sort.c revision 1.30
1 /* $NetBSD: sort.c,v 1.30 2002/12/24 13:09:38 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 1.30 2002/12/24 13:09:38 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; 110 111 setlocale(LC_ALL, ""); 112 113 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 many_files(); 118 119 fixit(&argc, argv); 120 if (!(tmpdir = getenv("TMPDIR"))) 121 tmpdir = _PATH_TMP; 122 123 while ((ch = getopt(argc, argv, "bcdfik:mHno:rR:sSt:T:ux")) != -1) { 124 switch (ch) { 125 case 'b': 126 fldtab->flags |= BI | BT; 127 break; 128 case 'c': 129 cflag = 1; 130 break; 131 case 'd': case 'f': case 'i': case 'n': case 'r': 132 tmp |= optval(ch, 0); 133 if ((tmp & R) && (tmp & F)) 134 fldtab->weights = RFtable; 135 else if (tmp & F) 136 fldtab->weights = Ftable; 137 else if (tmp & R) 138 fldtab->weights = Rascii; 139 fldtab->flags |= tmp; 140 break; 141 case 'H': 142 PANIC = 0; 143 break; 144 case 'k': 145 setfield(optarg, ++ftpos, fldtab->flags); 146 break; 147 case 'm': 148 mflag = 1; 149 break; 150 case 'o': 151 outpath = optarg; 152 break; 153 case 's': 154 /* for GNU sort compatibility (this is our default) */ 155 stable_sort = 1; 156 break; 157 case 'S': 158 stable_sort = 0; 159 break; 160 case 't': 161 if (SEP_FLAG) 162 usage("multiple field delimiters"); 163 SEP_FLAG = 1; 164 d_mask[' '] &= ~FLD_D; 165 d_mask['\t'] &= ~FLD_D; 166 d_mask[(u_char)*optarg] |= FLD_D; 167 if (d_mask[(u_char)*optarg] & REC_D_F) 168 errx(2, "record/field delimiter clash"); 169 break; 170 case 'R': 171 if (REC_D != '\n') 172 usage("multiple record delimiters"); 173 if ('\n' == (REC_D = *optarg)) 174 break; 175 d_mask['\n'] = d_mask[' ']; 176 d_mask[REC_D] = REC_D_F; 177 break; 178 case 'T': 179 /* -T tmpdir */ 180 tmpdir = optarg; 181 break; 182 case 'u': 183 UNIQUE = 1; 184 break; 185 case '?': 186 default: 187 usage(NULL); 188 } 189 } 190 if (cflag && argc > optind+1) 191 errx(2, "too many input files for -c option"); 192 if (argc - 2 > optind && !strcmp(argv[argc-2], "-o")) { 193 outpath = argv[argc-1]; 194 argc -= 2; 195 } 196 if (mflag && argc - optind > (MAXFCT - (16+1))*16) 197 errx(2, "too many input files for -m option"); 198 for (i = optind; i < argc; i++) { 199 /* allow one occurrence of /dev/stdin */ 200 if (!strcmp(argv[i], "-") || !strcmp(argv[i], _PATH_STDIN)) { 201 if (stdinflag) 202 warnx("ignoring extra \"%s\" in file list", 203 argv[i]); 204 else 205 stdinflag = 1; 206 207 /* change to /dev/stdin if '-' */ 208 if (argv[i][0] == '-') 209 argv[i] = _PATH_STDIN; 210 211 } else if ((ch = access(argv[i], R_OK))) 212 err(2, "%s", argv[i]); 213 } 214 if (!(fldtab->flags & (I|D|N) || fldtab[1].icol.num)) { 215 SINGL_FLD = 1; 216 fldtab[0].icol.num = 1; 217 } else { 218 if (!fldtab[1].icol.num) { 219 fldtab[0].flags &= ~(BI|BT); 220 setfield("1", ++ftpos, fldtab->flags); 221 } 222 fldreset(fldtab); 223 fldtab[0].flags &= ~F; 224 } 225 settables(fldtab[0].flags); 226 num_init(); 227 fldtab->weights = gweights; 228 if (optind == argc) { 229 static const char * const names[] = { _PATH_STDIN, NULL }; 230 231 filelist.names = names; 232 optind--; 233 } else 234 filelist.names = (const char * const *) &argv[optind]; 235 236 if (SINGL_FLD) 237 get = makeline; 238 else 239 get = makekey; 240 241 if (cflag) { 242 order(&filelist, get, fldtab); 243 /* NOT REACHED */ 244 } 245 if (!outpath) { 246 (void)snprintf(toutpath, 247 sizeof(toutpath), "%sstdout", _PATH_DEV); 248 outfile = outpath = toutpath; 249 outfp = stdout; 250 } else if (!(ch = access(outpath, 0)) && 251 strncmp(_PATH_DEV, outpath, 5)) { 252 struct sigaction act; 253 static const int sigtable[] = {SIGHUP, SIGINT, SIGPIPE, 254 SIGXCPU, SIGXFSZ, SIGVTALRM, SIGPROF, 0}; 255 int outfd; 256 errno = 0; 257 if (access(outpath, W_OK)) 258 err(2, "%s", outpath); 259 (void)snprintf(toutpath, sizeof(toutpath), "%sXXXXXX", 260 outpath); 261 if ((outfd = mkstemp(toutpath)) == -1) 262 err(2, "Cannot create temporary file `%s'", toutpath); 263 if ((outfp = fdopen(outfd, "w")) == NULL) 264 err(2, "Cannot open temporary file `%s'", toutpath); 265 outfile = toutpath; 266 (void)atexit(cleanup); 267 act.sa_handler = onsignal; 268 (void) sigemptyset(&act.sa_mask); 269 act.sa_flags = SA_RESTART | SA_RESETHAND; 270 for (i = 0; sigtable[i]; ++i) /* always unlink toutpath */ 271 sigaction(sigtable[i], &act, 0); 272 } else 273 outfile = outpath; 274 275 if (outfp == NULL && (outfp = fopen(outfile, "w")) == NULL) 276 err(2, "output file %s", outfile); 277 278 if (mflag) { 279 fmerge(-1, 0, &filelist, argc-optind, get, outfp, putline, 280 fldtab); 281 } else 282 fsort(-1, 0, 0, &filelist, argc-optind, outfp, fldtab); 283 284 if (outfile != outpath) { 285 if (access(outfile, 0)) 286 err(2, "%s", outfile); 287 (void)unlink(outpath); 288 if (link(outfile, outpath)) 289 err(2, "cannot link %s: output left in %s", 290 outpath, outfile); 291 (void)unlink(outfile); 292 } 293 exit(0); 294 } 295 296 static void 297 onsignal(sig) 298 int sig; 299 { 300 cleanup(); 301 } 302 303 static void 304 cleanup() 305 { 306 if (toutpath[0]) 307 (void)unlink(toutpath); 308 } 309 310 static void 311 usage(msg) 312 const char *msg; 313 { 314 if (msg != NULL) 315 (void)fprintf(stderr, "sort: %s\n", msg); 316 (void)fprintf(stderr, "usage: [-o output] [-cmubdfinrsS] [-t char] "); 317 (void)fprintf(stderr, "[-R char] [-k keydef] ... [files]\n"); 318 exit(2); 319 } 320 321 static void 322 many_files() 323 { 324 #if 0 325 struct rlimit rlp_many_files[1]; 326 327 if (getrlimit(RLIMIT_NOFILE, rlp_many_files) == 0) { 328 rlp_many_files->rlim_cur = rlp_many_files->rlim_max; 329 setrlimit(RLIMIT_NOFILE, rlp_many_files); 330 } 331 #endif 332 } 333
Indexes created Sat Sep 27 14:09:57 GMT 2025