sort.c revision 1.5
1 /* $NetBSD: sort.c,v 1.5 2000/10/07 21:13:56 bjh21 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 1.5 2000/10/07 21:13:56 bjh21 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 char devstdin[] = _PATH_STDIN; 83 char toutpath[_POSIX_PATH_MAX]; 84 char *tmpdir = _PATH_VARTMP; 85 86 static void cleanup __P((void)); 87 static void onsignal __P((int)); 88 static void usage __P((char *)); 89 90 int main __P((int argc, char **argv)); 91 92 int 93 main(argc, argv) 94 int argc; 95 char *argv[]; 96 { 97 extern int optind; 98 extern char *optarg; 99 int (*get) __P((int, union f_handle, int, struct recheader *, u_char *, 100 struct field *)); 101 int ch, i, stdinflag = 0, tmp = 0; 102 char cflag = 0, mflag = 0, nflag = 0; 103 char *outfile, *outpath = 0; 104 struct field fldtab[ND+2], *ftpos; 105 union f_handle filelist; 106 FILE *outfp = NULL; 107 memset(fldtab, 0, (ND+2)*sizeof(struct field)); 108 memset(d_mask, 0, NBINS); 109 d_mask[REC_D = '\n'] = REC_D_F; 110 SINGL_FLD = SEP_FLAG = 0; 111 d_mask['\t'] = d_mask[' '] = BLANK | FLD_D; 112 ftpos = fldtab; 113 fixit(&argc, argv); 114 while ((ch = getopt(argc, argv, "bcdfik:mHno:rt:T:ux")) != EOF) { 115 if ((outfile = getenv("TMPDIR"))) 116 tmpdir = outfile; 117 switch (ch) { 118 case 'b': fldtab->flags |= BI | BT; 119 break; 120 case 'd': 121 case 'i': 122 case 'f': 123 case 'r': tmp |= optval(ch, 0); 124 if (tmp & R && tmp & F) 125 fldtab->weights = RFtable; 126 else if (tmp & F) 127 fldtab->weights = Ftable; 128 else if(tmp & R) 129 fldtab->weights = Rascii; 130 fldtab->flags |= tmp; 131 break; 132 case 'o': 133 outpath = optarg; 134 break; 135 case 'n': 136 nflag = 1; 137 setfield("1n", ++ftpos, fldtab->flags&(~R)); 138 break; 139 case 'k': 140 setfield(optarg, ++ftpos, fldtab->flags); 141 break; 142 case 't': 143 if (SEP_FLAG) 144 usage("multiple field delimiters"); 145 SEP_FLAG = 1; 146 d_mask[' '] &= ~FLD_D; 147 d_mask['\t'] &= ~FLD_D; 148 d_mask[(u_char)*optarg] |= FLD_D; 149 if (d_mask[(u_char)*optarg] & REC_D_F) 150 err(2, "record/field delimiter clash"); 151 break; 152 case 'T': 153 if (REC_D != '\n') 154 usage("multiple record delimiters"); 155 if ('\n' == (REC_D = *optarg)) 156 break; 157 d_mask['\n'] = d_mask[' ']; 158 d_mask[REC_D] = REC_D_F; 159 break; 160 case 'u': 161 UNIQUE = 1; 162 break; 163 case 'c': 164 cflag = 1; 165 break; 166 case 'm': 167 mflag = 1; 168 break; 169 case 'H': 170 PANIC = 0; 171 break; 172 case '?': 173 default: usage(""); 174 } 175 } 176 if (cflag && argc > optind+1) 177 errx(2, "too many input files for -c option"); 178 if (argc - 2 > optind && !strcmp(argv[argc-2], "-o")) { 179 outpath = argv[argc-1]; 180 argc -= 2; 181 } 182 if (mflag && argc - optind > (MAXFCT - (16+1))*16) 183 errx(2, "too many input files for -m option"); 184 for (i = optind; i < argc; i++) { 185 /* allow one occurrence of /dev/stdin */ 186 if (!strcmp(argv[i], "-") || !strcmp(argv[i], devstdin)) { 187 if (stdinflag) 188 warnx("ignoring extra \"%s\" in file list", 189 argv[i]); 190 else { 191 stdinflag = 1; 192 argv[i] = devstdin; 193 } 194 } else if ((ch = access(argv[i], R_OK))) 195 err(2, argv[i]); 196 } 197 if (!(fldtab->flags & (I|D) || fldtab[1].icol.num)) { 198 SINGL_FLD = 1; 199 fldtab[0].icol.num = 1; 200 } else { 201 if (!fldtab[1].icol.num) { 202 fldtab[0].flags &= ~(BI|BT); 203 setfield("1", ++ftpos, fldtab->flags); 204 } 205 if (nflag) 206 fldtab[1].flags |= fldtab->flags; 207 fldreset(fldtab); 208 fldtab[0].flags &= ~F; 209 } 210 settables(fldtab[0].flags); 211 num_init(); 212 fldtab->weights = gweights; 213 if (optind == argc) { 214 static char *names[2]; 215 216 names[0] = devstdin; 217 names[1] = NULL; 218 filelist.names = names; 219 optind--; 220 } else 221 filelist.names = argv+optind; 222 if (SINGL_FLD) 223 get = makeline; 224 else 225 get = makekey; 226 if (cflag) { 227 order(filelist, get, fldtab); 228 /* NOT REACHED */ 229 } 230 if (!outpath) { 231 (void)snprintf(toutpath, 232 sizeof(toutpath), "%sstdout", _PATH_DEV); 233 outfile = outpath = toutpath; 234 } else if (!(ch = access(outpath, 0)) && 235 strncmp(_PATH_DEV, outpath, 5)) { 236 struct sigaction act = {0}; 237 int sigtable[] = {SIGHUP, SIGINT, SIGPIPE, SIGXCPU, SIGXFSZ, 238 SIGVTALRM, SIGPROF, 0}; 239 int outfd; 240 errno = 0; 241 if (access(outpath, W_OK)) 242 err(2, "%s", outpath); 243 act.sa_handler = onsignal; 244 act.sa_flags = SA_RESTART | SA_RESETHAND; 245 (void)snprintf(toutpath, sizeof(toutpath), "%sXXXX", outpath); 246 if ((outfd = mkstemp(toutpath)) < 0 || 247 (outfp = fdopen(outfd, "w")) == 0) 248 err(2, toutpath); 249 outfile = toutpath; 250 (void)atexit(cleanup); 251 for (i = 0; sigtable[i]; ++i) /* always unlink toutpath */ 252 sigaction(sigtable[i], &act, 0); 253 } else 254 outfile = outpath; 255 if (outfp == NULL && (outfp = fopen(outfile, "w")) == NULL) 256 err(2, outfile); 257 if (mflag) 258 fmerge(-1, filelist, argc-optind, get, outfp, putline, fldtab); 259 else 260 fsort(-1, 0, filelist, argc-optind, outfp, fldtab); 261 if (outfile != outpath) { 262 if (access(outfile, 0)) 263 err(2, outfile); 264 (void)unlink(outpath); 265 if (link(outfile, outpath)) 266 err(2, "cannot link %s: output left in %s", 267 outpath, outfile); 268 (void)unlink(outfile); 269 } 270 exit(0); 271 } 272 273 static void 274 onsignal(sig) 275 int sig; 276 { 277 cleanup(); 278 } 279 280 static void 281 cleanup() 282 { 283 if (toutpath[0]) 284 (void)unlink(toutpath); 285 } 286 287 static void 288 usage(msg) 289 char *msg; 290 { 291 if (msg) 292 (void)fprintf(stderr, "sort: %s\n", msg); 293 (void)fprintf(stderr, "usage: [-o output] [-cmubdfinr] [-t char] "); 294 (void)fprintf(stderr, "[-T char] [-k keydef] ... [files]\n"); 295 exit(2); 296 } 297
Indexes created Tue Sep 23 07:09:52 GMT 2025