Annotation of src/usr.bin/sort/sort.c, Revision 1.8
1.8 ! deraadt 1: /* $OpenBSD: sort.c,v 1.7 1998/07/20 19:14:37 mickey Exp $ */
1.1 millert 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: #ifndef lint
40: static char copyright[] =
41: "@(#) Copyright (c) 1993\n\
42: The Regents of the University of California. All rights reserved.\n";
43: #endif /* not lint */
44:
45: #ifndef lint
46: #if 0
47: static char sccsid[] = "@(#)sort.c 8.1 (Berkeley) 6/6/93";
48: #else
1.8 ! deraadt 49: static char rcsid[] = "$OpenBSD: sort.c,v 1.7 1998/07/20 19:14:37 mickey Exp $";
1.1 millert 50: #endif
51: #endif /* not lint */
52:
1.4 millert 53: /*
54: * Sort sorts a file using an optional user-defined key.
1.1 millert 55: * Sort uses radix sort for internal sorting, and allows
56: * a choice of merge sort and radix sort for external sorting.
57: */
58:
59: #include "sort.h"
60: #include "fsort.h"
61: #include "pathnames.h"
62:
63: #include <paths.h>
64: #include <signal.h>
65: #include <stdlib.h>
66: #include <string.h>
67: #include <unistd.h>
68:
69: int REC_D = '\n';
70: u_char d_mask[NBINS]; /* flags for rec_d, field_d, <blank> */
1.4 millert 71:
1.1 millert 72: /*
73: * weight tables. Gweights is one of ascii, Rascii..
74: * modified to weight rec_d = 0 (or 255)
75: */
76: extern u_char gweights[NBINS];
77: u_char ascii[NBINS], Rascii[NBINS], RFtable[NBINS], Ftable[NBINS];
1.4 millert 78:
1.1 millert 79: /*
80: * masks of ignored characters. Alltable is 256 ones
81: */
82: u_char dtable[NBINS], itable[NBINS], alltable[NBINS];
83: int SINGL_FLD = 0, SEP_FLAG = 0, UNIQUE = 0;
84: struct coldesc clist[(ND+1)*2];
85: int ncols = 0;
86: extern struct coldesc clist[(ND+1)*2];
87: extern int ncols;
88:
89: char devstdin[] = _PATH_STDIN;
90: char toutpath[_POSIX_PATH_MAX];
91: char *tmpdir = _PATH_VARTMP;
92:
93: static void cleanup __P((void));
94: static void onsig __P((int));
95: static void usage __P((char *));
96:
97: int
98: main(argc, argv)
99: int argc;
100: char *argv[];
101: {
102: int (*get)();
103: int ch, i, stdinflag = 0, tmp = 0;
1.6 millert 104: char cflag = 0, mflag = 0;
1.1 millert 105: char *outfile, *outpath = 0;
106: struct field fldtab[ND+2], *ftpos;
107: union f_handle filelist;
108: FILE *outfp = NULL;
1.4 millert 109:
1.1 millert 110: memset(fldtab, 0, (ND+2)*sizeof(struct field));
111: memset(d_mask, 0, NBINS);
112: d_mask[REC_D = '\n'] = REC_D_F;
113: SINGL_FLD = SEP_FLAG = 0;
114: d_mask['\t'] = d_mask[' '] = BLANK | FLD_D;
115: ftpos = fldtab;
116: fixit(&argc, argv);
117: if (!issetugid() && (outfile = getenv("TMPDIR")))
118: tmpdir = outfile;
119: while ((ch = getopt(argc, argv, "bcdfik:mHno:rR:t:T:uy:")) != -1) {
1.4 millert 120: switch (ch) {
1.1 millert 121: case 'b': fldtab->flags |= BI | BT;
122: break;
123: case 'd':
1.4 millert 124: case 'f':
1.1 millert 125: case 'i':
1.5 millert 126: case 'n':
1.1 millert 127: case 'r': tmp |= optval(ch, 0);
128: if (tmp & R && tmp & F)
129: fldtab->weights = RFtable;
130: else if (tmp & F)
131: fldtab->weights = Ftable;
1.4 millert 132: else if (tmp & R)
1.1 millert 133: fldtab->weights = Rascii;
134: fldtab->flags |= tmp;
135: break;
136: case 'o':
137: outpath = optarg;
138: break;
139: case 'k':
1.5 millert 140: setfield(optarg, ++ftpos, fldtab->flags);
1.1 millert 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[(int)*optarg] |= FLD_D;
149: if (d_mask[(int)*optarg] & REC_D_F)
150: err(2, "record/field delimiter clash");
151: break;
152: case 'R':
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 'T':
161: tmpdir = optarg;
162: break;
163: case 'u':
164: UNIQUE = 1;
165: break;
166: case 'c':
167: cflag = 1;
168: break;
169: case 'm':
170: mflag = 1;
171: break;
172: case 'H':
173: PANIC = 0;
174: break;
175: case 'y':
176: /* accept -y for backwards compat. */
177: break;
178: case '?':
1.8 ! deraadt 179: default:
! 180: usage(NULL);
1.1 millert 181: }
182: }
1.4 millert 183:
1.1 millert 184: if (cflag && argc > optind+1)
185: errx(2, "too many input files for -c option");
1.4 millert 186:
1.1 millert 187: if (argc - 2 > optind && !strcmp(argv[argc-2], "-o")) {
188: outpath = argv[argc-1];
189: argc -= 2;
190: }
1.4 millert 191:
1.1 millert 192: if (mflag && argc - optind > (MAXFCT - (16+1))*16)
193: errx(2, "too many input files for -m option");
1.4 millert 194:
1.1 millert 195: for (i = optind; i < argc; i++) {
196: /* allow one occurrence of /dev/stdin */
197: if (!strcmp(argv[i], "-") || !strcmp(argv[i], devstdin)) {
198: if (stdinflag)
199: warnx("ignoring extra \"%s\" in file list",
200: argv[i]);
201: else {
202: stdinflag = 1;
203: argv[i] = devstdin;
204: }
205: } else if ((ch = access(argv[i], R_OK)))
1.2 millert 206: err(2, argv[i]);
1.1 millert 207: }
1.4 millert 208:
1.5 millert 209: if (!(fldtab->flags & (I|D|N) || fldtab[1].icol.num)) {
1.1 millert 210: SINGL_FLD = 1;
211: fldtab[0].icol.num = 1;
212: } else {
213: if (!fldtab[1].icol.num) {
214: fldtab[0].flags &= ~(BI|BT);
215: setfield("1", ++ftpos, fldtab->flags);
216: }
217: fldreset(fldtab);
218: fldtab[0].flags &= ~F;
219: }
220: settables(fldtab[0].flags);
221: num_init();
222: fldtab->weights = gweights;
1.4 millert 223:
1.3 deraadt 224: if (optind == argc) {
225: static char *names[2];
226:
227: names[0] = devstdin;
228: names[1] = NULL;
229: filelist.names = names;
230: optind--;
231: } else
232: filelist.names = argv+optind;
1.4 millert 233:
1.1 millert 234: if (SINGL_FLD)
235: get = makeline;
236: else
237: get = makekey;
1.4 millert 238:
1.1 millert 239: if (cflag) {
240: order(filelist, get, fldtab);
241: /* NOT REACHED */
242: }
1.4 millert 243:
1.1 millert 244: if (!outpath) {
245: (void)snprintf(toutpath,
246: sizeof(toutpath), "%sstdout", _PATH_DEV);
247: outfile = outpath = toutpath;
248: } else if (!(ch = access(outpath, 0)) &&
249: strncmp(_PATH_DEV, outpath, 5)) {
250: struct sigaction act = {0, SIG_BLOCK, 6};
251: int sigtable[] = {SIGHUP, SIGINT, SIGPIPE, SIGXCPU, SIGXFSZ,
252: SIGVTALRM, SIGPROF, 0};
253: int outfd;
1.4 millert 254:
1.1 millert 255: errno = 0;
1.4 millert 256:
1.1 millert 257: if (access(outpath, W_OK))
1.2 millert 258: err(2, outpath);
1.1 millert 259: act.sa_handler = onsig;
1.4 millert 260: (void)snprintf(toutpath, sizeof(toutpath), "%sXXXXXXXXXX",
261: outpath);
1.1 millert 262: if ((outfd = mkstemp(toutpath)) < 0 ||
263: (outfp = fdopen(outfd, "w")) == 0)
264: err(2, toutpath);
265: outfile = toutpath;
1.4 millert 266:
1.1 millert 267: (void)atexit(cleanup);
268: for (i = 0; sigtable[i]; ++i) /* always unlink toutpath */
269: sigaction(sigtable[i], &act, 0);
270: } else
271: outfile = outpath;
272: if (outfp == NULL && (outfp = fopen(outfile, "w")) == NULL)
273: err(2, outfile);
274: if (mflag)
275: fmerge(-1, filelist, argc-optind, get, outfp, putline, fldtab);
276: else
277: fsort(-1, 0, filelist, argc-optind, outfp, fldtab);
278: if (outfile != outpath) {
279: if (access(outfile, 0))
1.2 millert 280: err(2, outfile);
1.1 millert 281: (void)unlink(outpath);
282: if (link(outfile, outpath))
283: err(2, "cannot link %s: output left in %s",
284: outpath, outfile);
285: (void)unlink(outfile);
286: }
287: exit(0);
288: }
289:
290: static void
291: onsig(s)
292: int s;
293: {
1.4 millert 294:
1.1 millert 295: cleanup();
296: exit(2); /* return 2 on error/interrupt */
297: }
298:
299: static void
300: cleanup()
301: {
1.4 millert 302:
1.1 millert 303: if (toutpath[0])
304: (void)unlink(toutpath);
305: }
306:
307: static void
308: usage(msg)
309: char *msg;
310: {
1.8 ! deraadt 311: extern char *__progname;
1.4 millert 312:
1.1 millert 313: if (msg)
314: (void)fprintf(stderr, "sort: %s\n", msg);
1.8 ! deraadt 315: (void)fprintf(stderr, "usage: %s [-T dir] [-o output] [-cmubdfinr] "
! 316: "[-t char] [-R char] [-k keydef] ... [files]\n", __progname);
1.1 millert 317: exit(2);
318: }