1 /* charset.h -- header file for the charset module. 2 Copyright (C) 2003, 2004, 2005, 2006, 2007, 2008, 2009, 2010, 2011, 2012 3 National Institute of Advanced Industrial Science and Technology (AIST) 4 Registration Number H15PRO112 5 6 This file is part of the m17n library. 7 8 The m17n library is free software; you can redistribute it and/or 9 modify it under the terms of the GNU Lesser General Public License 10 as published by the Free Software Foundation; either version 2.1 of 11 the License, or (at your option) any later version. 12 13 The m17n library is distributed in the hope that it will be useful, 14 but WITHOUT ANY WARRANTY; without even the implied warranty of 15 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU 16 Lesser General Public License for more details. 17 18 You should have received a copy of the GNU Lesser General Public 19 License along with the m17n library; if not, write to the Free 20 Software Foundation, Inc., 51 Franklin Street, Fifth Floor, 21 Boston, MA 02110-1301 USA. */ 22 23 #ifndef _M17N_CHARSET_H_ 24 #define _M17N_CHARSET_H_ 25 26 /** @file charset.h 27 @brief Header for charset handlers. 28 */ 29 30 enum mcharset_method 31 { 32 MCHARSET_METHOD_OFFSET, 33 MCHARSET_METHOD_MAP, 34 MCHARSET_METHOD_DEFERRED, 35 MCHARSET_METHOD_SUBSET, 36 MCHARSET_METHOD_SUPERSET, 37 MCHARSET_METHOD_MAX 38 }; 39 40 /** Structure for charset. */ 41 42 typedef struct MCharset MCharset; 43 44 struct MCharset 45 { 46 /** The value is always 0 because all charsets are static. */ 47 unsigned ref_count; 48 49 /** Symbol indicating the name of the charset. */ 50 MSymbol name; 51 52 /** Number of dimensions of the charset. It must be 1, 2, 3, or 53 4. */ 54 int dimension; 55 56 /** Byte code range of each dimension. <code_range>[4N] is a 57 minimum byte code of the (N+1)th dimension, <code_range>[4N+1] 58 is a maximum byte code of the (N+1)th dimension, 59 <code_range>[4N+2] is (<code_range>[4N+1] - <code_range>[4N] + 60 1), <code_range>[4N+3] is a number of characters contained in the 61 first to (N+1)th dimensions. We get "char-index" of a 62 "code-point" from this information. */ 63 int code_range[16]; 64 65 /** The minimum code-point calculated from <code_range>. It may be 66 smaller than <min_code>. */ 67 int code_range_min_code; 68 69 /** Nonzero means there is no gap in code points of the charset. If 70 <dimension> is 1, <no_code_gap> is always 1. Otherwise, 71 <no_code_gap> is 1 iff <code_range>[4N] is zero and 72 <code_range>[4N+1] is 256 for N = 0..<dimension>-2. If 73 <no_code_gap> is nonzero, "char-index" is "code-point" - 74 <min_code>. */ 75 int no_code_gap; 76 77 /** If the byte code B is valid in the (N+1)th dimension, 78 (<code_range_mask>[B] & (1 << N)) is 1. Otherwise, 79 (<code_range_mask>[B] & (1 << N)) is 0. */ 80 unsigned char code_range_mask[256]; 81 82 /** Minimum and maximum code-point of the charset. */ 83 unsigned min_code, max_code; 84 85 /** Nonzero means the charset encodes ASCII characters as is. */ 86 int ascii_compatible; 87 88 /** Minimum and maximum character of the charset. If 89 <ascii_compatible> is nonzero, <min_char> is actually the 90 minimum non-ASCII character of the charset. */ 91 int min_char, max_char; 92 93 /** ISO 2022 final byte of the charset. It must be in the range 94 48..127, or -1. The value -1 means that the charset is not 95 encodable by ISO 2022 based coding systems. */ 96 int final_byte; 97 98 /** ISO 2022 revision number of the charset, or -1. The value -1 99 means that the charset has no revision number. Used only when 100 <final_byte> is not -1. */ 101 int revision; 102 103 /** Specify how to encode/decode code-point of the charset. It must 104 be Moffset, Mmap, Munify, Msubset, or Msuperset. */ 105 MSymbol method; 106 107 /** Array of integers to decode a code-point of the charset. It is 108 indexed by a "char-index" of the code-point, and the 109 corresponding element is a character of the charset, or -1 if 110 the code point is not valid in the charset. Used only when 111 <method> is Mmap or Munify. */ 112 int *decoder; 113 114 /** Char-table to encode a character of the charset. It is indexed 115 by a character code, and the corresponding element is a code 116 point of the character in the charset, or 117 MCHAR_INVALID_CODE if the character is not included in the 118 charset. Used only when <method> is Mmap or Munify. */ 119 MCharTable *encoder; 120 121 int unified_max; 122 123 /** Array of pointers to parent charsets. Used only when <method> 124 is Msubset or Msuperset. Atmost 8 parents are supported. */ 125 MCharset *parents[8]; 126 127 /* Number of parent charsets. */ 128 int nparents; 129 130 unsigned subset_min_code, subset_max_code; 131 int subset_offset; 132 133 int simple; 134 135 /** If the charset is fully loaded (i.e. all the above member are 136 set to correct values), the value is 1. Otherwise, the value is 137 0. */ 138 int fully_loaded; 139 }; 140 141 extern MPlist *mcharset__cache; 142 143 /** Return a charset associated with the symbol CHARSET_SYM. */ 144 145 #define MCHARSET(charset_sym) \ 146 (((charset_sym) == MPLIST_KEY (mcharset__cache) \ 147 || (MPLIST_KEY (mcharset__cache) = (charset_sym), \ 148 MPLIST_VAL (mcharset__cache) \ 149 = (MCharset *) msymbol_get ((charset_sym), Mcharset))) \ 150 ? MPLIST_VAL (mcharset__cache) \ 151 : mcharset__find (charset_sym)) 152 153 154 /** Return index of a character whose code-point in CHARSET is CODE. 155 If CODE is not valid, return -1. */ 156 157 #define CODE_POINT_TO_INDEX(charset, code) \ 158 ((charset)->no_code_gap \ 159 ? (code) - (charset)->min_code \ 160 : (((charset)->code_range_mask[(code) >> 24] & 0x8) \ 161 && ((charset)->code_range_mask[((code) >> 16) & 0xFF] & 0x4) \ 162 && ((charset)->code_range_mask[((code) >> 8) & 0xFF] & 0x2) \ 163 && ((charset)->code_range_mask[(code) & 0xFF] & 0x1)) \ 164 ? (((((code) >> 24) - (charset)->code_range[12]) \ 165 * (charset)->code_range[11]) \ 166 + (((((code) >> 16) & 0xFF) - (charset)->code_range[8]) \ 167 * (charset)->code_range[7]) \ 168 + (((((code) >> 8) & 0xFF) - (charset)->code_range[4]) \ 169 * (charset)->code_range[3]) \ 170 + (((code) & 0xFF) - (charset)->code_range[0]) \ 171 - ((charset)->min_code - (charset)->code_range_min_code)) \ 172 : -1) 173 174 175 /* Return code-point of a character whose index is IDX. 176 The validness of IDX is not checked. IDX may be modified. */ 177 178 #define INDEX_TO_CODE_POINT(charset, idx) \ 179 ((charset)->no_code_gap \ 180 ? (idx) + (charset)->min_code \ 181 : (idx += (charset)->min_code - (charset)->code_range_min_code, \ 182 (((charset)->code_range[0] + (idx) % (charset)->code_range[2]) \ 183 | (((charset)->code_range[4] \ 184 + ((idx) / (charset)->code_range[3] % (charset)->code_range[6])) \ 185 << 8) \ 186 | (((charset)->code_range[8] \ 187 + ((idx) / (charset)->code_range[7] % (charset)->code_range[10])) \ 188 << 16) \ 189 | (((charset)->code_range[12] + ((idx) / (charset)->code_range[11])) \ 190 << 24)))) 191 192 193 /** Return a character whose code-point in CHARSET is CODE. If CODE 194 is invalid, return -1. */ 195 196 #define DECODE_CHAR(charset, code) \ 197 (((code) < 128 && (charset)->ascii_compatible) \ 198 ? (int) (code) \ 199 : ((code) < (charset)->min_code || (code) > (charset)->max_code) \ 200 ? -1 \ 201 : ! (charset)->simple \ 202 ? mcharset__decode_char ((charset), (code)) \ 203 : (charset)->method == Moffset \ 204 ? (code) - (charset)->min_code + (charset)->min_char \ 205 : (charset)->decoder[(code) - (charset)->min_code]) 206 207 208 /** Return a code-point in CHARSET for character C. If CHARSET 209 does not contain C, return MCHAR_INVALID_CODE. */ 210 211 #define ENCODE_CHAR(charset, c) \ 212 (! (charset)->simple \ 213 ? mcharset__encode_char ((charset), (c)) \ 214 : ((c) < (charset)->min_char || (c) > (charset)->max_char) \ 215 ? MCHAR_INVALID_CODE \ 216 : (charset)->method == Moffset \ 217 ? (c) - (charset)->min_char + (charset)->min_code \ 218 : (unsigned) mchartable_lookup ((charset)->encoder, (c))) 219 220 221 extern MCharset *mcharset__ascii; 222 extern MCharset *mcharset__binary; 223 extern MCharset *mcharset__m17n; 224 extern MCharset *mcharset__unicode; 225 226 #define ISO_MAX_DIMENSION 3 227 #define ISO_MAX_CHARS 2 228 #define ISO_MAX_FINAL 0x80 /* only 0x30..0xFF are used */ 229 230 typedef struct 231 { 232 /* Table of ISO-2022 charsets. */ 233 int size, inc, used; 234 MCharset **charsets; 235 236 /** A 3-dimensional table indexed by "dimension", "chars", and 237 "final byte" of an ISO-2022 charset to get the correponding 238 charset. A charset that has a revision number is not stored in 239 this table. */ 240 MCharset *classified[ISO_MAX_DIMENSION][ISO_MAX_CHARS][ISO_MAX_FINAL]; 241 } MCharsetISO2022Table; 242 243 extern MCharsetISO2022Table mcharset__iso_2022_table; 244 245 #define MCHARSET_ISO_2022(dim, chars, final) \ 246 mcharset__iso_2022_table.classified[(dim) - 1][(chars) == 96][(final)] 247 248 extern MCharset *mcharset__find (MSymbol name); 249 extern int mcharset__decode_char (MCharset *charset, unsigned code); 250 extern unsigned mcharset__encode_char (MCharset *charset, int c); 251 extern int mcharset__load_from_database (); 252 253 #endif /* _M17N_CHARSET_H_ */ 254