sort.c revision 1.8
1 /* $NetBSD: sort.c,v 1.8 2000/10/16 21:48:15 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 1.8 2000/10/16 21:48:15 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 char devstdin[] = _PATH_STDIN; 83 char toutpath[_POSIX_PATH_MAX]; 84 const char *tmpdir = _PATH_TMP; 85 86 static void cleanup __P((void)); 87 static void onsignal __P((int)); 88 static void usage __P((const 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; 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 'n': 124 case 'r': tmp |= optval(ch, 0); 125 if (tmp & R && tmp & F) 126 fldtab->weights = RFtable; 127 else if (tmp & F) 128 fldtab->weights = Ftable; 129 else if(tmp & R) 130 fldtab->weights = Rascii; 131 fldtab->flags |= tmp; 132 break; 133 case 'o': 134 outpath = optarg; 135 break; 136 case 'k': 137 setfield(optarg, ++ftpos, fldtab->flags); 138 break; 139 case 't': 140 if (SEP_FLAG) 141 usage("multiple field delimiters"); 142 SEP_FLAG = 1; 143 d_mask[' '] &= ~FLD_D; 144 d_mask['\t'] &= ~FLD_D; 145 d_mask[(u_char)*optarg] |= FLD_D; 146 if (d_mask[(u_char)*optarg] & REC_D_F) 147 err(2, "record/field delimiter clash"); 148 break; 149 case 'T': 150 if (REC_D != '\n') 151 usage("multiple record delimiters"); 152 if ('\n' == (REC_D = *optarg)) 153 break; 154 d_mask['\n'] = d_mask[' ']; 155 d_mask[REC_D] = REC_D_F; 156 break; 157 case 'u': 158 UNIQUE = 1; 159 break; 160 case 'c': 161 cflag = 1; 162 break; 163 case 'm': 164 mflag = 1; 165 break; 166 case 'H': 167 PANIC = 0; 168 break; 169 case '?': 170 default: usage(""); 171 } 172 } 173 if (cflag && argc > optind+1) 174 errx(2, "too many input files for -c option"); 175 if (argc - 2 > optind && !strcmp(argv[argc-2], "-o")) { 176 outpath = argv[argc-1]; 177 argc -= 2; 178 } 179 if (mflag && argc - optind > (MAXFCT - (16+1))*16) 180 errx(2, "too many input files for -m option"); 181 for (i = optind; i < argc; i++) { 182 /* allow one occurrence of /dev/stdin */ 183 if (!strcmp(argv[i], "-") || !strcmp(argv[i], devstdin)) { 184 if (stdinflag) 185 warnx("ignoring extra \"%s\" in file list", 186 argv[i]); 187 else { 188 stdinflag = 1; 189 argv[i] = devstdin; 190 } 191 } else if ((ch = access(argv[i], R_OK))) 192 err(2, "%s", argv[i]); 193 } 194 if (!(fldtab->flags & (I|D|N) || fldtab[1].icol.num)) { 195 SINGL_FLD = 1; 196 fldtab[0].icol.num = 1; 197 } else { 198 if (!fldtab[1].icol.num) { 199 fldtab[0].flags &= ~(BI|BT); 200 setfield("1", ++ftpos, fldtab->flags); 201 } 202 fldreset(fldtab); 203 fldtab[0].flags &= ~F; 204 } 205 settables(fldtab[0].flags); 206 num_init(); 207 fldtab->weights = gweights; 208 if (optind == argc) { 209 static char *names[2]; 210 211 names[0] = devstdin; 212 names[1] = NULL; 213 filelist.names = names; 214 optind--; 215 } else 216 filelist.names = argv+optind; 217 if (SINGL_FLD) 218 get = makeline; 219 else 220 get = makekey; 221 if (cflag) { 222 order(filelist, get, fldtab); 223 /* NOT REACHED */ 224 } 225 if (!outpath) { 226 (void)snprintf(toutpath, 227 sizeof(toutpath), "%sstdout", _PATH_DEV); 228 outfile = outpath = toutpath; 229 } else if (!(ch = access(outpath, 0)) && 230 strncmp(_PATH_DEV, outpath, 5)) { 231 struct sigaction act = {0}; 232 int sigtable[] = {SIGHUP, SIGINT, SIGPIPE, SIGXCPU, SIGXFSZ, 233 SIGVTALRM, SIGPROF, 0}; 234 int outfd; 235 errno = 0; 236 if (access(outpath, W_OK)) 237 err(2, "%s", outpath); 238 act.sa_handler = onsignal; 239 act.sa_flags = SA_RESTART | SA_RESETHAND; 240 (void)snprintf(toutpath, sizeof(toutpath), "%sXXXX", outpath); 241 if ((outfd = mkstemp(toutpath)) < 0 || 242 (outfp = fdopen(outfd, "w")) == 0) 243 err(2, "temporary file %s", toutpath); 244 outfile = toutpath; 245 (void)atexit(cleanup); 246 for (i = 0; sigtable[i]; ++i) /* always unlink toutpath */ 247 sigaction(sigtable[i], &act, 0); 248 } else 249 outfile = outpath; 250 if (outfp == NULL && (outfp = fopen(outfile, "w")) == NULL) 251 err(2, "output file %s", outfile); 252 if (mflag) 253 fmerge(-1, filelist, argc-optind, get, outfp, putline, fldtab); 254 else 255 fsort(-1, 0, filelist, argc-optind, outfp, fldtab); 256 if (outfile != outpath) { 257 if (access(outfile, 0)) 258 err(2, "%s", outfile); 259 (void)unlink(outpath); 260 if (link(outfile, outpath)) 261 err(2, "cannot link %s: output left in %s", 262 outpath, outfile); 263 (void)unlink(outfile); 264 } 265 exit(0); 266 } 267 268 static void 269 onsignal(sig) 270 int sig; 271 { 272 cleanup(); 273 } 274 275 static void 276 cleanup() 277 { 278 if (toutpath[0]) 279 (void)unlink(toutpath); 280 } 281 282 static void 283 usage(msg) 284 const char *msg; 285 { 286 if (msg) 287 (void)fprintf(stderr, "sort: %s\n", msg); 288 (void)fprintf(stderr, "usage: [-o output] [-cmubdfinr] [-t char] "); 289 (void)fprintf(stderr, "[-T char] [-k keydef] ... [files]\n"); 290 exit(2); 291 } 292
Indexes created Tue Sep 23 14:10:03 GMT 2025