/* udbuild.c - build udict & table */ #include "all.h" #define UCHAR unsigned char long lseek(); char *stat[228] = { "A", /* 928 */ "C", /* 721 */ "D", /* 5494 */ "E", /* 5967 */ "G", /* 3836 */ "H", /* 733 */ "K", /* 526 */ "L", /* 1735 */ "M", /* 761 */ "N", /* 3115 */ "O", /* 371 */ "P", /* 332 */ "R", /* 2950 */ "S", /* 9498 */ "T", /* 2945 */ "W", /* 176 */ "Y", /* 4542 */ "AB", /* 781 */ "AC", /* 672 */ "AD", /* 417 */ "AG", /* 434 */ "AI", /* 436 */ "AK", /* 191 */ "AL", /* 2331 */ "AM", /* 392 */ "AN", /* 2103 */ "AP", /* 328 */ "AR", /* 1369 */ "AS", /* 621 */ "AT", /* 3590 */ "AY", /* 251 */ "BA", /* 233 */ "BE", /* 382 */ "BI", /* 380 */ "BL", /* 1028 */ "BO", /* 282 */ "BU", /* 150 */ "CA", /* 1028 */ "CE", /* 1313 */ "CH", /* 840 */ "CI", /* 825 */ "CK", /* 558 */ "CL", /* 184 */ "CO", /* 404 */ "CR", /* 206 */ "CT", /* 1037 */ "CU", /* 348 */ "CY", /* 164 */ "DA", /* 330 */ "DE", /* 1405 */ "DI", /* 901 */ "DL", /* 357 */ "DO", /* 253 */ "DS", /* 370 */ "DU", /* 156 */ "EA", /* 666 */ "EC", /* 627 */ "ED", /* 4878 */ "EE", /* 340 */ "EF", /* 160 */ "EG", /* 204 */ "EI", /* 147 */ "EL", /* 1177 */ "EM", /* 490 */ "EN", /* 3045 */ "EO", /* 163 */ "EP", /* 226 */ "ER", /* 5692 */ "ES", /* 4732 */ "ET", /* 1054 */ "EV", /* 154 */ "EY", /* 172 */ "FE", /* 273 */ "FF", /* 171 */ "FI", /* 449 */ "FO", /* 165 */ "FU", /* 325 */ "GA", /* 325 */ "GE", /* 1040 */ "GH", /* 327 */ "GI", /* 519 */ "GL", /* 340 */ "GN", /* 164 */ "GO", /* 168 */ "GR", /* 308 */ "GS", /* 275 */ "GU", /* 173 */ "HA", /* 366 */ "HE", /* 1101 */ "HI", /* 606 */ "HO", /* 451 */ "HT", /* 326 */ "IA", /* 1229 */ "IB", /* 340 */ "IC", /* 2201 */ "ID", /* 571 */ "IE", /* 1508 */ "IF", /* 421 */ "IG", /* 418 */ "IL", /* 940 */ "IM", /* 344 */ "IN", /* 5946 */ "IO", /* 2526 */ "IP", /* 346 */ "IR", /* 339 */ "IS", /* 1867 */ "IT", /* 1870 */ "IV", /* 767 */ "IZ", /* 575 */ "KE", /* 717 */ "KI", /* 295 */ "KL", /* 147 */ "KS", /* 241 */ "LA", /* 1143 */ "LD", /* 215 */ "LE", /* 3029 */ "LI", /* 2141 */ "LL", /* 1155 */ "LO", /* 760 */ "LS", /* 415 */ "LT", /* 200 */ "LU", /* 222 */ "LY", /* 2166 */ "MA", /* 813 */ "MB", /* 178 */ "ME", /* 1273 */ "MI", /* 757 */ "MM", /* 155 */ "MO", /* 306 */ "MP", /* 283 */ "MS", /* 215 */ "NA", /* 1111 */ "NC", /* 896 */ "ND", /* 952 */ "NE", /* 2169 */ "NG", /* 4413 */ "NI", /* 1224 */ "NK", /* 168 */ "NN", /* 163 */ "NO", /* 317 */ "NS", /* 1375 */ "NT", /* 2276 */ "OA", /* 150 */ "OC", /* 283 */ "OD", /* 274 */ "OG", /* 432 */ "OI", /* 207 */ "OL", /* 718 */ "OM", /* 582 */ "ON", /* 3487 */ "OO", /* 297 */ "OP", /* 362 */ "OR", /* 1770 */ "OS", /* 523 */ "OT", /* 380 */ "OU", /* 1053 */ "OV", /* 155 */ "OW", /* 275 */ "PA", /* 318 */ "PE", /* 796 */ "PH", /* 382 */ "PI", /* 444 */ "PL", /* 365 */ "PO", /* 423 */ "PP", /* 240 */ "PR", /* 200 */ "PS", /* 209 */ "PT", /* 302 */ "QU", /* 214 */ "RA", /* 1816 */ "RC", /* 252 */ "RD", /* 454 */ "RE", /* 2014 */ "RG", /* 174 */ "RI", /* 2100 */ "RL", /* 219 */ "RM", /* 341 */ "RN", /* 325 */ "RO", /* 932 */ "RR", /* 217 */ "RS", /* 1352 */ "RT", /* 664 */ "RU", /* 261 */ "RY", /* 574 */ "SA", /* 300 */ "SC", /* 270 */ "SE", /* 1376 */ "SH", /* 728 */ "SI", /* 1316 */ "SL", /* 282 */ "SM", /* 364 */ "SO", /* 390 */ "SP", /* 209 */ "SS", /* 1648 */ "ST", /* 2193 */ "SU", /* 191 */ "TA", /* 1295 */ "TE", /* 3770 */ "TH", /* 732 */ "TI", /* 4842 */ "TL", /* 466 */ "TO", /* 1009 */ "TR", /* 749 */ "TS", /* 995 */ "TT", /* 429 */ "TU", /* 547 */ "TY", /* 691 */ "UA", /* 299 */ "UC", /* 248 */ "UD", /* 186 */ "UE", /* 347 */ "UI", /* 259 */ "UL", /* 876 */ "UM", /* 425 */ "UN", /* 364 */ "UP", /* 163 */ "UR", /* 796 */ "US", /* 1273 */ "UT", /* 440 */ "VA", /* 274 */ "VE", /* 1283 */ "VI", /* 396 */ "WA", /* 216 */ "WE", /* 189 */ "YI", /* 159 */ "YS", /* 155 */ "ZA", /* 150 */ "ZE"}; /* 483 */ unsigned long ptable[4000]; UCHAR dbuf[8192],pdict[8192]; long loc,wtot,wdict; UCHAR word[256],lastword[34],trans[34]; int first,second,oldfirst,oldsecond,third,oldthird; unsigned short dindex; int fudict,fudtable,fd = -1; unsigned short ctable[28][28]; int dictlen,dictindex = 0; UCHAR pathdict[80]; unsigned short tsize = 0; main(argc,argv) int argc; char **argv; { UCHAR *udict,*udtable,*p; register int i,j,k,l; int c1,ch,rtn,err; rtn = FALSE; udtable = (UCHAR *)"Udtable"; udict = (UCHAR *)"Udict"; pathdict[0] = 0x00; dindex = 0; puts("Uedit dictionary builder, (C) 1988 by Rick Stiles"); puts(" usage: Udbuild <-ttable> <-ddict>"); if (argc>1) { for (c1=1; c11) { j = pathdict[i-1]; if (j!=':' && j!='/') { pathdict[i] = '/'; i++; pathdict[i] = 0x00; } } } } } strcat(pathdict,"dict. "); unlink(udict); unlink(udtable); fudict = open(udict,O_CREAT | O_RDWR); if (fudict== -1) { printf("Can't create %s\n",udict); goto BADOUT; } oldthird = third = oldfirst = oldsecond = first = second= -1; loc = 0L; wtot = 0L; for (i=0; i<28; i++) for (j=0; j<28; j++) ctable[i][j] = 0; for (i=0; i<228; i++) { p = (UCHAR *)stat[i]; if (*p) *p -= ('A' - 2); if (*(p+1)) *(p+1) -= ('A' - 2); } for (i='A'; i<='Z'; i++) { if (!newdict(i)) goto BADOUT; wdict = 0L; lastword[0]=0x00; word[0]=0x00; while (getWord(i)) { wdict++; if (!buildTable()) goto BADOUT; } wtot += wdict; } OUT: ptable[tsize] = (loc << 11L) | 0x0700; tsize++; if (fudict && dindex>0) if (write(fudict,dbuf,dindex)!=dindex) goto BADOUT; dindex = 0; fudtable = open(udtable,O_CREAT | O_RDWR); if (fudtable == -1) { printf("Can't create %s\n",udtable); goto BADOUT; } ctable[0][0] = 65533; if (write(fudtable,&(ctable[0][0]),sizeof(ctable))!=sizeof(ctable)) goto BADMSG; for (i=0; i<228; i++) if (write(fudtable,stat[i],2)!=2) goto BADMSG; if (write(fudtable,ptable,tsize << 2)!=(tsize<<2)) { BADMSG: close(fudtable); puts("No room for table"); goto BADOUT; } close(fudtable); printf("UDICT: %ld, UDTABLE: %ld, WORDS: %ld\n",loc,(long)(tsize << 2L) + 456L + sizeof(ctable),wtot); rtn = TRUE; BADOUT: if (fudict!=-1) close(fudict); if (fd!=-1) close(fd); if (!rtn) { unlink(udict); unlink(udtable); } exit(0); } getWord(cdict) int cdict; { register UCHAR *p,*q; register int j; MORE: p = pdict + dictindex; while ( dictindex < dictlen && *p==0x00 ) { dictindex++; p++; } if (dictindex>=dictlen) { if (!readMore()) return(FALSE); goto MORE; } MOREWORD: q = word; j = 0; p = pdict + dictindex; while (*p && j<255) { *(q++) = *(p++); j++; dictindex++; if ( dictindex >= dictlen ) { p = pdict; if (!readMore()) { if (j>0) break; return(FALSE); } } } *q = 0x00; if (j>=32 && *p) { msg("long word: ",word); goto MOREWORD; } q = word; first = *(q++); second = *(q++); if (second) third = *q; else third = 0; if ((first + 'A' - 2)!=cdict) { if (cdict=='A' && first==1) goto TRYIT; goto BADORDER; } TRYIT: if (first>oldfirst) goto RTN; if (first==oldfirst) { if (second>oldsecond) goto RTN; if (second==oldsecond && third>=oldthird) goto RTN; } BADORDER: msg("Out of order and skipping:",word); goto MORE; RTN: return(TRUE); } buildTable() { register UCHAR *p; register int i,c1,c2; unsigned long s; i = strlen(word); c1 = first; c2 = second; if (c1==0 || c2==0) goto NOSTORE; if (c1!=oldfirst || c2!=oldsecond || third!=oldthird) { /*======== FLAGS: 0x40: len<3 0x20: len == 3 0x10: START or END of s-rgn ==================== */ s = (loc << 11L) | third; if (i<3) s |= 0x400; else if (i==3) s |= 0x200; if (c1!=oldfirst || c2!=oldsecond) { ctable[c1][c2] = tsize + 1; s |= 0x100; } ptable[tsize] = s; tsize++; } if (i>3) { /* translate & store the word */ i = transWord(i); if ((i + dindex)>=8192) { if (write(fudict,dbuf,dindex)!=dindex) { puts("No room for dictionary"); return(FALSE); } dindex = 0; } movmem(trans,&(dbuf[dindex]),i); dindex += i; loc += i; } NOSTORE: strcpy(lastword,word); oldfirst = first; oldsecond = second; oldthird = third; return(TRUE); } #define MINSTAT 0 #define MAXSTAT 227 #define MAXNULL 16 #define ADDSTAT 28 #define ZERO 0 #define APOSTROPHE 27 #define MINALPHA 1 #define MAXALPHA 26 transWord(len) int len; { register UCHAR *p,*q; register long bits; register int j,n,c1; int k; UCHAR hits[32]; p = &(word[3]); len -= 3; /* chopped off 1st 3 letters */ bits = 0L; hits[len] = 0; for (j=0; j>= 1L; } } else { /* must take it */ TAKEONE: q[n++] = c1; if (c1==0) break; } bits >>= 1L; } return(n); } pair(c1,c2) register int c1,c2; { register UCHAR *r; register int k; for (k=MINSTAT; k<=MAXSTAT; k++) { r = (UCHAR *)stat[k]; if (*r==c1 && *(r+1)==c2) return(k+ADDSTAT); } return(0); } newdict(cdict) int cdict; { register int i; if (fd!=-1) close(fd); i = strlen(pathdict); pathdict[i-1] = cdict; printf("%s\n",pathdict); fd = open(pathdict,O_RDONLY); if (fd == -1) { printf("Can't find %s\n",pathdict); return(FALSE); } readMore(); return(TRUE); } readMore() { register UCHAR *p; register int i,c,d; dictindex = 0; dictlen = 0; d = read(fd,pdict,8192); if (d<1) return(FALSE); dictlen = d; p = pdict; i = 0; while (1) { c = *p; if (c>='a' && c<='z') { c -= ('a' - 2); goto TAKEIT; } if (c>='A' && c<='Z') { c -= ('A' - 2); goto TAKEIT; } if (c<=' ') { c=0; goto TAKEIT; } if (c==39) { c = 1; goto TAKEIT; } printf("Illegal char %c in %s\n",c, pathdict); c = 0; TAKEIT: *p = c; BUMPIT: if (++i>=dictlen) break; p++; } return(TRUE); } msg(x,y) UCHAR *x,*y; { register UCHAR *z; UCHAR zz[80]; strcpy(zz,y); z = zz; while (*z) { if (*z==1) *z = 39; else *z += 63; z++; } printf("%s %s\n",x,zz); }