sort.c revision 1.10
1 /* $NetBSD: sort.c,v 1.10 2001/01/08 18:35:49 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 1.10 2001/01/08 18:35:49 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 64 int REC_D = '\n'; 65 u_char d_mask[NBINS]; /* flags for rec_d, field_d, <blank> */ 66 /* 67 * weight tables. Gweights is one of ascii, Rascii.. 68 * modified to weight rec_d = 0 (or 255) 69 */ 70 extern u_char gweights[NBINS]; 71 u_char ascii[NBINS], Rascii[NBINS], RFtable[NBINS], Ftable[NBINS]; 72 /* 73 * masks of ignored characters. Alltable is 256 ones 74 */ 75 u_char dtable[NBINS], itable[NBINS], alltable[NBINS]; 76 int SINGL_FLD = 0, SEP_FLAG = 0, UNIQUE = 0; 77 struct coldesc clist[(ND+1)*2]; 78 int ncols = 0; 79 extern struct coldesc clist[(ND+1)*2]; 80 extern int ncols; 81 82 /* 83 * Default to stable sort. 84 */ 85 int stable_sort = 1; 86 87 char toutpath[_POSIX_PATH_MAX]; 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 95 int main __P((int argc, char **argv)); 96 97 int 98 main(argc, argv) 99 int argc; 100 char *argv[]; 101 { 102 int (*get) __P((int, union f_handle, int, struct recheader *, u_char *, 103 struct field *)); 104 int ch, i, stdinflag = 0, tmp = 0; 105 char cflag = 0, mflag = 0; 106 char *outfile, *outpath = 0; 107 struct field fldtab[ND+2], *ftpos; 108 union f_handle filelist; 109 FILE *outfp = NULL; 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 if (!(tmpdir = getenv("TMPDIR"))) 120 tmpdir = _PATH_TMP; 121 122 while ((ch = getopt(argc, argv, "bcdfik:mHno:rsSt:T:ux")) != -1) { 123 switch (ch) { 124 case 'b': fldtab->flags |= BI | BT; 125 break; 126 case 'd': 127 case 'i': 128 case 'f': 129 case 'n': 130 case 'r': tmp |= optval(ch, 0); 131 if (tmp & R && tmp & F) 132 fldtab->weights = RFtable; 133 else if (tmp & F) 134 fldtab->weights = Ftable; 135 else if(tmp & R) 136 fldtab->weights = Rascii; 137 fldtab->flags |= tmp; 138 break; 139 case 'o': 140 outpath = optarg; 141 break; 142 case 'k': 143 setfield(optarg, ++ftpos, fldtab->flags); 144 break; 145 case 's': 146 /* for GNU sort compatibility (this is our default) */ 147 stable_sort = 1; 148 break; 149 case 'S': 150 stable_sort = 0; 151 break; 152 case 't': 153 if (SEP_FLAG) 154 usage("multiple field delimiters"); 155 SEP_FLAG = 1; 156 d_mask[' '] &= ~FLD_D; 157 d_mask['\t'] &= ~FLD_D; 158 d_mask[(u_char)*optarg] |= FLD_D; 159 if (d_mask[(u_char)*optarg] & REC_D_F) 160 err(2, "record/field delimiter clash"); 161 break; 162 case 'T': 163 if (REC_D != '\n') 164 usage("multiple record delimiters"); 165 if ('\n' == (REC_D = *optarg)) 166 break; 167 d_mask['\n'] = d_mask[' ']; 168 d_mask[REC_D] = REC_D_F; 169 break; 170 case 'u': 171 UNIQUE = 1; 172 break; 173 case 'c': 174 cflag = 1; 175 break; 176 case 'm': 177 mflag = 1; 178 break; 179 case 'H': 180 PANIC = 0; 181 break; 182 case '?': 183 default: 184 usage(""); 185 } 186 } 187 if (cflag && argc > optind+1) 188 errx(2, "too many input files for -c option"); 189 if (argc - 2 > optind && !strcmp(argv[argc-2], "-o")) { 190 outpath = argv[argc-1]; 191 argc -= 2; 192 } 193 if (mflag && argc - optind > (MAXFCT - (16+1))*16) 194 errx(2, "too many input files for -m option"); 195 for (i = optind; i < argc; i++) { 196 /* allow one occurrence of /dev/stdin */ 197 if (!strcmp(argv[i], "-") || !strcmp(argv[i], _PATH_STDIN)) { 198 if (stdinflag) 199 warnx("ignoring extra \"%s\" in file list", 200 argv[i]); 201 else 202 stdinflag = 1; 203 } else if ((ch = access(argv[i], R_OK))) 204 err(2, "%s", argv[i]); 205 } 206 if (!(fldtab->flags & (I|D|N) || fldtab[1].icol.num)) { 207 SINGL_FLD = 1; 208 fldtab[0].icol.num = 1; 209 } else { 210 if (!fldtab[1].icol.num) { 211 fldtab[0].flags &= ~(BI|BT); 212 setfield("1", ++ftpos, fldtab->flags); 213 } 214 fldreset(fldtab); 215 fldtab[0].flags &= ~F; 216 } 217 settables(fldtab[0].flags); 218 num_init(); 219 fldtab->weights = gweights; 220 if (optind == argc) { 221 static const char * const names[] = { _PATH_STDIN, NULL }; 222 223 filelist.names = names; 224 optind--; 225 } else 226 filelist.names = (const char * const *) &argv[optind]; 227 if (SINGL_FLD) 228 get = makeline; 229 else 230 get = makekey; 231 if (cflag) { 232 order(filelist, get, fldtab); 233 /* NOT REACHED */ 234 } 235 if (!outpath) { 236 (void)snprintf(toutpath, 237 sizeof(toutpath), "%sstdout", _PATH_DEV); 238 outfile = outpath = toutpath; 239 } else if (!(ch = access(outpath, 0)) && 240 strncmp(_PATH_DEV, outpath, 5)) { 241 struct sigaction act = {0}; 242 int sigtable[] = {SIGHUP, SIGINT, SIGPIPE, SIGXCPU, SIGXFSZ, 243 SIGVTALRM, SIGPROF, 0}; 244 int outfd; 245 errno = 0; 246 if (access(outpath, W_OK)) 247 err(2, "%s", outpath); 248 act.sa_handler = onsignal; 249 act.sa_flags = SA_RESTART | SA_RESETHAND; 250 (void)snprintf(toutpath, sizeof(toutpath), "%sXXXX", outpath); 251 if ((outfd = mkstemp(toutpath)) < 0 || 252 (outfp = fdopen(outfd, "w")) == 0) 253 err(2, "temporary file %s", toutpath); 254 outfile = toutpath; 255 (void)atexit(cleanup); 256 for (i = 0; sigtable[i]; ++i) /* always unlink toutpath */ 257 sigaction(sigtable[i], &act, 0); 258 } else 259 outfile = outpath; 260 if (outfp == NULL && (outfp = fopen(outfile, "w")) == NULL) 261 err(2, "output file %s", outfile); 262 if (mflag) 263 fmerge(-1, filelist, argc-optind, get, outfp, putline, fldtab); 264 else 265 fsort(-1, 0, filelist, argc-optind, outfp, fldtab); 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 Thu Sep 25 06:09:39 GMT 2025