|
fltk 1.3.0rc3
About: FLTK (Fast Light Tool Kit) is a cross-platform C++ GUI toolkit for UNIX/Linux (X11), Microsoft Windows, and MacOS X. Release candidate.
 SfR Fresh Dox: fltk-1.3.0rc3-source.tar.gz ("inofficial" and yet experimental doxygen-generated source code documentation)  |
00001 /* $XFree86: xc/lib/X11/lcUniConv/8bit_tab_to_h.c,v 1.3 2001/02/09 00:02:54 dawes Exp $ */ 00002 00003 /* 00004 * Generates an 8-bit character set table from a .TXT table as found on 00005 * ftp.unicode.org or from a table containing the 256 Unicode values as 00006 * hexadecimal integers. 00007 * Examples: 00008 * 00009 * ./8bit_tab_to_h ISO-8859-1 iso8859_1 < tab8859_1 00010 * ./8bit_tab_to_h ISO-8859-2 iso8859_2 < tab8859_2 00011 * ./8bit_tab_to_h ISO-8859-3 iso8859_3 < tab8859_3 00012 * ./8bit_tab_to_h ISO-8859-4 iso8859_4 < tab8859_4 00013 * ./8bit_tab_to_h ISO-8859-5 iso8859_5 < tab8859_5 00014 * ./8bit_tab_to_h ISO-8859-6 iso8859_6 < tab8859_6 00015 * ./8bit_tab_to_h ISO-8859-7 iso8859_7 < tab8859_7 00016 * ./8bit_tab_to_h ISO-8859-8 iso8859_8 < tab8859_8 00017 * ./8bit_tab_to_h ISO-8859-9 iso8859_9 < tab8859_9 00018 * ./8bit_tab_to_h ISO-8859-10 iso8859_10 < tab8859_10 00019 * ./8bit_tab_to_h ISO-8859-14 iso8859_14 < tab8859_14 00020 * ./8bit_tab_to_h ISO-8859-15 iso8859_15 < tab8859_15 00021 * ./8bit_tab_to_h JISX0201.1976-0 jisx0201 < jis0201 00022 * ./8bit_tab_to_h TIS620-0 tis620 < tabtis620 00023 * ./8bit_tab_to_h KOI8-R koi8_r < tabkoi8_r 00024 * ./8bit_tab_to_h KOI8-U koi8_u < tabkoi8_u 00025 * ./8bit_tab_to_h ARMSCII-8 armscii_8 < tabarmscii_8 00026 * ./8bit_tab_to_h CP1133 cp1133 < tabibm_cp1133 00027 * ./8bit_tab_to_h MULELAO-1 mulelao < tabmulelao_1 00028 * ./8bit_tab_to_h VISCII1.1-1 viscii1 < tabviscii 00029 * ./8bit_tab_to_h TCVN-5712 tcvn < tabtcvn 00030 * ./8bit_tab_to_h GEORGIAN-ACADEMY georgian_ac < tabgeorgian_academy 00031 * ./8bit_tab_to_h GEORGIAN-PS georgian_ps < tabgeorgian_ps 00032 * 00033 * ./8bit_tab_to_h ISO-8859-1 iso8859_1 < 8859-1.TXT 00034 * ./8bit_tab_to_h ISO-8859-2 iso8859_2 < 8859-2.TXT 00035 * ./8bit_tab_to_h ISO-8859-3 iso8859_3 < 8859-3.TXT 00036 * ./8bit_tab_to_h ISO-8859-4 iso8859_4 < 8859-4.TXT 00037 * ./8bit_tab_to_h ISO-8859-5 iso8859_5 < 8859-5.TXT 00038 * ./8bit_tab_to_h ISO-8859-6 iso8859_6 < 8859-6.TXT 00039 * ./8bit_tab_to_h ISO-8859-7 iso8859_7 < 8859-7.TXT 00040 * ./8bit_tab_to_h ISO-8859-8 iso8859_8 < 8859-8.TXT 00041 * ./8bit_tab_to_h ISO-8859-9 iso8859_9 < 8859-9.TXT 00042 * ./8bit_tab_to_h ISO-8859-10 iso8859_10 < 8859-10.TXT 00043 * ./8bit_tab_to_h ISO-8859-14 iso8859_14 < 8859-14.TXT 00044 * ./8bit_tab_to_h ISO-8859-15 iso8859_15 < 8859-15.TXT 00045 * ./8bit_tab_to_h JISX0201.1976-0 jisx0201 < JIS0201.TXT 00046 * ./8bit_tab_to_h KOI8-R koi8_r < KOI8-R.TXT 00047 */ 00048 00049 #include <stdio.h> 00050 #include <stdlib.h> 00051 #include <stdbool.h> 00052 #include <string.h> 00053 00054 int main (int argc, char *argv[]) 00055 { 00056 const char* charsetname; 00057 const char* c_charsetname; 00058 const char* filename; 00059 const char* directory; 00060 int charset2uni[0x100]; 00061 00062 if (argc != 3 && argc != 4 && argc != 5) 00063 exit(1); 00064 charsetname = argv[1]; 00065 c_charsetname = argv[2]; 00066 if (argc > 3) { 00067 filename = argv[3]; 00068 } else { 00069 char* s = (char*) malloc(strlen(c_charsetname)+strlen(".h")+1); 00070 strcpy(s,c_charsetname); strcat(s,".h"); 00071 filename = s; 00072 } 00073 directory = (argc > 4 ? argv[4] : ""); 00074 00075 fprintf(stderr, "Creating %s%s\n", directory, filename); 00076 00077 { 00078 int i, c; 00079 c = getc(stdin); 00080 ungetc(c,stdin); 00081 if (c == '#') { 00082 /* Read a unicode.org style .TXT file. */ 00083 for (i = 0; i < 0x100; i++) 00084 charset2uni[i] = 0xfffd; 00085 for (;;) { 00086 c = getc(stdin); 00087 if (c == EOF) 00088 break; 00089 if (c == '\n' || c == ' ' || c == '\t') 00090 continue; 00091 if (c == '#') { 00092 do { c = getc(stdin); } while (!(c == EOF || c == '\n')); 00093 continue; 00094 } 00095 ungetc(c,stdin); 00096 if (scanf("0x%x", &i) != 1 || !(i >= 0 && i < 0x100)) 00097 exit(1); 00098 do { c = getc(stdin); } while (c == ' ' || c == '\t'); 00099 if (c != EOF) 00100 ungetc(c,stdin); 00101 if (c == '\n' || c == '#') 00102 continue; 00103 if (scanf("0x%x", &charset2uni[i]) != 1) 00104 exit(1); 00105 } 00106 } else { 00107 /* Read a table of hexadecimal Unicode values. */ 00108 for (i = 0; i < 0x100; i++) { 00109 if (scanf("%x", &charset2uni[i]) != 1) 00110 exit(1); 00111 if (charset2uni[i] < 0 || charset2uni[i] == 0xffff) 00112 charset2uni[i] = 0xfffd; 00113 } 00114 if (scanf("%x", &i) != EOF) 00115 exit(1); 00116 } 00117 } 00118 00119 /* Write the output file. */ 00120 { 00121 FILE* f; 00122 00123 { 00124 char* fname = malloc(strlen(directory)+strlen(filename)+1); 00125 strcpy(fname,directory); strcat(fname,filename); 00126 f = fopen(fname,"w"); 00127 if (f == NULL) 00128 exit(1); 00129 } 00130 00131 fprintf(f, "\n"); 00132 fprintf(f, "/*\n"); 00133 fprintf(f, " * %s\n", charsetname); 00134 fprintf(f, " */\n"); 00135 fprintf(f, "\n"); 00136 00137 { 00138 int i, i1, i2, i3; 00139 int line[16]; 00140 int tableno; 00141 struct { int minline; int maxline; } tables[16]; 00142 bool some_invalid; 00143 bool final_ret_reached; 00144 00145 for (i1 = 0; i1 < 16; i1++) { 00146 bool all_invalid = true; 00147 bool all_identity = true; 00148 for (i2 = 0; i2 < 16; i2++) { 00149 i = 16*i1+i2; 00150 if (charset2uni[i] != 0xfffd) 00151 all_invalid = false; 00152 if (charset2uni[i] != i) 00153 all_identity = false; 00154 } 00155 if (all_invalid) 00156 line[i1] = -2; 00157 else if (all_identity) 00158 line[i1] = -1; 00159 else 00160 line[i1] = 0; 00161 } 00162 tableno = 0; 00163 for (i1 = 0; i1 < 16; i1++) { 00164 if (line[i1] >= 0) { 00165 if (i1 > 0 && tableno > 0 && line[i1-1] == tableno-1) { 00166 line[i1] = tableno-1; 00167 tables[tableno-1].maxline = i1; 00168 } else { 00169 tableno++; 00170 line[i1] = tableno-1; 00171 tables[tableno-1].minline = tables[tableno-1].maxline = i1; 00172 } 00173 } 00174 } 00175 some_invalid = false; 00176 for (i = 0; i < 0x100; i++) 00177 if (charset2uni[i] == 0xfffd) 00178 some_invalid = true; 00179 if (tableno > 0) { 00180 int t; 00181 for (t = 0; t < tableno; t++) { 00182 fprintf(f, "static const unsigned short %s_2uni", c_charsetname); 00183 if (tableno > 1) 00184 fprintf(f, "_%d", t+1); 00185 fprintf(f, "[%d] = {\n", 16*(tables[t].maxline-tables[t].minline+1)); 00186 for (i1 = tables[t].minline; i1 <= tables[t].maxline; i1++) { 00187 fprintf(f, " /* 0x%02x */\n", 16*i1); 00188 for (i2 = 0; i2 < 2; i2++) { 00189 fprintf(f, " "); 00190 for (i3 = 0; i3 < 8; i3++) { 00191 i = 16*i1+8*i2+i3; 00192 fprintf(f, " 0x%04x,", charset2uni[i]); 00193 } 00194 fprintf(f, "\n"); 00195 } 00196 } 00197 fprintf(f, "};\n"); 00198 } 00199 fprintf(f, "\n"); 00200 } 00201 final_ret_reached = false; 00202 fprintf(f, "static int\n%s_mbtowc (conv_t conv, ucs4_t *pwc, const unsigned char *s, int n)\n", c_charsetname); 00203 fprintf(f, "{\n"); 00204 fprintf(f, " unsigned char c = *s;\n"); 00205 if (some_invalid) { 00206 for (i1 = 0; i1 < 16;) { 00207 int t = line[i1]; 00208 const char* indent; 00209 for (i2 = i1; i2 < 16 && line[i2] == t; i2++); 00210 indent = (i1 == 0 && i2 == 16 ? " " : " "); 00211 if (i1 == 0) { 00212 if (i2 == 16) { 00213 } else { 00214 fprintf(f, " if (c < 0x%02x) {\n", 16*i2); 00215 } 00216 } else { 00217 if (i2 == 16) { 00218 fprintf(f, " else {\n"); 00219 } else { 00220 fprintf(f, " else if (c < 0x%02x) {\n", 16*i2); 00221 } 00222 } 00223 if (t == -2) { 00224 final_ret_reached = true; 00225 } else if (t == -1) { 00226 fprintf(f, "%s*pwc = (ucs4_t) c;\n", indent); 00227 fprintf(f, "%sreturn 1;\n", indent); 00228 } else { 00229 fprintf(f, "%s", indent); 00230 some_invalid = false; 00231 for (i = 16*i1; i < 16*i2; i++) 00232 if (charset2uni[i] == 0xfffd) 00233 some_invalid = true; 00234 if (some_invalid) 00235 fprintf(f, "unsigned short wc = "); 00236 else 00237 fprintf(f, "*pwc = (ucs4_t) "); 00238 fprintf(f, "%s_2uni", c_charsetname); 00239 if (tableno > 1) 00240 fprintf(f, "_%d", t+1); 00241 fprintf(f, "[c"); 00242 if (tables[t].minline > 0) 00243 fprintf(f, "-0x%02x", 16*tables[t].minline); 00244 fprintf(f, "];\n"); 00245 if (some_invalid) { 00246 fprintf(f, "%sif (wc != 0xfffd) {\n", indent); 00247 fprintf(f, "%s *pwc = (ucs4_t) wc;\n", indent); 00248 fprintf(f, "%s return 1;\n", indent); 00249 fprintf(f, "%s}\n", indent); 00250 final_ret_reached = true; 00251 } else { 00252 fprintf(f, "%sreturn 1;\n", indent); 00253 } 00254 } 00255 if (!(i1 == 0 && i2 == 16)) 00256 fprintf(f, " }\n"); 00257 i1 = i2; 00258 } 00259 if (final_ret_reached) 00260 fprintf(f, " return RET_ILSEQ;\n"); 00261 } else { 00262 for (i1 = 0; i1 < 16;) { 00263 int t = line[i1]; 00264 for (i2 = i1; i2 < 16 && line[i2] == t; i2++); 00265 if (i1 == 0) { 00266 if (i2 == 16) { 00267 fprintf(f, " "); 00268 } else { 00269 fprintf(f, " if (c < 0x%02x)\n ", 16*i2); 00270 } 00271 } else { 00272 if (i2 == 16) { 00273 fprintf(f, " else\n "); 00274 } else { 00275 fprintf(f, " else if (c < 0x%02x)\n ", 16*i2); 00276 } 00277 } 00278 if (t == -1) 00279 fprintf(f, "*pwc = (ucs4_t) c;\n"); 00280 else { 00281 fprintf(f, "*pwc = (ucs4_t) %s_2uni", c_charsetname); 00282 if (tableno > 1) 00283 fprintf(f, "_%d", t+1); 00284 fprintf(f, "[c"); 00285 if (tables[t].minline > 0) 00286 fprintf(f, "-0x%02x", 16*tables[t].minline); 00287 fprintf(f, "];\n"); 00288 } 00289 i1 = i2; 00290 } 00291 fprintf(f, " return 1;\n"); 00292 } 00293 fprintf(f, "}\n"); 00294 00295 } 00296 00297 fprintf(f, "\n"); 00298 00299 { 00300 int uni2charset[0x10000]; 00301 bool pages[0x100]; 00302 int line[0x2000]; 00303 int tableno; 00304 struct { int minline; int maxline; int usecount; const char* suffix; } tables[0x2000]; 00305 bool need_c; 00306 bool fix_0000; 00307 int i, j, p, j1, j2, t; 00308 00309 for (j = 0; j < 0x10000; j++) 00310 uni2charset[j] = 0; 00311 for (p = 0; p < 0x100; p++) 00312 pages[p] = false; 00313 for (i = 0; i < 0x100; i++) { 00314 j = charset2uni[i]; 00315 if (j != 0xfffd) { 00316 uni2charset[j] = i; 00317 pages[j>>8] = true; 00318 } 00319 } 00320 for (j1 = 0; j1 < 0x2000; j1++) { 00321 bool all_invalid = true; 00322 bool all_identity = true; 00323 for (j2 = 0; j2 < 8; j2++) { 00324 j = 8*j1+j2; 00325 if (uni2charset[j] != 0) 00326 all_invalid = false; 00327 if (uni2charset[j] != j) 00328 all_identity = false; 00329 } 00330 if (all_invalid) 00331 line[j1] = -2; 00332 else if (all_identity) 00333 line[j1] = -1; 00334 else 00335 line[j1] = 0; 00336 } 00337 tableno = 0; 00338 for (j1 = 0; j1 < 0x2000; j1++) { 00339 if (line[j1] >= 0) { 00340 if (tableno > 0 00341 && ((j1 > 0 && line[j1-1] == tableno-1) 00342 || ((tables[tableno-1].maxline >> 5) == (j1 >> 5) 00343 && j1 - tables[tableno-1].maxline <= 8))) { 00344 line[j1] = tableno-1; 00345 tables[tableno-1].maxline = j1; 00346 } else { 00347 tableno++; 00348 line[j1] = tableno-1; 00349 tables[tableno-1].minline = tables[tableno-1].maxline = j1; 00350 } 00351 } 00352 } 00353 for (t = 0; t < tableno; t++) { 00354 tables[t].usecount = 0; 00355 j1 = 8*tables[t].minline; 00356 j2 = 8*(tables[t].maxline+1); 00357 for (j = j1; j < j2; j++) 00358 if (uni2charset[j] != 0) 00359 tables[t].usecount++; 00360 } 00361 for (t = 0, p = -1, i = 0; t < tableno; t++) { 00362 if (tables[t].usecount > 1) { 00363 char* s; 00364 if (p == tables[t].minline >> 5) { 00365 s = (char*) malloc(5+1); 00366 sprintf(s, "%02x_%d", p, ++i); 00367 } else { 00368 p = tables[t].minline >> 5; 00369 s = (char*) malloc(2+1); 00370 sprintf(s, "%02x", p); 00371 } 00372 tables[t].suffix = s; 00373 } else 00374 tables[t].suffix = NULL; 00375 } 00376 { 00377 p = -1; 00378 for (t = 0; t < tableno; t++) 00379 if (tables[t].usecount > 1) { 00380 p = 0; 00381 fprintf(f, "static const unsigned char %s_page%s[%d] = {\n", c_charsetname, tables[t].suffix, 8*(tables[t].maxline-tables[t].minline+1)); 00382 for (j1 = tables[t].minline; j1 <= tables[t].maxline; j1++) { 00383 if ((j1 % 0x20) == 0 && j1 > tables[t].minline) 00384 fprintf(f, " /* 0x%04x */\n", 8*j1); 00385 fprintf(f, " "); 00386 for (j2 = 0; j2 < 8; j2++) { 00387 j = 8*j1+j2; 00388 fprintf(f, " 0x%02x,", uni2charset[j]); 00389 } 00390 fprintf(f, " /* 0x%02x-0x%02x */\n", 8*(j1 % 0x20), 8*(j1 % 0x20)+7); 00391 } 00392 fprintf(f, "};\n"); 00393 } 00394 if (p >= 0) 00395 fprintf(f, "\n"); 00396 } 00397 need_c = false; 00398 for (j1 = 0; j1 < 0x2000;) { 00399 t = line[j1]; 00400 for (j2 = j1; j2 < 0x2000 && line[j2] == t; j2++); 00401 if (t >= 0) 00402 j2 = tables[t].maxline+1; 00403 if (!(t == -2 || (t == -1 && j1 == 0))) 00404 need_c = true; 00405 j1 = j2; 00406 } 00407 fix_0000 = false; 00408 fprintf(f, "static int\n%s_wctomb (conv_t conv, unsigned char *r, ucs4_t wc, int n)\n", c_charsetname); 00409 fprintf(f, "{\n"); 00410 if (need_c) 00411 fprintf(f, " unsigned char c = 0;\n"); 00412 for (j1 = 0; j1 < 0x2000;) { 00413 t = line[j1]; 00414 for (j2 = j1; j2 < 0x2000 && line[j2] == t; j2++); 00415 if (t >= 0) { 00416 if (j1 != tables[t].minline) abort(); 00417 if (j2 > tables[t].maxline+1) abort(); 00418 j2 = tables[t].maxline+1; 00419 } 00420 if (t == -2) { 00421 } else { 00422 if (j1 == 0) 00423 fprintf(f, " "); 00424 else 00425 fprintf(f, " else "); 00426 if (t >= 0 && tables[t].usecount == 0) abort(); 00427 if (t >= 0 && tables[t].usecount == 1) { 00428 if (j2 != j1+1) abort(); 00429 for (j = 8*j1; j < 8*j2; j++) 00430 if (uni2charset[j] != 0) { 00431 fprintf(f, "if (wc == 0x%04x)\n c = 0x%02x;\n", j, uni2charset[j]); 00432 break; 00433 } 00434 } else { 00435 if (j1 == 0) { 00436 fprintf(f, "if (wc < 0x%04x)", 8*j2); 00437 } else { 00438 fprintf(f, "if (wc >= 0x%04x && wc < 0x%04x)", 8*j1, 8*j2); 00439 } 00440 if (t == -1) { 00441 if (j1 == 0) 00442 /* If wc == 0, the function must return 1, not -1. */ 00443 fprintf(f, " {\n *r = wc;\n return 1;\n }\n"); 00444 else 00445 fprintf(f, "\n c = wc;\n"); 00446 } else { 00447 fprintf(f, "\n c = %s_page%s[wc", c_charsetname, tables[t].suffix); 00448 if (tables[t].minline > 0) 00449 fprintf(f, "-0x%04x", 8*j1); 00450 fprintf(f, "];\n"); 00451 if (j1 == 0 && uni2charset[0] == 0) 00452 /* If wc == 0, the function must return 1, not -1. */ 00453 fix_0000 = true; 00454 } 00455 } 00456 } 00457 j1 = j2; 00458 } 00459 if (need_c) { 00460 if (fix_0000) 00461 fprintf(f, " if (c != 0 || wc == 0) {\n"); 00462 else 00463 fprintf(f, " if (c != 0) {\n"); 00464 fprintf(f, " *r = c;\n"); 00465 fprintf(f, " return 1;\n"); 00466 fprintf(f, " }\n"); 00467 } 00468 fprintf(f, " return RET_ILSEQ;\n"); 00469 fprintf(f, "}\n"); 00470 00471 } 00472 00473 if (ferror(f) || fclose(f)) 00474 exit(1); 00475 } 00476 00477 #if 0 00478 00479 int i1, i2, i3, i1_min, i1_max, j1, j2; 00480 00481 i1_min = 16; 00482 i1_max = -1; 00483 for (i1 = 0; i1 < 16; i1++) 00484 for (i2 = 0; i2 < 16; i2++) 00485 if (charset2uni[16*i1+i2] != 0xfffd) { 00486 if (i1_min > i1) i1_min = i1; 00487 if (i1_max < i1) i1_max = i1; 00488 } 00489 printf("static const unsigned short %s_2uni[%d] = {\n", 00490 name, 16*(i1_max-i1_min+1)); 00491 for (i1 = i1_min; i1 <= i1_max; i1++) { 00492 printf(" /""* 0x%02x *""/\n", 16*i1); 00493 for (i2 = 0; i2 < 2; i2++) { 00494 printf(" "); 00495 for (i3 = 0; i3 < 8; i3++) { 00496 if (i3 > 0) printf(" "); 00497 printf("0x%04x,", charset2uni[16*i1+8*i2+i3]); 00498 } 00499 printf("\n"); 00500 } 00501 } 00502 printf("};\n"); 00503 printf("\n"); 00504 00505 for (p = 0; p < 0x100; p++) 00506 pages[p] = 0; 00507 for (i = 0; i < 0x100; i++) 00508 if (charset2uni[i] != 0xfffd) 00509 pages[charset2uni[i]>>8] = 1; 00510 for (p = 0; p < 0x100; p++) 00511 if (pages[p]) { 00512 int j1_min = 32; 00513 int j1_max = -1; 00514 for (j1 = 0; j1 < 32; j1++) 00515 for (j2 = 0; j2 < 8; j2++) 00516 if (uni2charset[256*p+8*j1+j2] != 0) { 00517 if (j1_min > j1) j1_min = j1; 00518 if (j1_max < j1) j1_max = j1; 00519 } 00520 printf("static const unsigned char %s_page%02x[%d] = {\n", 00521 name, p, 8*(j1_max-j1_min+1)); 00522 for (j1 = j1_min; j1 <= j1_max; j1++) { 00523 printf(" "); 00524 for (j2 = 0; j2 < 8; j2++) 00525 printf("0x%02x, ", uni2charset[256*p+8*j1+j2]); 00526 printf("/""* 0x%02x-0x%02x *""/\n", 8*j1, 8*j1+7); 00527 } 00528 printf("};\n"); 00529 } 00530 printf("\n"); 00531 00532 } 00533 #endif 00534 00535 exit(0); 00536 }
1.7.3
