1 /* $NetBSD: init.c,v 1.8 2003/08/07 11:15:54 agc 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. Neither the name of the University nor the names of its contributors 19 * may be used to endorse or promote products derived from this software 20 * without specific prior written permission. 21 * 22 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND 23 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 24 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 25 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE 26 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 27 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 28 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 29 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 30 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 31 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 32 * SUCH DAMAGE. 33 */ 34 35 #include "sort.h" 36 37 #ifndef lint 38 __RCSID("$NetBSD: init.c,v 1.8 2003/08/07 11:15:54 agc Exp $"); 39 __SCCSID("@(#)init.c 8.1 (Berkeley) 6/6/93"); 40 #endif /* not lint */ 41 42 #include <ctype.h> 43 #include <string.h> 44 45 static void insertcol __P((struct field *)); 46 static const char *setcolumn __P((const char *, struct field *, int)); 47 int setfield __P((const char *, struct field *, int)); 48 49 u_char gweights[NBINS]; 50 51 /* 52 * masks of ignored characters. Alltable is 256 ones. 53 */ 54 static u_char alltable[NBINS], dtable[NBINS], itable[NBINS]; 55 56 /* 57 * clist (list of columns which correspond to one or more icol or tcol) 58 * is in increasing order of columns. 59 * Fields are kept in increasing order of fields. 60 */ 61 62 /* 63 * keep clist in order--inserts a column in a sorted array 64 */ 65 static void 66 insertcol(field) 67 struct field *field; 68 { 69 int i; 70 for (i = 0; i < ncols; i++) 71 if (field->icol.num <= clist[i].num) 72 break; 73 if (field->icol.num != clist[i].num) { 74 memmove(clist+i+1, clist+i, sizeof(COLDESC)*(ncols-i)); 75 clist[i].num = field->icol.num; 76 ncols++; 77 } 78 if (field->tcol.num && field->tcol.num != field->icol.num) { 79 for (i = 0; i < ncols; i++) 80 if (field->tcol.num <= clist[i].num) 81 break; 82 if (field->tcol.num != clist[i].num) { 83 memmove(clist+i+1, clist+i,sizeof(COLDESC)*(ncols-i)); 84 clist[i].num = field->tcol.num; 85 ncols++; 86 } 87 } 88 } 89 90 /* 91 * matches fields with the appropriate columns--n^2 but who cares? 92 */ 93 void 94 fldreset(fldtab) 95 struct field *fldtab; 96 { 97 int i; 98 fldtab[0].tcol.p = clist+ncols-1; 99 for (++fldtab; fldtab->icol.num; ++fldtab) { 100 for (i = 0; fldtab->icol.num != clist[i].num; i++) 101 ; 102 fldtab->icol.p = clist + i; 103 if (!fldtab->tcol.num) 104 continue; 105 for (i = 0; fldtab->tcol.num != clist[i].num; i++) 106 ; 107 fldtab->tcol.p = clist + i; 108 } 109 } 110 111 /* 112 * interprets a column in a -k field 113 */ 114 static const char * 115 setcolumn(pos, cur_fld, gflag) 116 const char *pos; 117 struct field *cur_fld; 118 int gflag; 119 { 120 struct column *col; 121 int tmp; 122 col = cur_fld->icol.num ? (&(*cur_fld).tcol) : (&(*cur_fld).icol); 123 pos += sscanf(pos, "%d", &(col->num)); 124 while (isdigit(*pos)) 125 pos++; 126 if (col->num <= 0 && !(col->num == 0 && col == &(cur_fld->tcol))) 127 errx(2, "field numbers must be positive"); 128 if (*pos == '.') { 129 if (!col->num) 130 errx(2, "cannot indent end of line"); 131 ++pos; 132 pos += sscanf(pos, "%d", &(col->indent)); 133 while (isdigit(*pos)) 134 pos++; 135 if (&cur_fld->icol == col) 136 col->indent--; 137 if (col->indent < 0) 138 errx(2, "illegal offset"); 139 } 140 if (optval(*pos, cur_fld->tcol.num)) 141 while ((tmp = optval(*pos, cur_fld->tcol.num))) { 142 cur_fld->flags |= tmp; 143 pos++; 144 } 145 if (cur_fld->icol.num == 0) 146 cur_fld->icol.num = 1; 147 return (pos); 148 } 149 150 int 151 setfield(pos, cur_fld, gflag) 152 const char *pos; 153 struct field *cur_fld; 154 int gflag; 155 { 156 static int nfields = 0; 157 int tmp; 158 159 if (++nfields == ND) 160 errx(2, "too many sort keys. (Limit is %d)", ND-1); 161 162 cur_fld->weights = ascii; 163 cur_fld->mask = alltable; 164 165 pos = setcolumn(pos, cur_fld, gflag); 166 if (*pos == '\0') /* key extends to EOL. */ 167 cur_fld->tcol.num = 0; 168 else { 169 if (*pos != ',') 170 errx(2, "illegal field descriptor"); 171 setcolumn((++pos), cur_fld, gflag); 172 } 173 if (!cur_fld->flags) 174 cur_fld->flags = gflag; 175 tmp = cur_fld->flags; 176 177 /* 178 * Assign appropriate mask table and weight table. 179 * If the global weights are reversed, the local field 180 * must be "re-reversed". 181 */ 182 if (((tmp & R) ^ (gflag & R)) && (tmp & F)) 183 cur_fld->weights = RFtable; 184 else if (tmp & F) 185 cur_fld->weights = Ftable; 186 else if ((tmp & R) ^ (gflag & R)) 187 cur_fld->weights = Rascii; 188 189 if (tmp & I) 190 cur_fld->mask = itable; 191 else if (tmp & D) 192 cur_fld->mask = dtable; 193 194 cur_fld->flags |= (gflag & (BI | BT)); 195 if (!cur_fld->tcol.indent) /* BT has no meaning at end of field */ 196 cur_fld->flags &= ~BT; 197 198 if (cur_fld->tcol.num && !(!(cur_fld->flags & BI) 199 && cur_fld->flags & BT) && (cur_fld->tcol.num <= cur_fld->icol.num 200 && cur_fld->tcol.indent < cur_fld->icol.indent)) 201 errx(2, "fields out of order"); 202 insertcol(cur_fld); 203 return (cur_fld->tcol.num); 204 } 205 206 int 207 optval(desc, tcolflag) 208 int desc, tcolflag; 209 { 210 switch(desc) { 211 case 'b': 212 if (!tcolflag) 213 return (BI); 214 else 215 return (BT); 216 case 'd': return (D); 217 case 'f': return (F); 218 case 'i': return (I); 219 case 'n': return (N); 220 case 'r': return (R); 221 default: return (0); 222 } 223 } 224 225 void 226 fixit(argc, argv) 227 int *argc; 228 char **argv; 229 { 230 int i, j, v, w, x; 231 static char vbuf[ND*20], *vpos, *tpos; 232 vpos = vbuf; 233 234 for (i = 1; i < *argc; i++) { 235 if (argv[i][0] == '+') { 236 tpos = argv[i]+1; 237 argv[i] = vpos; 238 vpos += sprintf(vpos, "-k"); 239 tpos += sscanf(tpos, "%d", &v); 240 while (isdigit(*tpos)) 241 tpos++; 242 vpos += sprintf(vpos, "%d", v+1); 243 if (*tpos == '.') { 244 ++tpos; 245 tpos += sscanf(tpos, "%d", &x); 246 vpos += sprintf(vpos, ".%d", x+1); 247 } 248 while (*tpos) 249 *vpos++ = *tpos++; 250 vpos += sprintf(vpos, ","); 251 if (argv[i+1] && 252 argv[i+1][0] == '-' && isdigit(argv[i+1][1])) { 253 tpos = argv[i+1] + 1; 254 tpos += sscanf(tpos, "%d", &w); 255 while (isdigit(*tpos)) 256 tpos++; 257 x = 0; 258 if (*tpos == '.') { 259 ++tpos; 260 tpos += sscanf(tpos, "%d", &x); 261 while (isdigit(*tpos)) 262 tpos++; 263 } 264 if (x) { 265 vpos += sprintf(vpos, "%d", w+1); 266 vpos += sprintf(vpos, ".%d", x); 267 } else 268 vpos += sprintf(vpos, "%d", w); 269 while (*tpos) 270 *vpos++ = *tpos++; 271 for (j= i+1; j < *argc; j++) 272 argv[j] = argv[j+1]; 273 *argc -= 1; 274 } 275 } 276 } 277 } 278 279 /* 280 * ascii, Rascii, Ftable, and RFtable map 281 * REC_D -> REC_D; {not REC_D} -> {not REC_D}. 282 * gweights maps REC_D -> (0 or 255); {not REC_D} -> {not gweights[REC_D]}. 283 * Note: when sorting in forward order, to encode character zero in a key, 284 * use \001\001; character 1 becomes \001\002. In this case, character 0 285 * is reserved for the field delimiter. Analagously for -r (fld_d = 255). 286 * Note: this is only good for ASCII sorting. For different LC 's, 287 * all bets are off. See also num_init in number.c 288 */ 289 void 290 settables(gflags) 291 int gflags; 292 { 293 u_char *wts; 294 int i, incr; 295 for (i=0; i < 256; i++) { 296 ascii[i] = i; 297 if (i > REC_D && i < 255 - REC_D+1) 298 Rascii[i] = 255 - i + 1; 299 else 300 Rascii[i] = 255 - i; 301 if (islower(i)) { 302 Ftable[i] = Ftable[toupper(i)]; 303 RFtable[i] = RFtable[toupper(i)]; 304 } else if (REC_D>= 'A' && REC_D < 'Z' && i < 'a' && i > REC_D) { 305 Ftable[i] = i + 1; 306 RFtable[i] = Rascii[i] - 1; 307 } else { 308 Ftable[i] = i; 309 RFtable[i] = Rascii[i]; 310 } 311 alltable[i] = 1; 312 313 if (i == '\n' || isprint(i)) 314 itable[i] = 1; 315 else 316 itable[i] = 0; 317 318 if (i == '\n' || i == '\t' || i == ' ' || isalnum(i)) 319 dtable[i] = 1; 320 else 321 dtable[i] = 0; 322 } 323 324 Rascii[REC_D] = RFtable[REC_D] = REC_D; 325 if (isupper(REC_D)) 326 Ftable[tolower(REC_D)]++; 327 328 if ((gflags & R) && !((gflags & F) && SINGL_FLD)) 329 wts = Rascii; 330 else if (!((gflags & F) && SINGL_FLD)) 331 wts = ascii; 332 else if (gflags & R) 333 wts = RFtable; 334 else 335 wts = Ftable; 336 337 memmove(gweights, wts, sizeof(gweights)); 338 incr = (gflags & R) ? -1 : 1; 339 for (i = 0; i < REC_D; i++) 340 gweights[i] += incr; 341 gweights[REC_D] = ((gflags & R) ? 255 : 0); 342 if (SINGL_FLD && (gflags & F)) { 343 for (i = 0; i < REC_D; i++) { 344 ascii[i] += incr; 345 Rascii[i] += incr; 346 } 347 ascii[REC_D] = Rascii[REC_D] = gweights[REC_D]; 348 } 349 } 350