sort.c revision 1.33
1 /* $NetBSD: sort.c,v 1.33 2003/08/07 11:15:55 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 /* Sort sorts a file using an optional user-defined key. 36 * Sort uses radix sort for internal sorting, and allows 37 * a choice of merge sort and radix sort for external sorting. 38 */ 39 40 #include "sort.h" 41 #include "fsort.h" 42 #include "pathnames.h" 43 44 #ifndef lint 45 __COPYRIGHT("@(#) Copyright (c) 1993\n\ 46 The Regents of the University of California. All rights reserved.\n"); 47 #endif /* not lint */ 48 49 #ifndef lint 50 __RCSID("$NetBSD: sort.c,v 1.33 2003/08/07 11:15:55 agc Exp $"); 51 __SCCSID("@(#)sort.c 8.1 (Berkeley) 6/6/93"); 52 #endif /* not lint */ 53 54 #include <sys/types.h> 55 #include <sys/time.h> 56 #include <sys/resource.h> 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 /* 69 * weight tables. Gweights is one of ascii, Rascii.. 70 * modified to weight rec_d = 0 (or 255) 71 */ 72 u_char ascii[NBINS], Rascii[NBINS], RFtable[NBINS], Ftable[NBINS]; 73 int SINGL_FLD = 0, SEP_FLAG = 0, UNIQUE = 0; 74 struct coldesc clist[(ND+1)*2]; 75 int ncols = 0; 76 77 /* 78 * Default to stable sort. 79 */ 80 int stable_sort = 1; 81 82 static char toutpath[MAXPATHLEN]; 83 static struct field fldtab[ND+2]; 84 85 const char *tmpdir; /* where temporary files should be put */ 86 87 static void cleanup __P((void)); 88 static void onsignal __P((int)); 89 static void usage __P((const char *)); 90 91 int main __P((int argc, char **argv)); 92 93 int 94 main(argc, argv) 95 int argc; 96 char *argv[]; 97 { 98 get_func_t get; 99 int ch, i, stdinflag = 0, tmp = 0; 100 char cflag = 0, mflag = 0; 101 char *outfile, *outpath = 0; 102 struct field *ftpos; 103 struct filelist filelist; 104 FILE *outfp = NULL; 105 struct rlimit rl; 106 107 setlocale(LC_ALL, ""); 108 109 /* bump RLIMIT_NOFILE to maximum our hard limit allows */ 110 if (getrlimit(RLIMIT_NOFILE, &rl) < 0) 111 err(2, "getrlimit"); 112 rl.rlim_cur = rl.rlim_max; 113 if (setrlimit(RLIMIT_NOFILE, &rl) < 0) 114 err(2, "setrlimit"); 115 116 d_mask[REC_D = '\n'] = REC_D_F; 117 SINGL_FLD = SEP_FLAG = 0; 118 d_mask['\t'] = d_mask[' '] = BLANK | FLD_D; 119 ftpos = fldtab; 120 121 fixit(&argc, argv); 122 if (!(tmpdir = getenv("TMPDIR"))) 123 tmpdir = _PATH_TMP; 124 125 while ((ch = getopt(argc, argv, "bcdfik:mHno:rR:sSt:T:ux")) != -1) { 126 switch (ch) { 127 case 'b': 128 fldtab->flags |= BI | BT; 129 break; 130 case 'c': 131 cflag = 1; 132 break; 133 case 'd': case 'f': case 'i': case 'n': case 'r': 134 tmp |= optval(ch, 0); 135 if ((tmp & R) && (tmp & F)) 136 fldtab->weights = RFtable; 137 else if (tmp & F) 138 fldtab->weights = Ftable; 139 else if (tmp & R) 140 fldtab->weights = Rascii; 141 fldtab->flags |= tmp; 142 break; 143 case 'H': 144 PANIC = 0; 145 break; 146 case 'k': 147 setfield(optarg, ++ftpos, fldtab->flags); 148 break; 149 case 'm': 150 mflag = 1; 151 break; 152 case 'o': 153 outpath = optarg; 154 break; 155 case 's': 156 /* for GNU sort compatibility (this is our default) */ 157 stable_sort = 1; 158 break; 159 case 'S': 160 stable_sort = 0; 161 break; 162 case 't': 163 if (SEP_FLAG) 164 usage("multiple field delimiters"); 165 SEP_FLAG = 1; 166 d_mask[' '] &= ~FLD_D; 167 d_mask['\t'] &= ~FLD_D; 168 d_mask[(u_char)*optarg] |= FLD_D; 169 if (d_mask[(u_char)*optarg] & REC_D_F) 170 errx(2, "record/field delimiter clash"); 171 break; 172 case 'R': 173 if (REC_D != '\n') 174 usage("multiple record delimiters"); 175 if ('\n' == (REC_D = *optarg)) 176 break; 177 d_mask['\n'] = d_mask[' ']; 178 d_mask[REC_D] = REC_D_F; 179 break; 180 case 'T': 181 /* -T tmpdir */ 182 tmpdir = optarg; 183 break; 184 case 'u': 185 UNIQUE = 1; 186 break; 187 case '?': 188 default: 189 usage(NULL); 190 } 191 } 192 if (cflag && argc > optind+1) 193 errx(2, "too many input files for -c option"); 194 if (argc - 2 > optind && !strcmp(argv[argc-2], "-o")) { 195 outpath = argv[argc-1]; 196 argc -= 2; 197 } 198 if (mflag && argc - optind > (MAXFCT - (16+1))*16) 199 errx(2, "too many input files for -m option"); 200 for (i = optind; i < argc; i++) { 201 /* allow one occurrence of /dev/stdin */ 202 if (!strcmp(argv[i], "-") || !strcmp(argv[i], _PATH_STDIN)) { 203 if (stdinflag) 204 warnx("ignoring extra \"%s\" in file list", 205 argv[i]); 206 else 207 stdinflag = 1; 208 209 /* change to /dev/stdin if '-' */ 210 if (argv[i][0] == '-') 211 argv[i] = _PATH_STDIN; 212 213 } else if ((ch = access(argv[i], R_OK))) 214 err(2, "%s", argv[i]); 215 } 216 if (!(fldtab->flags & (I|D|N) || fldtab[1].icol.num)) { 217 SINGL_FLD = 1; 218 fldtab[0].icol.num = 1; 219 } else { 220 if (!fldtab[1].icol.num) { 221 fldtab[0].flags &= ~(BI|BT); 222 setfield("1", ++ftpos, fldtab->flags); 223 } 224 fldreset(fldtab); 225 fldtab[0].flags &= ~F; 226 } 227 settables(fldtab[0].flags); 228 num_init(); 229 fldtab->weights = gweights; 230 if (optind == argc) { 231 static const char * const names[] = { _PATH_STDIN, NULL }; 232 233 filelist.names = names; 234 optind--; 235 } else 236 filelist.names = (const char * const *) &argv[optind]; 237 238 if (SINGL_FLD) 239 get = makeline; 240 else 241 get = makekey; 242 243 if (cflag) { 244 order(&filelist, get, fldtab); 245 /* NOT REACHED */ 246 } 247 if (!outpath) { 248 (void)snprintf(toutpath, 249 sizeof(toutpath), "%sstdout", _PATH_DEV); 250 outfile = outpath = toutpath; 251 outfp = stdout; 252 } else if (!(ch = access(outpath, 0)) && 253 strncmp(_PATH_DEV, outpath, 5)) { 254 struct sigaction act; 255 static const int sigtable[] = {SIGHUP, SIGINT, SIGPIPE, 256 SIGXCPU, SIGXFSZ, SIGVTALRM, SIGPROF, 0}; 257 int outfd; 258 errno = 0; 259 if (access(outpath, W_OK)) 260 err(2, "%s", outpath); 261 (void)snprintf(toutpath, sizeof(toutpath), "%sXXXXXX", 262 outpath); 263 if ((outfd = mkstemp(toutpath)) == -1) 264 err(2, "Cannot create temporary file `%s'", toutpath); 265 if ((outfp = fdopen(outfd, "w")) == NULL) 266 err(2, "Cannot open temporary file `%s'", toutpath); 267 outfile = toutpath; 268 (void)atexit(cleanup); 269 act.sa_handler = onsignal; 270 (void) sigemptyset(&act.sa_mask); 271 act.sa_flags = SA_RESTART | SA_RESETHAND; 272 for (i = 0; sigtable[i]; ++i) /* always unlink toutpath */ 273 sigaction(sigtable[i], &act, 0); 274 } else 275 outfile = outpath; 276 277 if (outfp == NULL && (outfp = fopen(outfile, "w")) == NULL) 278 err(2, "output file %s", outfile); 279 280 if (mflag) { 281 fmerge(-1, 0, &filelist, argc-optind, get, outfp, putline, 282 fldtab); 283 } else 284 fsort(-1, 0, 0, &filelist, argc-optind, outfp, fldtab); 285 286 if (outfile != outpath) { 287 if (access(outfile, 0)) 288 err(2, "%s", outfile); 289 (void)unlink(outpath); 290 if (link(outfile, outpath)) 291 err(2, "cannot link %s: output left in %s", 292 outpath, outfile); 293 (void)unlink(outfile); 294 } 295 exit(0); 296 } 297 298 static void 299 onsignal(sig) 300 int sig; 301 { 302 cleanup(); 303 } 304 305 static void 306 cleanup() 307 { 308 if (toutpath[0]) 309 (void)unlink(toutpath); 310 } 311 312 static void 313 usage(msg) 314 const char *msg; 315 { 316 if (msg != NULL) 317 (void)fprintf(stderr, "sort: %s\n", msg); 318 (void)fprintf(stderr, "usage: [-o output] [-cmubdfinrsS] [-t char] "); 319 (void)fprintf(stderr, "[-R char] [-k keydef] ... [files]\n"); 320 exit(2); 321 } 322
Indexes created Fri Sep 26 08:10:20 GMT 2025