| 1 |
/* |
| 2 |
* The contents of this file are subject to the Interbase Public |
| 3 |
* License Version 1.0 (the "License"); you may not use this file |
| 4 |
* except in compliance with the License. You may obtain a copy |
| 5 |
* of the License at http://www.Inprise.com/IPL.html |
| 6 |
* |
| 7 |
* Software distributed under the License is distributed on an |
| 8 |
* "AS IS" basis, WITHOUT WARRANTY OF ANY KIND, either express |
| 9 |
* or implied. See the License for the specific language governing |
| 10 |
* rights and limitations under the License. |
| 11 |
* |
| 12 |
* The Original Code was created by Inprise Corporation |
| 13 |
* and its predecessors. Portions created by Inprise Corporation are |
| 14 |
* Copyright (C) Inprise Corporation. |
| 15 |
* |
| 16 |
* All Rights Reserved. |
| 17 |
* Contributor(s): ______________________________________. |
| 18 |
*/ |
| 19 |
|
| 20 |
#include "firebird.h" |
| 21 |
#include <stdio.h> |
| 22 |
#include <string.h> |
| 23 |
|
| 24 |
typedef struct { |
| 25 |
unsigned int codepoint; |
| 26 |
unsigned int equivilant; |
| 27 |
unsigned int exists; /* Count of times in input table */ |
| 28 |
unsigned int not_defined; /* Defined as NOT DEFINED */ |
| 29 |
char name[80]; /* Name of the character */ |
| 30 |
} CONVERSION; |
| 31 |
|
| 32 |
typedef struct { |
| 33 |
unsigned int high_point; |
| 34 |
unsigned int low_point; |
| 35 |
unsigned int count; |
| 36 |
CONVERSION table[0xFFFF + 1]; |
| 37 |
} TABLE; |
| 38 |
|
| 39 |
TABLE codepoint_conversion; |
| 40 |
TABLE unicode_conversion; |
| 41 |
|
| 42 |
#define UNICODE_REPLACEMENT_CHARACTER 0xFFFD |
| 43 |
#define CANT_MAP_CHARACTER 0 |
| 44 |
|
| 45 |
void print_condensed_indexed_table(char *name, TABLE * table, |
| 46 |
unsigned short replacement); |
| 47 |
void print_indexed_table(char *name, TABLE * table, |
| 48 |
unsigned short replacement); |
| 49 |
void print_direct_table(char *name, TABLE * table); |
| 50 |
void declare(unsigned short codepoint, unsigned short unicode, char *name); |
| 51 |
|
| 52 |
main(int argc, char *argv[]) |
| 53 |
{ |
| 54 |
FILE *f = NULL; |
| 55 |
char line[100]; |
| 56 |
int i; |
| 57 |
int linecount = 0; |
| 58 |
int option_euc = 0; |
| 59 |
int option_sjis = 0; |
| 60 |
int option_condensed = 0; |
| 61 |
int usage = 0; |
| 62 |
char *filename = NULL; |
| 63 |
|
| 64 |
for (i = 1; i < argc && !usage; i++) |
| 65 |
{ |
| 66 |
if (argv[i][0] == '-') |
| 67 |
{ |
| 68 |
if (strcmp(argv[i], "-euc") == 0) |
| 69 |
option_euc++; |
| 70 |
else if (strcmp(argv[i], "-sjis") == 0) |
| 71 |
option_sjis++; |
| 72 |
else if (strcmp(argv[i], "-condense") == 0) |
| 73 |
option_condensed++; |
| 74 |
else |
| 75 |
usage++; |
| 76 |
} |
| 77 |
else if (filename) |
| 78 |
usage++; |
| 79 |
else |
| 80 |
filename = argv[i]; |
| 81 |
} |
| 82 |
|
| 83 |
if (usage) |
| 84 |
{ |
| 85 |
fprintf(stderr, |
| 86 |
"usage: mapgen4 [-condense | -euc | -sjis] <name of unicode map file>\n"); |
| 87 |
exit(1); |
| 88 |
} |
| 89 |
|
| 90 |
f = fopen(filename, "r"); |
| 91 |
if (!f) |
| 92 |
{ |
| 93 |
fprintf(stderr, "Unable to open file '%s'\n", filename); |
| 94 |
exit(1); |
| 95 |
} |
| 96 |
|
| 97 |
memset((void *) &codepoint_conversion, 0, sizeof(codepoint_conversion)); |
| 98 |
memset((void *) &unicode_conversion, 0, sizeof(unicode_conversion)); |
| 99 |
codepoint_conversion.low_point = 0xFFFF; |
| 100 |
unicode_conversion.low_point = 0xFFFF; |
| 101 |
|
| 102 |
printf("/* -------------------------------------------- \n"); |
| 103 |
printf(" THIS FILE WAS GENERATED BY intl/mapgen4.c\n"); |
| 104 |
printf(" "); |
| 105 |
for (i = 0; i < argc; i++) |
| 106 |
printf("%s ", argv[i]); |
| 107 |
printf("\n\n"); |
| 108 |
printf(" Unicode mapping table generated from file %s \n", |
| 109 |
filename); |
| 110 |
printf("\n"); |
| 111 |
|
| 112 |
|
| 113 |
while (fgets(line, sizeof(line), f)) |
| 114 |
{ |
| 115 |
unsigned int unicode; |
| 116 |
unsigned int codepoint; |
| 117 |
char *p; |
| 118 |
|
| 119 |
/* Strip trailing control codes from the line */ |
| 120 |
p = &line[strlen(line) - 1]; |
| 121 |
while (p >= line && (*p == '\r' || *p == '\n' || *p == 26)) |
| 122 |
*p-- = 0; |
| 123 |
|
| 124 |
printf("%s\n", line); |
| 125 |
linecount++; |
| 126 |
/* Skip blank lines */ |
| 127 |
if (line[0] == 0) |
| 128 |
continue; |
| 129 |
/* Skip comment lines */ |
| 130 |
if (line[0] == '#') |
| 131 |
continue; |
| 132 |
p = line; |
| 133 |
|
| 134 |
if (option_sjis) |
| 135 |
{ |
| 136 |
unsigned int sjis_point; |
| 137 |
if (1 != sscanf(p, "0x%04x\t", &sjis_point)) |
| 138 |
{ |
| 139 |
fprintf(stderr, "Problem in input file - line %d\n", |
| 140 |
linecount); |
| 141 |
exit(1); |
| 142 |
} |
| 143 |
p = strchr(p, '\t') + 1; |
| 144 |
} |
| 145 |
|
| 146 |
codepoint = 0; |
| 147 |
if (1 != sscanf(p, "0x%04x\t", &codepoint)) |
| 148 |
{ |
| 149 |
fprintf(stderr, "Problem in input file - line %d\n", |
| 150 |
linecount); |
| 151 |
exit(1); |
| 152 |
} |
| 153 |
p = strchr(p, '\t') + 1; |
| 154 |
|
| 155 |
if (option_euc) |
| 156 |
codepoint += 0x8080; |
| 157 |
|
| 158 |
unicode = 0; |
| 159 |
if (0 == sscanf(p, "0x%04x\t", &unicode)) |
| 160 |
{ |
| 161 |
/* Conversion is not defined */ |
| 162 |
codepoint_conversion.table[codepoint].not_defined++; |
| 163 |
declare(codepoint, unicode, "NOT DEFINED"); |
| 164 |
} |
| 165 |
else |
| 166 |
{ |
| 167 |
/* Find start of comment defining Unicode name */ |
| 168 |
p = strchr(p, '#'); |
| 169 |
if (p) |
| 170 |
p++; /* Skip comment character */ |
| 171 |
else |
| 172 |
p = ""; |
| 173 |
declare(codepoint, unicode, p); |
| 174 |
} |
| 175 |
|
| 176 |
} |
| 177 |
fclose(f); |
| 178 |
printf(" -------------------------------------------- */\n"); |
| 179 |
|
| 180 |
/* Declare any standard ASCII characters that didn't exist in the table */ |
| 181 |
for (i = 0; i <= 0x7f; i++) |
| 182 |
{ |
| 183 |
if (!codepoint_conversion.table[i].exists) |
| 184 |
declare(i, i, "ASCII"); |
| 185 |
} |
| 186 |
|
| 187 |
printf("\n"); |
| 188 |
printf("\n"); |
| 189 |
printf("#ifndef UNICODE_REPLACEMENT_CHARACTER\n"); |
| 190 |
printf("#define UNICODE_REPLACEMENT_CHARACTER 0xFFFD\n"); |
| 191 |
printf("#endif\n"); |
| 192 |
printf("\n"); |
| 193 |
|
| 194 |
/* Small tables print with all comments */ |
| 195 |
if ((codepoint_conversion.high_point < 256) || (!option_condensed)) |
| 196 |
print_indexed_table("to_unicode", &codepoint_conversion, |
| 197 |
UNICODE_REPLACEMENT_CHARACTER); |
| 198 |
else |
| 199 |
print_condensed_indexed_table("to_unicode", &codepoint_conversion, |
| 200 |
UNICODE_REPLACEMENT_CHARACTER); |
| 201 |
|
| 202 |
|
| 203 |
printf("\n"); |
| 204 |
printf("\n"); |
| 205 |
printf("#ifndef CANT_MAP_CHARACTER\n"); |
| 206 |
printf("#define CANT_MAP_CHARACTER 0\n"); |
| 207 |
printf("#endif\n"); |
| 208 |
printf("\n"); |
| 209 |
|
| 210 |
/* Small tables print with all comments */ |
| 211 |
if ((codepoint_conversion.count < 256) || (!option_condensed)) |
| 212 |
print_indexed_table("from_unicode", &unicode_conversion, |
| 213 |
CANT_MAP_CHARACTER); |
| 214 |
else |
| 215 |
print_condensed_indexed_table("from_unicode", &unicode_conversion, |
| 216 |
CANT_MAP_CHARACTER); |
| 217 |
|
| 218 |
return 0; |
| 219 |
} |
| 220 |
|
| 221 |
void declare(unsigned short codepoint, unsigned short unicode, char *name) |
| 222 |
{ |
| 223 |
if (codepoint < codepoint_conversion.low_point) |
| 224 |
codepoint_conversion.low_point = codepoint; |
| 225 |
if (codepoint > codepoint_conversion.high_point) |
| 226 |
codepoint_conversion.high_point = codepoint; |
| 227 |
|
| 228 |
if (codepoint_conversion.table[codepoint].exists++) |
| 229 |
{ |
| 230 |
if (unicode != codepoint_conversion.table[codepoint].equivilant) |
| 231 |
fprintf(stderr, |
| 232 |
"Error: duplicate unequal mappings for 0x%04x : 0x%04x and 0x%04x\n", |
| 233 |
codepoint, unicode, |
| 234 |
codepoint_conversion.table[codepoint].equivilant); |
| 235 |
else |
| 236 |
fprintf(stderr, |
| 237 |
"Warning: duplicate entries for codepoint 0x%04x to 0x%04x\n", |
| 238 |
codepoint, unicode); |
| 239 |
} |
| 240 |
else |
| 241 |
codepoint_conversion.count++; |
| 242 |
|
| 243 |
codepoint_conversion.table[codepoint].codepoint = codepoint; |
| 244 |
strcpy(codepoint_conversion.table[codepoint].name, name); |
| 245 |
|
| 246 |
if (!codepoint_conversion.table[codepoint].not_defined) |
| 247 |
{ |
| 248 |
codepoint_conversion.table[codepoint].equivilant = unicode; |
| 249 |
|
| 250 |
if (unicode < unicode_conversion.low_point) |
| 251 |
unicode_conversion.low_point = unicode; |
| 252 |
if (unicode > unicode_conversion.high_point) |
| 253 |
unicode_conversion.high_point = unicode; |
| 254 |
if (!unicode_conversion.table[unicode].exists++) |
| 255 |
{ |
| 256 |
unicode_conversion.table[unicode].codepoint = unicode; |
| 257 |
if (unicode != UNICODE_REPLACEMENT_CHARACTER) |
| 258 |
{ |
| 259 |
unicode_conversion.table[unicode].equivilant = codepoint; |
| 260 |
strcpy(unicode_conversion.table[unicode].name, name); |
| 261 |
} |
| 262 |
} |
| 263 |
else |
| 264 |
{ |
| 265 |
if (unicode != UNICODE_REPLACEMENT_CHARACTER) |
| 266 |
fprintf(stderr, |
| 267 |
"Warning: Multiple mappings to Unicode 0x%04x : 0x%04x and 0x%04x\n", |
| 268 |
unicode, codepoint, |
| 269 |
unicode_conversion.table[unicode].equivilant); |
| 270 |
} |
| 271 |
} |
| 272 |
} |
| 273 |
|
| 274 |
|
| 275 |
void print_direct_table(char *name, TABLE* table) |
| 276 |
{ |
| 277 |
int i; |
| 278 |
|
| 279 |
printf("static const USHORT %s_map[256] = {\n", name); |
| 280 |
for (i = 0; i < 256; i++) |
| 281 |
{ |
| 282 |
if (table->table[i].exists != 1 || table->table[i].not_defined) |
| 283 |
printf("/* %02X */ UNICODE_REPLACEMENT_CHARACTER,\n", i); |
| 284 |
else |
| 285 |
printf("/* %02X */ 0x%04X /* %45s */,\n", |
| 286 |
i, table->table[i].equivilant, table->table[i].name); |
| 287 |
} |
| 288 |
printf("};\n"); |
| 289 |
} |
| 290 |
|
| 291 |
|
| 292 |
void print_indexed_table(char *name, TABLE* table, |
| 293 |
unsigned short replacement) |
| 294 |
{ |
| 295 |
unsigned int index; |
| 296 |
unsigned int codepoint; |
| 297 |
unsigned short upper_byte[256]; |
| 298 |
int i; |
| 299 |
|
| 300 |
printf("static const USHORT %s_mapping_array[] = {\n", |
| 301 |
name); |
| 302 |
for (index = 0; index < 256; index++) |
| 303 |
{ |
| 304 |
if (replacement == UNICODE_REPLACEMENT_CHARACTER) |
| 305 |
printf("/* U+XX%02X */\tUNICODE_REPLACEMENT_CHARACTER,\t/* %d */\n", |
| 306 |
index % 256, index); |
| 307 |
else |
| 308 |
printf("/* U+XX%02X */\tCANT_MAP_CHARACTER,\t/* %d */\n", |
| 309 |
index % 256, index); |
| 310 |
} |
| 311 |
|
| 312 |
memset(upper_byte, 0, sizeof(upper_byte)); |
| 313 |
codepoint = 0; |
| 314 |
while (codepoint < 0xFFFF + 1) |
| 315 |
{ |
| 316 |
if (!table->table[codepoint].exists) |
| 317 |
{ |
| 318 |
codepoint++; |
| 319 |
continue; |
| 320 |
} |
| 321 |
upper_byte[codepoint / 256] = index; |
| 322 |
while ((index % 256) < (codepoint % 256)) |
| 323 |
{ |
| 324 |
if (replacement == UNICODE_REPLACEMENT_CHARACTER) |
| 325 |
printf("/* U+%04X */\tUNICODE_REPLACEMENT_CHARACTER,\t/* %d */\n", |
| 326 |
(codepoint & 0xFF00) + (index % 256), index); |
| 327 |
else |
| 328 |
printf("/* U+%04X */\tCANT_MAP_CHARACTER,\t/* %d */\n", |
| 329 |
(codepoint & 0xFF00) + (index % 256), index); |
| 330 |
index++; |
| 331 |
} |
| 332 |
do { |
| 333 |
if (!table->table[codepoint].exists) |
| 334 |
{ |
| 335 |
if (replacement == UNICODE_REPLACEMENT_CHARACTER) |
| 336 |
printf("/* U+%04X */\tUNICODE_REPLACEMENT_CHARACTER,\t/* %d */\n", |
| 337 |
codepoint, index); |
| 338 |
else |
| 339 |
printf("/* U+%04X */\tCANT_MAP_CHARACTER,\t/* %d */\n", |
| 340 |
codepoint, index); |
| 341 |
} |
| 342 |
else |
| 343 |
printf("/* U+%04X */\t0x%04X\t/* %45s */,/* %d */\n", |
| 344 |
codepoint, table->table[codepoint].equivilant, |
| 345 |
table->table[codepoint].name, index); |
| 346 |
index++; |
| 347 |
codepoint++; |
| 348 |
} while (index % 256); |
| 349 |
} |
| 350 |
printf("\t0 /* END OF MAP TABLE */\n"); |
| 351 |
printf("};\n"); |
| 352 |
|
| 353 |
printf("static const USHORT %s_map[256] = {\n", name); |
| 354 |
for (i = 0; i < 256; i++) |
| 355 |
printf("/* U+%02X-- */\t%d,\n", i, upper_byte[i]); |
| 356 |
printf("};\n"); |
| 357 |
|
| 358 |
} |
| 359 |
|
| 360 |
|
| 361 |
void print_condensed_indexed_table(char *name, |
| 362 |
TABLE* table, unsigned short replacement) |
| 363 |
{ |
| 364 |
unsigned int index; |
| 365 |
unsigned int codepoint; |
| 366 |
unsigned short upper_byte[256]; |
| 367 |
int i; |
| 368 |
|
| 369 |
printf("static const USHORT %s_mapping_array[] = {\n", |
| 370 |
name); |
| 371 |
|
| 372 |
printf("\n"); |
| 373 |
printf("/* %5d to %5d */\n", 0, 255); |
| 374 |
for (index = 0; index < 256; index++) |
| 375 |
{ |
| 376 |
if (replacement) |
| 377 |
printf("0x%04X,", replacement); |
| 378 |
else |
| 379 |
printf("0, "); |
| 380 |
if (index % 8 == 7) |
| 381 |
printf("\n"); |
| 382 |
else |
| 383 |
printf(" "); |
| 384 |
} |
| 385 |
|
| 386 |
memset(upper_byte, 0, sizeof(upper_byte)); |
| 387 |
codepoint = 0; |
| 388 |
while (codepoint < 0xFFFF + 1) |
| 389 |
{ |
| 390 |
if (!table->table[codepoint].exists) |
| 391 |
{ |
| 392 |
codepoint++; |
| 393 |
continue; |
| 394 |
} |
| 395 |
upper_byte[codepoint / 256] = index; |
| 396 |
|
| 397 |
printf("\n"); |
| 398 |
printf("/* %5d to %5d */\n", index, index + 255); |
| 399 |
|
| 400 |
while ((index % 256) < (codepoint % 256)) |
| 401 |
{ |
| 402 |
if (replacement) |
| 403 |
printf("0x%04X,", replacement); |
| 404 |
else |
| 405 |
printf("0, "); |
| 406 |
if (index % 8 == 7) |
| 407 |
printf("\n"); |
| 408 |
else |
| 409 |
printf(" "); |
| 410 |
index++; |
| 411 |
} |
| 412 |
do { |
| 413 |
if (!table->table[codepoint].exists) |
| 414 |
{ |
| 415 |
if (replacement) |
| 416 |
printf("0x%04X,", replacement); |
| 417 |
else |
| 418 |
printf("0, "); |
| 419 |
} |
| 420 |
else |
| 421 |
printf("0x%04X,", table->table[codepoint].equivilant); |
| 422 |
if (index % 8 == 7) |
| 423 |
printf("\n"); |
| 424 |
else |
| 425 |
printf(" "); |
| 426 |
index++; |
| 427 |
codepoint++; |
| 428 |
} while (index % 256); |
| 429 |
} |
| 430 |
printf("\t0 /* END OF MAP TABLE */\n"); |
| 431 |
printf("};\n"); |
| 432 |
|
| 433 |
printf("static const USHORT %s_map[256] = {\n", name); |
| 434 |
for (i = 0; i < 256; i++) |
| 435 |
printf("%d,\n", upper_byte[i]); |
| 436 |
printf("};\n"); |
| 437 |
|
| 438 |
} |