assigned in ISCII Telugu and Unicode */
1320 if (converterData.currentDeltaFromUnicode != (TELUGU_DELTA) || sourceChar != VOCALLIC_RR) {
1321 targetByteUnit = UConverterConstants.missingCharMarker;
1326 /* we are in a script block which is different than
1327 * previous sourceChar's script block write ATR and language codes
1330 temp = (char)(ATR << 8);
1331 temp += (char)(lookupInitialData[range].isciiLang & UConverterConstants.UNSIGNED_BYTE_MASK);
1333 deltaChanged = false;
1334 /* now append ATR and language code */
1335 cr = WriteToTargetFromU(offsets, source, target, temp);
1336 if (cr.isOverflow()) {
1340 if (converterData.currentDeltaFromUnicode == PNJ_DELTA && (sourceChar + PNJ_DELTA) == PNJ_ADHAK) {
1344 /* reset context char */
1345 converterData.contextCharFromUnicode = 0x00;
1348 if (converterData.currentDeltaFromUnicode == PNJ_DELTA && tempContextFromUnicode == PNJ_ADHAK && PNJ_CONSONANT_SET.contains(sourceChar + PNJ_DELTA)) {
1349 /* If the previous codepoint is Adhak and the current codepoint is a consonant, the targetByteUnit should be C + Halant + C. */
1350 /* reset context char */
1351 converterData.contextCharFromUnicode = 0x0000;
1352 targetByteUnit = targetByteUnit << 16 | ISCII_HALANT << 8 | targetByteUnit;
1353 /*write targetByteUnit to target */
1354 cr = WriteToTargetFromU(offsets, source, target, targetByteUnit);
1355 if (cr.isOverflow()) {
1358 } else if (targetByteUnit != UConverterConstants.missingCharMarker) {
1359 if (targetByteUnit == ISCII_HALANT) {
1360 converterData.contextCharFromUnicode = (char)targetByteUnit;
1362 /*write targetByteUnit to target */
1363 cr = WriteToTargetFromU(offsets, source, target, targetByteUnit);
1364 if (cr.isOverflow()) {
1367 } else if (UTF16.isSurrogate((char)sourceChar)) {
1368 cr = handleSurrogates(source, (char) sourceChar);
1369 return (cr != null) ? cr : CoderResult.unmappableForLength(2);
1371 return CoderResult.unmappableForLength(1);
1373 } /* end of while */
1375 /* save the state and return */
1379 private CoderResult WriteToTargetFromU(IntBuffer offsets, CharBuffer source, ByteBuffer target, int targetByteUnit) {
1380 CoderResult cr = CoderResult.UNDERFLOW;
1381 int offset = source.position() - 1;
1382 /* write the targetUniChar to target */
1383 if (target.hasRemaining()) {
1384 if (targetByteUnit <= 0xFF) {
1385 target.put((byte)targetByteUnit);
1386 if (offsets != null) {
1387 offsets.put(offset);
1390 if (targetByteUnit > 0xFFFF) {
1391 target.put((byte)(targetByteUnit >> 16));
1392 if (offsets != null) {
1394 offsets.put(offset);
1397 if (!target.hasRemaining()) {
1398 errorBuffer[errorBufferLength++] = (byte)(targetByteUnit >> 8);
1399 errorBuffer[errorBufferLength++] = (byte)targetByteUnit;
1400 cr = CoderResult.OVERFLOW;
1403 target.put((byte)(targetByteUnit >> 8));
1404 if (offsets != null) {
1405 offsets.put(offset);
1407 if (target.hasRemaining()) {
1408 target.put((byte)targetByteUnit);
1409 if (offsets != null) {
1410 offsets.put(offset);
1413 errorBuffer[errorBufferLength++] = (byte)targetByteUnit;
1414 cr = CoderResult.OVERFLOW;
1418 if ((targetByteUnit > 0xFFFF)) {
1419 errorBuffer[errorBufferLength++] = (byte)(targetByteUnit >> 16);
1420 } else if ((targetByteUnit & 0xFF00) > 0) {
1421 errorBuffer[errorBufferLength++] = (byte)(targetByteUnit >> 8);
1423 errorBuffer[errorBufferLength++] = (byte)(targetByteUnit);
1424 cr = CoderResult.OVERFLOW;
1430 public CharsetDecoder newDecoder() {
1431 return new CharsetDecoderISCII(this);
1434 public CharsetEncoder newEncoder() {
1435 return new CharsetEncoderISCII(this);
1438 void getUnicodeSetImpl( UnicodeSet setFillIn, int which){
1442 setFillIn.add(0,ASCII_END );
1443 for(script = UniLang.DEVALANGARI ; script<= UniLang.MALAYALAM ;script++){
1444 mask = (char)lookupInitialData[script].maskEnum;
1445 for(idx=0; idx < UniLang.DELTA ; idx++){
1446 // Special check for telugu character
1447 if((validityTable[idx] & mask)!=0 || (script == UniLang.TELUGU && idx==0x31)){
1448 setFillIn.add(idx+(script*UniLang.DELTA)+INDIC_BLOCK_BEGIN );
1452 setFillIn.add(DANDA);
1453 setFillIn.add(DOUBLE_DANDA);
1454 setFillIn.add(ZWNJ);