sort.c revision 1.11
1 /* $NetBSD: sort.c,v 1.11 2001/01/08 18:58:56 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 1.11 2001/01/08 18:58:56 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 extern u_char gweights[NBINS]; 72 u_char ascii[NBINS], Rascii[NBINS], RFtable[NBINS], Ftable[NBINS]; 73 /* 74 * masks of ignored characters. Alltable is 256 ones 75 */ 76 u_char dtable[NBINS], itable[NBINS], alltable[NBINS]; 77 int SINGL_FLD = 0, SEP_FLAG = 0, UNIQUE = 0; 78 struct coldesc clist[(ND+1)*2]; 79 int ncols = 0; 80 extern struct coldesc clist[(ND+1)*2]; 81 extern int ncols; 82 83 /* 84 * Default to stable sort. 85 */ 86 int stable_sort = 1; 87 88 char toutpath[_POSIX_PATH_MAX]; 89 90 const char *tmpdir; /* where temporary files should be put */ 91 92 static void cleanup __P((void)); 93 static void onsignal __P((int)); 94 static void usage __P((const char *)); 95 96 int main __P((int argc, char **argv)); 97 98 int 99 main(argc, argv) 100 int argc; 101 char *argv[]; 102 { 103 int (*get) __P((int, union f_handle, int, struct recheader *, u_char *, 104 struct field *)); 105 int ch, i, stdinflag = 0, tmp = 0; 106 char cflag = 0, mflag = 0; 107 char *outfile, *outpath = 0; 108 struct field fldtab[ND+2], *ftpos; 109 union f_handle filelist; 110 FILE *outfp = NULL; 111 112 setlocale(LC_ALL, ""); 113 114 memset(fldtab, 0, (ND+2)*sizeof(struct field)); 115 memset(d_mask, 0, NBINS); 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:rsSt: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 err(2, "record/field delimiter clash"); 171 break; 172 case 'T': 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 'u': 181 UNIQUE = 1; 182 break; 183 case '?': 184 default: 185 usage(""); 186 } 187 } 188 if (cflag && argc > optind+1) 189 errx(2, "too many input files for -c option"); 190 if (argc - 2 > optind && !strcmp(argv[argc-2], "-o")) { 191 outpath = argv[argc-1]; 192 argc -= 2; 193 } 194 if (mflag && argc - optind > (MAXFCT - (16+1))*16) 195 errx(2, "too many input files for -m option"); 196 for (i = optind; i < argc; i++) { 197 /* allow one occurrence of /dev/stdin */ 198 if (!strcmp(argv[i], "-") || !strcmp(argv[i], _PATH_STDIN)) { 199 if (stdinflag) 200 warnx("ignoring extra \"%s\" in file list", 201 argv[i]); 202 else 203 stdinflag = 1; 204 } else if ((ch = access(argv[i], R_OK))) 205 err(2, "%s", argv[i]); 206 } 207 if (!(fldtab->flags & (I|D|N) || fldtab[1].icol.num)) { 208 SINGL_FLD = 1; 209 fldtab[0].icol.num = 1; 210 } else { 211 if (!fldtab[1].icol.num) { 212 fldtab[0].flags &= ~(BI|BT); 213 setfield("1", ++ftpos, fldtab->flags); 214 } 215 fldreset(fldtab); 216 fldtab[0].flags &= ~F; 217 } 218 settables(fldtab[0].flags); 219 num_init(); 220 fldtab->weights = gweights; 221 if (optind == argc) { 222 static const char * const names[] = { _PATH_STDIN, NULL }; 223 224 filelist.names = names; 225 optind--; 226 } else 227 filelist.names = (const char * const *) &argv[optind]; 228 if (SINGL_FLD) 229 get = makeline; 230 else 231 get = makekey; 232 if (cflag) { 233 order(filelist, get, fldtab); 234 /* NOT REACHED */ 235 } 236 if (!outpath) { 237 (void)snprintf(toutpath, 238 sizeof(toutpath), "%sstdout", _PATH_DEV); 239 outfile = outpath = toutpath; 240 } else if (!(ch = access(outpath, 0)) && 241 strncmp(_PATH_DEV, outpath, 5)) { 242 struct sigaction act = {0}; 243 int sigtable[] = {SIGHUP, SIGINT, SIGPIPE, SIGXCPU, SIGXFSZ, 244 SIGVTALRM, SIGPROF, 0}; 245 int outfd; 246 errno = 0; 247 if (access(outpath, W_OK)) 248 err(2, "%s", outpath); 249 act.sa_handler = onsignal; 250 act.sa_flags = SA_RESTART | SA_RESETHAND; 251 (void)snprintf(toutpath, sizeof(toutpath), "%sXXXX", outpath); 252 if ((outfd = mkstemp(toutpath)) < 0 || 253 (outfp = fdopen(outfd, "w")) == 0) 254 err(2, "temporary file %s", toutpath); 255 outfile = toutpath; 256 (void)atexit(cleanup); 257 for (i = 0; sigtable[i]; ++i) /* always unlink toutpath */ 258 sigaction(sigtable[i], &act, 0); 259 } else 260 outfile = outpath; 261 if (outfp == NULL && (outfp = fopen(outfile, "w")) == NULL) 262 err(2, "output file %s", outfile); 263 if (mflag) 264 fmerge(-1, filelist, argc-optind, get, outfp, putline, fldtab); 265 else 266 fsort(-1, 0, filelist, argc-optind, outfp, fldtab); 267 if (outfile != outpath) { 268 if (access(outfile, 0)) 269 err(2, "%s", outfile); 270 (void)unlink(outpath); 271 if (link(outfile, outpath)) 272 err(2, "cannot link %s: output left in %s", 273 outpath, outfile); 274 (void)unlink(outfile); 275 } 276 exit(0); 277 } 278 279 static void 280 onsignal(sig) 281 int sig; 282 { 283 cleanup(); 284 } 285 286 static void 287 cleanup() 288 { 289 if (toutpath[0]) 290 (void)unlink(toutpath); 291 } 292 293 static void 294 usage(msg) 295 const char *msg; 296 { 297 if (msg) 298 (void)fprintf(stderr, "sort: %s\n", msg); 299 (void)fprintf(stderr, "usage: [-o output] [-cmubdfinr] [-t char] "); 300 (void)fprintf(stderr, "[-T char] [-k keydef] ... [files]\n"); 301 exit(2); 302 } 303
Indexes created Fri Sep 26 08:10:20 GMT 2025