1 /* $NetBSD: init.c,v 1.5 2001/02/19 20:50:17 jdolecek Exp $ */ 2 3 /*- 4 * Copyright (c) 1993 5 * The Regents of the University of California. All rights reserved. 6 * 7 * This code is derived from software contributed to Berkeley by 8 * Peter McIlroy. 9 * 10 * Redistribution and use in source and binary forms, with or without 11 * modification, are permitted provided that the following conditions 12 * are met: 13 * 1. Redistributions of source code must retain the above copyright 14 * notice, this list of conditions and the following disclaimer. 15 * 2. Redistributions in binary form must reproduce the above copyright 16 * notice, this list of conditions and the following disclaimer in the 17 * documentation and/or other materials provided with the distribution. 18 * 3. All advertising materials mentioning features or use of this software 19 * must display the following acknowledgement: 20 * This product includes software developed by the University of 21 * California, Berkeley and its contributors. 22 * 4. Neither the name of the University nor the names of its contributors 23 * may be used to endorse or promote products derived from this software 24 * without specific prior written permission. 25 * 26 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND 27 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 28 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 29 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE 30 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 31 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 32 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 33 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 34 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 35 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 36 * SUCH DAMAGE. 37 */ 38 39 #include "sort.h" 40 41 #ifndef lint 42 __RCSID("$NetBSD: init.c,v 1.5 2001/02/19 20:50:17 jdolecek Exp $"); 43 __SCCSID("@(#)init.c 8.1 (Berkeley) 6/6/93"); 44 #endif /* not lint */ 45 46 #include <ctype.h> 47 #include <string.h> 48 49 static void insertcol __P((struct field *)); 50 static const char *setcolumn __P((const char *, struct field *, int)); 51 int setfield __P((const char *, struct field *, int)); 52 53 extern struct coldesc clist[(ND+1)*2]; 54 extern int ncols; 55 u_char gweights[NBINS]; 56 57 /* 58 * masks of ignored characters. Alltable is 256 ones. 59 */ 60 static u_char alltable[NBINS], dtable[NBINS], itable[NBINS]; 61 62 /* 63 * clist (list of columns which correspond to one or more icol or tcol) 64 * is in increasing order of columns. 65 * Fields are kept in increasing order of fields. 66 */ 67 68 /* 69 * keep clist in order--inserts a column in a sorted array 70 */ 71 static void 72 insertcol(field) 73 struct field *field; 74 { 75 int i; 76 for (i = 0; i < ncols; i++) 77 if (field->icol.num <= clist[i].num) 78 break; 79 if (field->icol.num != clist[i].num) { 80 memmove(clist+i+1, clist+i, sizeof(COLDESC)*(ncols-i)); 81 clist[i].num = field->icol.num; 82 ncols++; 83 } 84 if (field->tcol.num && field->tcol.num != field->icol.num) { 85 for (i = 0; i < ncols; i++) 86 if (field->tcol.num <= clist[i].num) 87 break; 88 if (field->tcol.num != clist[i].num) { 89 memmove(clist+i+1, clist+i,sizeof(COLDESC)*(ncols-i)); 90 clist[i].num = field->tcol.num; 91 ncols++; 92 } 93 } 94 } 95 96 /* 97 * matches fields with the appropriate columns--n^2 but who cares? 98 */ 99 void 100 fldreset(fldtab) 101 struct field *fldtab; 102 { 103 int i; 104 fldtab[0].tcol.p = clist+ncols-1; 105 for (++fldtab; fldtab->icol.num; ++fldtab) { 106 for (i = 0; fldtab->icol.num != clist[i].num; i++) 107 ; 108 fldtab->icol.p = clist + i; 109 if (!fldtab->tcol.num) 110 continue; 111 for (i = 0; fldtab->tcol.num != clist[i].num; i++) 112 ; 113 fldtab->tcol.p = clist + i; 114 } 115 } 116 117 /* 118 * interprets a column in a -k field 119 */ 120 static const char * 121 setcolumn(pos, cur_fld, gflag) 122 const char *pos; 123 struct field *cur_fld; 124 int gflag; 125 { 126 struct column *col; 127 int tmp; 128 col = cur_fld->icol.num ? (&(*cur_fld).tcol) : (&(*cur_fld).icol); 129 pos += sscanf(pos, "%d", &(col->num)); 130 while (isdigit(*pos)) 131 pos++; 132 if (col->num <= 0 && !(col->num == 0 && col == &(cur_fld->tcol))) 133 errx(2, "field numbers must be positive"); 134 if (*pos == '.') { 135 if (!col->num) 136 errx(2, "cannot indent end of line"); 137 pos += sscanf(++pos, "%d", &(col->indent)); 138 while (isdigit(*pos)) 139 pos++; 140 if (&cur_fld->icol == col) 141 col->indent--; 142 if (col->indent < 0) 143 errx(2, "illegal offset"); 144 } 145 if (optval(*pos, cur_fld->tcol.num)) 146 while ((tmp = optval(*pos, cur_fld->tcol.num))) { 147 cur_fld->flags |= tmp; 148 pos++; 149 } 150 if (cur_fld->icol.num == 0) 151 cur_fld->icol.num = 1; 152 return (pos); 153 } 154 155 int 156 setfield(pos, cur_fld, gflag) 157 const char *pos; 158 struct field *cur_fld; 159 int gflag; 160 { 161 static int nfields = 0; 162 int tmp; 163 164 if (++nfields == ND) 165 errx(2, "too many sort keys. (Limit is %d)", ND-1); 166 167 cur_fld->weights = ascii; 168 cur_fld->mask = alltable; 169 170 pos = setcolumn(pos, cur_fld, gflag); 171 if (*pos == '\0') /* key extends to EOL. */ 172 cur_fld->tcol.num = 0; 173 else { 174 if (*pos != ',') 175 errx(2, "illegal field descriptor"); 176 setcolumn((++pos), cur_fld, gflag); 177 } 178 if (!cur_fld->flags) 179 cur_fld->flags = gflag; 180 tmp = cur_fld->flags; 181 182 /* 183 * Assign appropriate mask table and weight table. 184 * If the global weights are reversed, the local field 185 * must be "re-reversed". 186 */ 187 if (((tmp & R) ^ (gflag & R)) && (tmp & F)) 188 cur_fld->weights = RFtable; 189 else if (tmp & F) 190 cur_fld->weights = Ftable; 191 else if ((tmp & R) ^ (gflag & R)) 192 cur_fld->weights = Rascii; 193 194 if (tmp & I) 195 cur_fld->mask = itable; 196 else if (tmp & D) 197 cur_fld->mask = dtable; 198 199 cur_fld->flags |= (gflag & (BI | BT)); 200 if (!cur_fld->tcol.indent) /* BT has no meaning at end of field */ 201 cur_fld->flags &= ~BT; 202 203 if (cur_fld->tcol.num && !(!(cur_fld->flags & BI) 204 && cur_fld->flags & BT) && (cur_fld->tcol.num <= cur_fld->icol.num 205 && cur_fld->tcol.indent < cur_fld->icol.indent)) 206 errx(2, "fields out of order"); 207 insertcol(cur_fld); 208 return (cur_fld->tcol.num); 209 } 210 211 int 212 optval(desc, tcolflag) 213 int desc, tcolflag; 214 { 215 switch(desc) { 216 case 'b': 217 if (!tcolflag) 218 return (BI); 219 else 220 return (BT); 221 case 'd': return (D); 222 case 'f': return (F); 223 case 'i': return (I); 224 case 'n': return (N); 225 case 'r': return (R); 226 default: return (0); 227 } 228 } 229 230 void 231 fixit(argc, argv) 232 int *argc; 233 char **argv; 234 { 235 int i, j, v, w, x; 236 static char vbuf[ND*20], *vpos, *tpos; 237 vpos = vbuf; 238 239 for (i = 1; i < *argc; i++) { 240 if (argv[i][0] == '+') { 241 tpos = argv[i]+1; 242 argv[i] = vpos; 243 vpos += sprintf(vpos, "-k"); 244 tpos += sscanf(tpos, "%d", &v); 245 while (isdigit(*tpos)) 246 tpos++; 247 vpos += sprintf(vpos, "%d", v+1); 248 if (*tpos == '.') { 249 tpos += sscanf(++tpos, "%d", &x); 250 vpos += sprintf(vpos, ".%d", x+1); 251 } 252 while (*tpos) 253 *vpos++ = *tpos++; 254 vpos += sprintf(vpos, ","); 255 if (argv[i+1] && 256 argv[i+1][0] == '-' && isdigit(argv[i+1][1])) { 257 tpos = argv[i+1] + 1; 258 tpos += sscanf(tpos, "%d", &w); 259 while (isdigit(*tpos)) 260 tpos++; 261 x = 0; 262 if (*tpos == '.') { 263 tpos += sscanf(++tpos, "%d", &x); 264 while (isdigit(*tpos)) 265 tpos++; 266 } 267 if (x) { 268 vpos += sprintf(vpos, "%d", w+1); 269 vpos += sprintf(vpos, ".%d", x); 270 } else 271 vpos += sprintf(vpos, "%d", w); 272 while (*tpos) 273 *vpos++ = *tpos++; 274 for (j= i+1; j < *argc; j++) 275 argv[j] = argv[j+1]; 276 *argc -= 1; 277 } 278 } 279 } 280 } 281 282 /* 283 * ascii, Rascii, Ftable, and RFtable map 284 * REC_D -> REC_D; {not REC_D} -> {not REC_D}. 285 * gweights maps REC_D -> (0 or 255); {not REC_D} -> {not gweights[REC_D]}. 286 * Note: when sorting in forward order, to encode character zero in a key, 287 * use \001\001; character 1 becomes \001\002. In this case, character 0 288 * is reserved for the field delimiter. Analagously for -r (fld_d = 255). 289 * Note: this is only good for ASCII sorting. For different LC 's, 290 * all bets are off. See also num_init in number.c 291 */ 292 void 293 settables(gflags) 294 int gflags; 295 { 296 u_char *wts; 297 int i, incr; 298 for (i=0; i < 256; i++) { 299 ascii[i] = i; 300 if (i > REC_D && i < 255 - REC_D+1) 301 Rascii[i] = 255 - i + 1; 302 else 303 Rascii[i] = 255 - i; 304 if (islower(i)) { 305 Ftable[i] = Ftable[toupper(i)]; 306 RFtable[i] = RFtable[toupper(i)]; 307 } else if (REC_D>= 'A' && REC_D < 'Z' && i < 'a' && i > REC_D) { 308 Ftable[i] = i + 1; 309 RFtable[i] = Rascii[i] - 1; 310 } else { 311 Ftable[i] = i; 312 RFtable[i] = Rascii[i]; 313 } 314 alltable[i] = 1; 315 316 if (i == '\n' || isprint(i)) 317 itable[i] = 1; 318 else 319 itable[i] = 0; 320 321 if (i == '\n' || i == '\t' || i == ' ' || isalnum(i)) 322 dtable[i] = 1; 323 else 324 dtable[i] = 0; 325 } 326 327 Rascii[REC_D] = RFtable[REC_D] = REC_D; 328 if (isupper(REC_D)) 329 Ftable[tolower(REC_D)]++; 330 331 if ((gflags & R) && !((gflags & F) && SINGL_FLD)) 332 wts = Rascii; 333 else if (!((gflags & F) && SINGL_FLD)) 334 wts = ascii; 335 else if (gflags & R) 336 wts = RFtable; 337 else 338 wts = Ftable; 339 340 memmove(gweights, wts, sizeof(gweights)); 341 incr = (gflags & R) ? -1 : 1; 342 for (i = 0; i < REC_D; i++) 343 gweights[i] += incr; 344 gweights[REC_D] = ((gflags & R) ? 255 : 0); 345 if (SINGL_FLD && (gflags & F)) { 346 for (i = 0; i < REC_D; i++) { 347 ascii[i] += incr; 348 Rascii[i] += incr; 349 } 350 ascii[REC_D] = Rascii[REC_D] = gweights[REC_D]; 351 } 352 } 353