sort.c revision 1.31
1 /* $NetBSD: sort.c,v 1.31 2002/12/24 14:55:46 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 1.31 2002/12/24 14:55:46 jdolecek Exp $"); 55 __SCCSID("@(#)sort.c 8.1 (Berkeley) 6/6/93"); 56 #endif /* not lint */ 57 58 #include <sys/types.h> 59 #include <sys/time.h> 60 #include <sys/resource.h> 61 62 #include <paths.h> 63 #include <signal.h> 64 #include <stdlib.h> 65 #include <string.h> 66 #include <unistd.h> 67 #include <locale.h> 68 69 int REC_D = '\n'; 70 u_char d_mask[NBINS]; /* flags for rec_d, field_d, <blank> */ 71 72 /* 73 * weight tables. Gweights is one of ascii, Rascii.. 74 * modified to weight rec_d = 0 (or 255) 75 */ 76 u_char ascii[NBINS], Rascii[NBINS], RFtable[NBINS], Ftable[NBINS]; 77 int SINGL_FLD = 0, SEP_FLAG = 0, UNIQUE = 0; 78 struct coldesc clist[(ND+1)*2]; 79 int ncols = 0; 80 81 /* 82 * Default to stable sort. 83 */ 84 int stable_sort = 1; 85 86 static char toutpath[MAXPATHLEN]; 87 static struct field fldtab[ND+2]; 88 89 const char *tmpdir; /* where temporary files should be put */ 90 91 static void cleanup __P((void)); 92 static void onsignal __P((int)); 93 static void usage __P((const char *)); 94 static void many_files __P((void)); 95 96 int main __P((int argc, char **argv)); 97 98 int 99 main(argc, argv) 100 int argc; 101 char *argv[]; 102 { 103 get_func_t get; 104 int ch, i, stdinflag = 0, tmp = 0; 105 char cflag = 0, mflag = 0; 106 char *outfile, *outpath = 0; 107 struct field *ftpos; 108 struct filelist filelist; 109 FILE *outfp = NULL; 110 struct rlimit rl; 111 112 setlocale(LC_ALL, ""); 113 114 /* bump RLIMIT_NOFILE to maximum our hard limit allows */ 115 if (getrlimit(RLIMIT_NOFILE, &rl) < 0) 116 err(2, "getrlimit"); 117 rl.rlim_cur = rl.rlim_max; 118 if (setrlimit(RLIMIT_NOFILE, &rl) < 0) 119 err(2, "setrlimit"); 120 121 d_mask[REC_D = '\n'] = REC_D_F; 122 SINGL_FLD = SEP_FLAG = 0; 123 d_mask['\t'] = d_mask[' '] = BLANK | FLD_D; 124 ftpos = fldtab; 125 many_files(); 126 127 fixit(&argc, argv); 128 if (!(tmpdir = getenv("TMPDIR"))) 129 tmpdir = _PATH_TMP; 130 131 while ((ch = getopt(argc, argv, "bcdfik:mHno:rR:sSt:T:ux")) != -1) { 132 switch (ch) { 133 case 'b': 134 fldtab->flags |= BI | BT; 135 break; 136 case 'c': 137 cflag = 1; 138 break; 139 case 'd': case 'f': case 'i': case 'n': case 'r': 140 tmp |= optval(ch, 0); 141 if ((tmp & R) && (tmp & F)) 142 fldtab->weights = RFtable; 143 else if (tmp & F) 144 fldtab->weights = Ftable; 145 else if (tmp & R) 146 fldtab->weights = Rascii; 147 fldtab->flags |= tmp; 148 break; 149 case 'H': 150 PANIC = 0; 151 break; 152 case 'k': 153 setfield(optarg, ++ftpos, fldtab->flags); 154 break; 155 case 'm': 156 mflag = 1; 157 break; 158 case 'o': 159 outpath = optarg; 160 break; 161 case 's': 162 /* for GNU sort compatibility (this is our default) */ 163 stable_sort = 1; 164 break; 165 case 'S': 166 stable_sort = 0; 167 break; 168 case 't': 169 if (SEP_FLAG) 170 usage("multiple field delimiters"); 171 SEP_FLAG = 1; 172 d_mask[' '] &= ~FLD_D; 173 d_mask['\t'] &= ~FLD_D; 174 d_mask[(u_char)*optarg] |= FLD_D; 175 if (d_mask[(u_char)*optarg] & REC_D_F) 176 errx(2, "record/field delimiter clash"); 177 break; 178 case 'R': 179 if (REC_D != '\n') 180 usage("multiple record delimiters"); 181 if ('\n' == (REC_D = *optarg)) 182 break; 183 d_mask['\n'] = d_mask[' ']; 184 d_mask[REC_D] = REC_D_F; 185 break; 186 case 'T': 187 /* -T tmpdir */ 188 tmpdir = optarg; 189 break; 190 case 'u': 191 UNIQUE = 1; 192 break; 193 case '?': 194 default: 195 usage(NULL); 196 } 197 } 198 if (cflag && argc > optind+1) 199 errx(2, "too many input files for -c option"); 200 if (argc - 2 > optind && !strcmp(argv[argc-2], "-o")) { 201 outpath = argv[argc-1]; 202 argc -= 2; 203 } 204 if (mflag && argc - optind > (MAXFCT - (16+1))*16) 205 errx(2, "too many input files for -m option"); 206 for (i = optind; i < argc; i++) { 207 /* allow one occurrence of /dev/stdin */ 208 if (!strcmp(argv[i], "-") || !strcmp(argv[i], _PATH_STDIN)) { 209 if (stdinflag) 210 warnx("ignoring extra \"%s\" in file list", 211 argv[i]); 212 else 213 stdinflag = 1; 214 215 /* change to /dev/stdin if '-' */ 216 if (argv[i][0] == '-') 217 argv[i] = _PATH_STDIN; 218 219 } else if ((ch = access(argv[i], R_OK))) 220 err(2, "%s", argv[i]); 221 } 222 if (!(fldtab->flags & (I|D|N) || fldtab[1].icol.num)) { 223 SINGL_FLD = 1; 224 fldtab[0].icol.num = 1; 225 } else { 226 if (!fldtab[1].icol.num) { 227 fldtab[0].flags &= ~(BI|BT); 228 setfield("1", ++ftpos, fldtab->flags); 229 } 230 fldreset(fldtab); 231 fldtab[0].flags &= ~F; 232 } 233 settables(fldtab[0].flags); 234 num_init(); 235 fldtab->weights = gweights; 236 if (optind == argc) { 237 static const char * const names[] = { _PATH_STDIN, NULL }; 238 239 filelist.names = names; 240 optind--; 241 } else 242 filelist.names = (const char * const *) &argv[optind]; 243 244 if (SINGL_FLD) 245 get = makeline; 246 else 247 get = makekey; 248 249 if (cflag) { 250 order(&filelist, get, fldtab); 251 /* NOT REACHED */ 252 } 253 if (!outpath) { 254 (void)snprintf(toutpath, 255 sizeof(toutpath), "%sstdout", _PATH_DEV); 256 outfile = outpath = toutpath; 257 outfp = stdout; 258 } else if (!(ch = access(outpath, 0)) && 259 strncmp(_PATH_DEV, outpath, 5)) { 260 struct sigaction act; 261 static const int sigtable[] = {SIGHUP, SIGINT, SIGPIPE, 262 SIGXCPU, SIGXFSZ, SIGVTALRM, SIGPROF, 0}; 263 int outfd; 264 errno = 0; 265 if (access(outpath, W_OK)) 266 err(2, "%s", outpath); 267 (void)snprintf(toutpath, sizeof(toutpath), "%sXXXXXX", 268 outpath); 269 if ((outfd = mkstemp(toutpath)) == -1) 270 err(2, "Cannot create temporary file `%s'", toutpath); 271 if ((outfp = fdopen(outfd, "w")) == NULL) 272 err(2, "Cannot open temporary file `%s'", toutpath); 273 outfile = toutpath; 274 (void)atexit(cleanup); 275 act.sa_handler = onsignal; 276 (void) sigemptyset(&act.sa_mask); 277 act.sa_flags = SA_RESTART | SA_RESETHAND; 278 for (i = 0; sigtable[i]; ++i) /* always unlink toutpath */ 279 sigaction(sigtable[i], &act, 0); 280 } else 281 outfile = outpath; 282 283 if (outfp == NULL && (outfp = fopen(outfile, "w")) == NULL) 284 err(2, "output file %s", outfile); 285 286 if (mflag) { 287 fmerge(-1, 0, &filelist, argc-optind, get, outfp, putline, 288 fldtab); 289 } else 290 fsort(-1, 0, 0, &filelist, argc-optind, outfp, fldtab); 291 292 if (outfile != outpath) { 293 if (access(outfile, 0)) 294 err(2, "%s", outfile); 295 (void)unlink(outpath); 296 if (link(outfile, outpath)) 297 err(2, "cannot link %s: output left in %s", 298 outpath, outfile); 299 (void)unlink(outfile); 300 } 301 exit(0); 302 } 303 304 static void 305 onsignal(sig) 306 int sig; 307 { 308 cleanup(); 309 } 310 311 static void 312 cleanup() 313 { 314 if (toutpath[0]) 315 (void)unlink(toutpath); 316 } 317 318 static void 319 usage(msg) 320 const char *msg; 321 { 322 if (msg != NULL) 323 (void)fprintf(stderr, "sort: %s\n", msg); 324 (void)fprintf(stderr, "usage: [-o output] [-cmubdfinrsS] [-t char] "); 325 (void)fprintf(stderr, "[-R char] [-k keydef] ... [files]\n"); 326 exit(2); 327 } 328 329 static void 330 many_files() 331 { 332 #if 0 333 struct rlimit rlp_many_files[1]; 334 335 if (getrlimit(RLIMIT_NOFILE, rlp_many_files) == 0) { 336 rlp_many_files->rlim_cur = rlp_many_files->rlim_max; 337 setrlimit(RLIMIT_NOFILE, rlp_many_files); 338 } 339 #endif 340 } 341
Indexes created Fri Sep 26 20:09:58 GMT 2025