1 /* $NetBSD: init.c,v 1.6 2001/12/31 18:45:04 thorpej 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 1.6 2001/12/31 18:45:04 thorpej 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; 138 pos += sscanf(pos, "%d", &(col->indent)); 139 while (isdigit(*pos)) 140 pos++; 141 if (&cur_fld->icol == col) 142 col->indent--; 143 if (col->indent < 0) 144 errx(2, "illegal offset"); 145 } 146 if (optval(*pos, cur_fld->tcol.num)) 147 while ((tmp = optval(*pos, cur_fld->tcol.num))) { 148 cur_fld->flags |= tmp; 149 pos++; 150 } 151 if (cur_fld->icol.num == 0) 152 cur_fld->icol.num = 1; 153 return (pos); 154 } 155 156 int 157 setfield(pos, cur_fld, gflag) 158 const char *pos; 159 struct field *cur_fld; 160 int gflag; 161 { 162 static int nfields = 0; 163 int tmp; 164 165 if (++nfields == ND) 166 errx(2, "too many sort keys. (Limit is %d)", ND-1); 167 168 cur_fld->weights = ascii; 169 cur_fld->mask = alltable; 170 171 pos = setcolumn(pos, cur_fld, gflag); 172 if (*pos == '\0') /* key extends to EOL. */ 173 cur_fld->tcol.num = 0; 174 else { 175 if (*pos != ',') 176 errx(2, "illegal field descriptor"); 177 setcolumn((++pos), cur_fld, gflag); 178 } 179 if (!cur_fld->flags) 180 cur_fld->flags = gflag; 181 tmp = cur_fld->flags; 182 183 /* 184 * Assign appropriate mask table and weight table. 185 * If the global weights are reversed, the local field 186 * must be "re-reversed". 187 */ 188 if (((tmp & R) ^ (gflag & R)) && (tmp & F)) 189 cur_fld->weights = RFtable; 190 else if (tmp & F) 191 cur_fld->weights = Ftable; 192 else if ((tmp & R) ^ (gflag & R)) 193 cur_fld->weights = Rascii; 194 195 if (tmp & I) 196 cur_fld->mask = itable; 197 else if (tmp & D) 198 cur_fld->mask = dtable; 199 200 cur_fld->flags |= (gflag & (BI | BT)); 201 if (!cur_fld->tcol.indent) /* BT has no meaning at end of field */ 202 cur_fld->flags &= ~BT; 203 204 if (cur_fld->tcol.num && !(!(cur_fld->flags & BI) 205 && cur_fld->flags & BT) && (cur_fld->tcol.num <= cur_fld->icol.num 206 && cur_fld->tcol.indent < cur_fld->icol.indent)) 207 errx(2, "fields out of order"); 208 insertcol(cur_fld); 209 return (cur_fld->tcol.num); 210 } 211 212 int 213 optval(desc, tcolflag) 214 int desc, tcolflag; 215 { 216 switch(desc) { 217 case 'b': 218 if (!tcolflag) 219 return (BI); 220 else 221 return (BT); 222 case 'd': return (D); 223 case 'f': return (F); 224 case 'i': return (I); 225 case 'n': return (N); 226 case 'r': return (R); 227 default: return (0); 228 } 229 } 230 231 void 232 fixit(argc, argv) 233 int *argc; 234 char **argv; 235 { 236 int i, j, v, w, x; 237 static char vbuf[ND*20], *vpos, *tpos; 238 vpos = vbuf; 239 240 for (i = 1; i < *argc; i++) { 241 if (argv[i][0] == '+') { 242 tpos = argv[i]+1; 243 argv[i] = vpos; 244 vpos += sprintf(vpos, "-k"); 245 tpos += sscanf(tpos, "%d", &v); 246 while (isdigit(*tpos)) 247 tpos++; 248 vpos += sprintf(vpos, "%d", v+1); 249 if (*tpos == '.') { 250 ++tpos; 251 tpos += sscanf(tpos, "%d", &x); 252 vpos += sprintf(vpos, ".%d", x+1); 253 } 254 while (*tpos) 255 *vpos++ = *tpos++; 256 vpos += sprintf(vpos, ","); 257 if (argv[i+1] && 258 argv[i+1][0] == '-' && isdigit(argv[i+1][1])) { 259 tpos = argv[i+1] + 1; 260 tpos += sscanf(tpos, "%d", &w); 261 while (isdigit(*tpos)) 262 tpos++; 263 x = 0; 264 if (*tpos == '.') { 265 ++tpos; 266 tpos += sscanf(tpos, "%d", &x); 267 while (isdigit(*tpos)) 268 tpos++; 269 } 270 if (x) { 271 vpos += sprintf(vpos, "%d", w+1); 272 vpos += sprintf(vpos, ".%d", x); 273 } else 274 vpos += sprintf(vpos, "%d", w); 275 while (*tpos) 276 *vpos++ = *tpos++; 277 for (j= i+1; j < *argc; j++) 278 argv[j] = argv[j+1]; 279 *argc -= 1; 280 } 281 } 282 } 283 } 284 285 /* 286 * ascii, Rascii, Ftable, and RFtable map 287 * REC_D -> REC_D; {not REC_D} -> {not REC_D}. 288 * gweights maps REC_D -> (0 or 255); {not REC_D} -> {not gweights[REC_D]}. 289 * Note: when sorting in forward order, to encode character zero in a key, 290 * use \001\001; character 1 becomes \001\002. In this case, character 0 291 * is reserved for the field delimiter. Analagously for -r (fld_d = 255). 292 * Note: this is only good for ASCII sorting. For different LC 's, 293 * all bets are off. See also num_init in number.c 294 */ 295 void 296 settables(gflags) 297 int gflags; 298 { 299 u_char *wts; 300 int i, incr; 301 for (i=0; i < 256; i++) { 302 ascii[i] = i; 303 if (i > REC_D && i < 255 - REC_D+1) 304 Rascii[i] = 255 - i + 1; 305 else 306 Rascii[i] = 255 - i; 307 if (islower(i)) { 308 Ftable[i] = Ftable[toupper(i)]; 309 RFtable[i] = RFtable[toupper(i)]; 310 } else if (REC_D>= 'A' && REC_D < 'Z' && i < 'a' && i > REC_D) { 311 Ftable[i] = i + 1; 312 RFtable[i] = Rascii[i] - 1; 313 } else { 314 Ftable[i] = i; 315 RFtable[i] = Rascii[i]; 316 } 317 alltable[i] = 1; 318 319 if (i == '\n' || isprint(i)) 320 itable[i] = 1; 321 else 322 itable[i] = 0; 323 324 if (i == '\n' || i == '\t' || i == ' ' || isalnum(i)) 325 dtable[i] = 1; 326 else 327 dtable[i] = 0; 328 } 329 330 Rascii[REC_D] = RFtable[REC_D] = REC_D; 331 if (isupper(REC_D)) 332 Ftable[tolower(REC_D)]++; 333 334 if ((gflags & R) && !((gflags & F) && SINGL_FLD)) 335 wts = Rascii; 336 else if (!((gflags & F) && SINGL_FLD)) 337 wts = ascii; 338 else if (gflags & R) 339 wts = RFtable; 340 else 341 wts = Ftable; 342 343 memmove(gweights, wts, sizeof(gweights)); 344 incr = (gflags & R) ? -1 : 1; 345 for (i = 0; i < REC_D; i++) 346 gweights[i] += incr; 347 gweights[REC_D] = ((gflags & R) ? 255 : 0); 348 if (SINGL_FLD && (gflags & F)) { 349 for (i = 0; i < REC_D; i++) { 350 ascii[i] += incr; 351 Rascii[i] += incr; 352 } 353 ascii[REC_D] = Rascii[REC_D] = gweights[REC_D]; 354 } 355 } 356