sort.c revision 1.13
1 /* $NetBSD: sort.c,v 1.13 2001/01/11 14:05:24 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 /* Sort sorts a file using an optional user-defined key. 40 * Sort uses radix sort for internal sorting, and allows 41 * a choice of merge sort and radix sort for external sorting. 42 */ 43 44 #include "sort.h" 45 #include "fsort.h" 46 #include "pathnames.h" 47 48 #ifndef lint 49 __COPYRIGHT("@(#) Copyright (c) 1993\n\ 50 The Regents of the University of California. All rights reserved.\n"); 51 #endif /* not lint */ 52 53 #ifndef lint 54 __RCSID("$NetBSD: sort.c,v 1.13 2001/01/11 14:05:24 jdolecek Exp $"); 55 __SCCSID("@(#)sort.c 8.1 (Berkeley) 6/6/93"); 56 #endif /* not lint */ 57 58 #include <paths.h> 59 #include <signal.h> 60 #include <stdlib.h> 61 #include <string.h> 62 #include <unistd.h> 63 #include <locale.h> 64 65 int REC_D = '\n'; 66 u_char d_mask[NBINS]; /* flags for rec_d, field_d, <blank> */ 67 /* 68 * weight tables. Gweights is one of ascii, Rascii.. 69 * modified to weight rec_d = 0 (or 255) 70 */ 71 extern u_char gweights[NBINS]; 72 u_char ascii[NBINS], Rascii[NBINS], RFtable[NBINS], Ftable[NBINS]; 73 /* 74 * masks of ignored characters. Alltable is 256 ones 75 */ 76 u_char dtable[NBINS], itable[NBINS], alltable[NBINS]; 77 int SINGL_FLD = 0, SEP_FLAG = 0, UNIQUE = 0; 78 struct coldesc clist[(ND+1)*2]; 79 int ncols = 0; 80 extern struct coldesc clist[(ND+1)*2]; 81 extern int ncols; 82 83 /* 84 * Default to stable sort. 85 */ 86 int stable_sort = 1; 87 88 char toutpath[_POSIX_PATH_MAX]; 89 90 static void cleanup __P((void)); 91 static void onsignal __P((int)); 92 static void usage __P((const char *)); 93 94 int main __P((int argc, char **argv)); 95 96 int 97 main(argc, argv) 98 int argc; 99 char *argv[]; 100 { 101 get_func_t get; 102 int ch, i, stdinflag = 0, tmp = 0; 103 char cflag = 0, mflag = 0; 104 char *outfile, *outpath = 0; 105 struct field fldtab[ND+2], *ftpos; 106 struct filelist filelist; 107 FILE *outfp = NULL; 108 109 setlocale(LC_ALL, ""); 110 111 memset(fldtab, 0, (ND+2)*sizeof(struct field)); 112 memset(d_mask, 0, NBINS); 113 d_mask[REC_D = '\n'] = REC_D_F; 114 SINGL_FLD = SEP_FLAG = 0; 115 d_mask['\t'] = d_mask[' '] = BLANK | FLD_D; 116 ftpos = fldtab; 117 118 fixit(&argc, argv); 119 120 while ((ch = getopt(argc, argv, "bcdfik:mHno:rsSt:T:ux")) != -1) { 121 switch (ch) { 122 case 'b': 123 fldtab->flags |= BI | BT; 124 break; 125 case 'c': 126 cflag = 1; 127 break; 128 case 'd': case 'f': case 'i': case 'n': case 'r': 129 tmp |= optval(ch, 0); 130 if (tmp & R && tmp & F) 131 fldtab->weights = RFtable; 132 else if (tmp & F) 133 fldtab->weights = Ftable; 134 else if(tmp & R) 135 fldtab->weights = Rascii; 136 fldtab->flags |= tmp; 137 break; 138 case 'H': 139 PANIC = 0; 140 break; 141 case 'k': 142 setfield(optarg, ++ftpos, fldtab->flags); 143 break; 144 case 'm': 145 mflag = 1; 146 break; 147 case 'o': 148 outpath = optarg; 149 break; 150 case 's': 151 /* for GNU sort compatibility (this is our default) */ 152 stable_sort = 1; 153 break; 154 case 'S': 155 stable_sort = 0; 156 break; 157 case 't': 158 if (SEP_FLAG) 159 usage("multiple field delimiters"); 160 SEP_FLAG = 1; 161 d_mask[' '] &= ~FLD_D; 162 d_mask['\t'] &= ~FLD_D; 163 d_mask[(u_char)*optarg] |= FLD_D; 164 if (d_mask[(u_char)*optarg] & REC_D_F) 165 err(2, "record/field delimiter clash"); 166 break; 167 case 'T': 168 if (REC_D != '\n') 169 usage("multiple record delimiters"); 170 if ('\n' == (REC_D = *optarg)) 171 break; 172 d_mask['\n'] = d_mask[' ']; 173 d_mask[REC_D] = REC_D_F; 174 break; 175 case 'u': 176 UNIQUE = 1; 177 break; 178 case '?': 179 default: 180 usage(""); 181 } 182 } 183 if (cflag && argc > optind+1) 184 errx(2, "too many input files for -c option"); 185 if (argc - 2 > optind && !strcmp(argv[argc-2], "-o")) { 186 outpath = argv[argc-1]; 187 argc -= 2; 188 } 189 if (mflag && argc - optind > (MAXFCT - (16+1))*16) 190 errx(2, "too many input files for -m option"); 191 for (i = optind; i < argc; i++) { 192 /* allow one occurrence of /dev/stdin */ 193 if (!strcmp(argv[i], "-") || !strcmp(argv[i], _PATH_STDIN)) { 194 if (stdinflag) 195 warnx("ignoring extra \"%s\" in file list", 196 argv[i]); 197 else 198 stdinflag = 1; 199 } else if ((ch = access(argv[i], R_OK))) 200 err(2, "%s", argv[i]); 201 } 202 if (!(fldtab->flags & (I|D|N) || fldtab[1].icol.num)) { 203 SINGL_FLD = 1; 204 fldtab[0].icol.num = 1; 205 } else { 206 if (!fldtab[1].icol.num) { 207 fldtab[0].flags &= ~(BI|BT); 208 setfield("1", ++ftpos, fldtab->flags); 209 } 210 fldreset(fldtab); 211 fldtab[0].flags &= ~F; 212 } 213 settables(fldtab[0].flags); 214 num_init(); 215 fldtab->weights = gweights; 216 if (optind == argc) { 217 static const char * const names[] = { _PATH_STDIN, NULL }; 218 219 filelist.names = names; 220 optind--; 221 } else 222 filelist.names = (const char * const *) &argv[optind]; 223 if (SINGL_FLD) 224 get = makeline; 225 else 226 get = makekey; 227 if (cflag) { 228 order(&filelist, get, fldtab); 229 /* NOT REACHED */ 230 } 231 if (!outpath) { 232 (void)snprintf(toutpath, 233 sizeof(toutpath), "%sstdout", _PATH_DEV); 234 outfile = outpath = toutpath; 235 } else if (!(ch = access(outpath, 0)) && 236 strncmp(_PATH_DEV, outpath, 5)) { 237 struct sigaction act = {0}; 238 int sigtable[] = {SIGHUP, SIGINT, SIGPIPE, SIGXCPU, SIGXFSZ, 239 SIGVTALRM, SIGPROF, 0}; 240 int outfd; 241 errno = 0; 242 if (access(outpath, W_OK)) 243 err(2, "%s", outpath); 244 act.sa_handler = onsignal; 245 act.sa_flags = SA_RESTART | SA_RESETHAND; 246 (void)snprintf(toutpath, sizeof(toutpath), "%sXXXX", outpath); 247 if ((outfd = mkstemp(toutpath)) < 0 || 248 (outfp = fdopen(outfd, "w")) == 0) 249 err(2, "temporary file %s", toutpath); 250 outfile = toutpath; 251 (void)atexit(cleanup); 252 for (i = 0; sigtable[i]; ++i) /* always unlink toutpath */ 253 sigaction(sigtable[i], &act, 0); 254 } else 255 outfile = outpath; 256 257 if (outfp == NULL && (outfp = fopen(outfile, "w")) == NULL) 258 err(2, "output file %s", outfile); 259 260 if (mflag) { 261 fmerge(-1, 0, &filelist, argc-optind, get, outfp, putline, 262 fldtab); 263 } else 264 fsort(-1, 0, 0, &filelist, argc-optind, outfp, fldtab); 265 266 if (outfile != outpath) { 267 if (access(outfile, 0)) 268 err(2, "%s", outfile); 269 (void)unlink(outpath); 270 if (link(outfile, outpath)) 271 err(2, "cannot link %s: output left in %s", 272 outpath, outfile); 273 (void)unlink(outfile); 274 } 275 exit(0); 276 } 277 278 static void 279 onsignal(sig) 280 int sig; 281 { 282 cleanup(); 283 } 284 285 static void 286 cleanup() 287 { 288 if (toutpath[0]) 289 (void)unlink(toutpath); 290 } 291 292 static void 293 usage(msg) 294 const char *msg; 295 { 296 if (msg) 297 (void)fprintf(stderr, "sort: %s\n", msg); 298 (void)fprintf(stderr, "usage: [-o output] [-cmubdfinr] [-t char] "); 299 (void)fprintf(stderr, "[-T char] [-k keydef] ... [files]\n"); 300 exit(2); 301 } 302
Indexes created Tue Sep 23 13:09:54 GMT 2025