sort.c revision 1.23
1 /* $NetBSD: sort.c,v 1.23 2001/02/19 15:53:07 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 1.23 2001/02/19 15:53:07 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 u_char ascii[NBINS], Rascii[NBINS], RFtable[NBINS], Ftable[NBINS]; 72 int SINGL_FLD = 0, SEP_FLAG = 0, UNIQUE = 0; 73 struct coldesc clist[(ND+1)*2]; 74 int ncols = 0; 75 extern struct coldesc clist[(ND+1)*2]; 76 extern int ncols; 77 78 /* 79 * Default to stable sort. 80 */ 81 int stable_sort = 1; 82 83 char toutpath[_POSIX_PATH_MAX]; 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 fldtab[ND+2], *ftpos; 103 struct filelist filelist; 104 FILE *outfp = NULL; 105 106 setlocale(LC_ALL, ""); 107 108 memset(fldtab, 0, (ND+2)*sizeof(struct field)); 109 memset(d_mask, 0, NBINS); 110 d_mask[REC_D = '\n'] = REC_D_F; 111 SINGL_FLD = SEP_FLAG = 0; 112 d_mask['\t'] = d_mask[' '] = BLANK | FLD_D; 113 ftpos = fldtab; 114 115 fixit(&argc, argv); 116 if (!(tmpdir = getenv("TMPDIR"))) 117 tmpdir = _PATH_TMP; 118 119 while ((ch = getopt(argc, argv, "bcdfik:mHno:rR:sSt:T:ux")) != -1) { 120 switch (ch) { 121 case 'b': 122 fldtab->flags |= BI | BT; 123 break; 124 case 'c': 125 cflag = 1; 126 break; 127 case 'd': case 'f': case 'i': case 'n': case 'r': 128 tmp |= optval(ch, 0); 129 if ((tmp & R) && (tmp & F)) 130 fldtab->weights = RFtable; 131 else if (tmp & F) 132 fldtab->weights = Ftable; 133 else if (tmp & R) 134 fldtab->weights = Rascii; 135 fldtab->flags |= tmp; 136 break; 137 case 'H': 138 PANIC = 0; 139 break; 140 case 'k': 141 setfield(optarg, ++ftpos, fldtab->flags); 142 break; 143 case 'm': 144 mflag = 1; 145 break; 146 case 'o': 147 outpath = optarg; 148 break; 149 case 's': 150 /* for GNU sort compatibility (this is our default) */ 151 stable_sort = 1; 152 break; 153 case 'S': 154 stable_sort = 0; 155 break; 156 case 't': 157 if (SEP_FLAG) 158 usage("multiple field delimiters"); 159 SEP_FLAG = 1; 160 d_mask[' '] &= ~FLD_D; 161 d_mask['\t'] &= ~FLD_D; 162 d_mask[(u_char)*optarg] |= FLD_D; 163 if (d_mask[(u_char)*optarg] & REC_D_F) 164 errx(2, "record/field delimiter clash"); 165 break; 166 case 'R': 167 if (REC_D != '\n') 168 usage("multiple record delimiters"); 169 if ('\n' == (REC_D = *optarg)) 170 break; 171 d_mask['\n'] = d_mask[' ']; 172 d_mask[REC_D] = REC_D_F; 173 break; 174 case 'T': 175 /* -T tmpdir */ 176 tmpdir = optarg; 177 break; 178 case 'u': 179 UNIQUE = 1; 180 break; 181 case '?': 182 default: 183 usage(NULL); 184 } 185 } 186 if (cflag && argc > optind+1) 187 errx(2, "too many input files for -c option"); 188 if (argc - 2 > optind && !strcmp(argv[argc-2], "-o")) { 189 outpath = argv[argc-1]; 190 argc -= 2; 191 } 192 if (mflag && argc - optind > (MAXFCT - (16+1))*16) 193 errx(2, "too many input files for -m option"); 194 for (i = optind; i < argc; i++) { 195 /* allow one occurrence of /dev/stdin */ 196 if (!strcmp(argv[i], "-") || !strcmp(argv[i], _PATH_STDIN)) { 197 if (stdinflag) 198 warnx("ignoring extra \"%s\" in file list", 199 argv[i]); 200 else 201 stdinflag = 1; 202 203 /* change to /dev/stdin if '-' */ 204 if (argv[i][0] == '-') 205 argv[i] = _PATH_STDIN; 206 207 } else if ((ch = access(argv[i], R_OK))) 208 err(2, "%s", argv[i]); 209 } 210 if (!(fldtab->flags & (I|D|N) || fldtab[1].icol.num)) { 211 SINGL_FLD = 1; 212 fldtab[0].icol.num = 1; 213 } else { 214 if (!fldtab[1].icol.num) { 215 fldtab[0].flags &= ~(BI|BT); 216 setfield("1", ++ftpos, fldtab->flags); 217 } 218 fldreset(fldtab); 219 fldtab[0].flags &= ~F; 220 } 221 settables(fldtab[0].flags); 222 num_init(); 223 fldtab->weights = gweights; 224 if (optind == argc) { 225 static const char * const names[] = { _PATH_STDIN, NULL }; 226 227 filelist.names = names; 228 optind--; 229 } else 230 filelist.names = (const char * const *) &argv[optind]; 231 232 if (SINGL_FLD) 233 get = makeline; 234 else 235 get = makekey; 236 237 if (cflag) { 238 order(&filelist, get, fldtab); 239 /* NOT REACHED */ 240 } 241 if (!outpath) { 242 (void)snprintf(toutpath, 243 sizeof(toutpath), "%sstdout", _PATH_DEV); 244 outfile = outpath = toutpath; 245 } else if (!(ch = access(outpath, 0)) && 246 strncmp(_PATH_DEV, outpath, 5)) { 247 static const struct sigaction act = 248 { onsignal, {{0}}, SA_RESTART | SA_RESETHAND }; 249 static const int sigtable[] = {SIGHUP, SIGINT, SIGPIPE, SIGXCPU, 250 SIGXFSZ, SIGVTALRM, SIGPROF, 0}; 251 int outfd; 252 errno = 0; 253 if (access(outpath, W_OK)) 254 err(2, "%s", outpath); 255 (void)snprintf(toutpath, sizeof(toutpath), "%sXXXX", outpath); 256 if ((outfd = mkstemp(toutpath)) < 0 || 257 (outfp = fdopen(outfd, "w")) == 0) 258 err(2, "temporary file %s", toutpath); 259 outfile = toutpath; 260 (void)atexit(cleanup); 261 for (i = 0; sigtable[i]; ++i) /* always unlink toutpath */ 262 sigaction(sigtable[i], &act, 0); 263 } else 264 outfile = outpath; 265 266 if (outfp == NULL && (outfp = fopen(outfile, "w")) == NULL) 267 err(2, "output file %s", outfile); 268 269 if (mflag) { 270 fmerge(-1, 0, &filelist, argc-optind, get, outfp, putline, 271 fldtab); 272 } else 273 fsort(-1, 0, 0, &filelist, argc-optind, outfp, fldtab); 274 275 if (outfile != outpath) { 276 if (access(outfile, 0)) 277 err(2, "%s", outfile); 278 (void)unlink(outpath); 279 if (link(outfile, outpath)) 280 err(2, "cannot link %s: output left in %s", 281 outpath, outfile); 282 (void)unlink(outfile); 283 } 284 exit(0); 285 } 286 287 static void 288 onsignal(sig) 289 int sig; 290 { 291 cleanup(); 292 } 293 294 static void 295 cleanup() 296 { 297 if (toutpath[0]) 298 (void)unlink(toutpath); 299 } 300 301 static void 302 usage(msg) 303 const char *msg; 304 { 305 if (msg != NULL) 306 (void)fprintf(stderr, "sort: %s\n", msg); 307 (void)fprintf(stderr, "usage: [-o output] [-cmubdfinrsS] [-t char] "); 308 (void)fprintf(stderr, "[-R char] [-k keydef] ... [files]\n"); 309 exit(2); 310 } 311
Indexes created Tue Sep 23 14:10:03 GMT 2025