Annotation of src/usr.bin/sort/sort.c, Revision 1.11
1.11 ! deraadt 1: /* $OpenBSD: sort.c,v 1.10 1998/07/20 20:02:19 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
1.11 ! deraadt 49: static char rcsid[] = "$OpenBSD: sort.c,v 1.10 1998/07/20 20:02:19 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>
1.10 mickey 68: #include <err.h>
1.1 millert 69:
70: int REC_D = '\n';
71: u_char d_mask[NBINS]; /* flags for rec_d, field_d, <blank> */
1.4 millert 72:
1.1 millert 73: /*
74: * weight tables. Gweights is one of ascii, Rascii..
75: * modified to weight rec_d = 0 (or 255)
76: */
77: extern u_char gweights[NBINS];
78: u_char ascii[NBINS], Rascii[NBINS], RFtable[NBINS], Ftable[NBINS];
1.4 millert 79:
1.1 millert 80: /*
81: * masks of ignored characters. Alltable is 256 ones
82: */
83: u_char dtable[NBINS], itable[NBINS], alltable[NBINS];
84: int SINGL_FLD = 0, SEP_FLAG = 0, UNIQUE = 0;
85: struct coldesc clist[(ND+1)*2];
86: int ncols = 0;
87: extern struct coldesc clist[(ND+1)*2];
88: extern int ncols;
89:
90: char devstdin[] = _PATH_STDIN;
91: char toutpath[_POSIX_PATH_MAX];
92: char *tmpdir = _PATH_VARTMP;
93:
94: static void cleanup __P((void));
95: static void onsig __P((int));
96: static void usage __P((char *));
97:
98: int
99: main(argc, argv)
100: int argc;
101: char *argv[];
102: {
103: int (*get)();
104: int ch, i, stdinflag = 0, tmp = 0;
1.6 millert 105: char cflag = 0, mflag = 0;
1.1 millert 106: char *outfile, *outpath = 0;
107: struct field fldtab[ND+2], *ftpos;
108: union f_handle filelist;
109: FILE *outfp = NULL;
1.4 millert 110:
1.1 millert 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: fixit(&argc, argv);
118: if (!issetugid() && (outfile = getenv("TMPDIR")))
119: tmpdir = outfile;
120: while ((ch = getopt(argc, argv, "bcdfik:mHno:rR:t:T:uy:")) != -1) {
1.4 millert 121: switch (ch) {
1.1 millert 122: case 'b': fldtab->flags |= BI | BT;
123: break;
124: case 'd':
1.4 millert 125: case 'f':
1.1 millert 126: case 'i':
1.5 millert 127: case 'n':
1.1 millert 128: case 'r': tmp |= optval(ch, 0);
129: if (tmp & R && tmp & F)
130: fldtab->weights = RFtable;
131: else if (tmp & F)
132: fldtab->weights = Ftable;
1.4 millert 133: else if (tmp & R)
1.1 millert 134: fldtab->weights = Rascii;
135: fldtab->flags |= tmp;
136: break;
137: case 'o':
138: outpath = optarg;
139: break;
140: case 'k':
1.5 millert 141: setfield(optarg, ++ftpos, fldtab->flags);
1.1 millert 142: break;
143: case 't':
144: if (SEP_FLAG)
145: usage("multiple field delimiters");
146: SEP_FLAG = 1;
147: d_mask[' '] &= ~FLD_D;
148: d_mask['\t'] &= ~FLD_D;
149: d_mask[(int)*optarg] |= FLD_D;
150: if (d_mask[(int)*optarg] & REC_D_F)
151: err(2, "record/field delimiter clash");
152: break;
153: case 'R':
154: if (REC_D != '\n')
155: usage("multiple record delimiters");
156: if ('\n' == (REC_D = *optarg))
157: break;
158: d_mask['\n'] = d_mask[' '];
159: d_mask[REC_D] = REC_D_F;
160: break;
161: case 'T':
162: tmpdir = optarg;
163: break;
164: case 'u':
165: UNIQUE = 1;
166: break;
167: case 'c':
168: cflag = 1;
169: break;
170: case 'm':
171: mflag = 1;
172: break;
173: case 'H':
174: PANIC = 0;
175: break;
176: case 'y':
177: /* accept -y for backwards compat. */
178: break;
179: case '?':
1.8 deraadt 180: default:
181: usage(NULL);
1.1 millert 182: }
183: }
1.4 millert 184:
1.1 millert 185: if (cflag && argc > optind+1)
186: errx(2, "too many input files for -c option");
1.4 millert 187:
1.1 millert 188: if (argc - 2 > optind && !strcmp(argv[argc-2], "-o")) {
189: outpath = argv[argc-1];
190: argc -= 2;
191: }
1.4 millert 192:
1.1 millert 193: if (mflag && argc - optind > (MAXFCT - (16+1))*16)
194: errx(2, "too many input files for -m option");
1.4 millert 195:
1.1 millert 196: for (i = optind; i < argc; i++) {
197: /* allow one occurrence of /dev/stdin */
198: if (!strcmp(argv[i], "-") || !strcmp(argv[i], devstdin)) {
199: if (stdinflag)
200: warnx("ignoring extra \"%s\" in file list",
201: argv[i]);
202: else {
203: stdinflag = 1;
204: argv[i] = devstdin;
205: }
206: } else if ((ch = access(argv[i], R_OK)))
1.2 millert 207: err(2, argv[i]);
1.1 millert 208: }
1.4 millert 209:
1.5 millert 210: if (!(fldtab->flags & (I|D|N) || fldtab[1].icol.num)) {
1.1 millert 211: SINGL_FLD = 1;
212: fldtab[0].icol.num = 1;
213: } else {
214: if (!fldtab[1].icol.num) {
215: fldtab[0].flags &= ~(BI|BT);
216: setfield("1", ++ftpos, fldtab->flags);
217: }
218: fldreset(fldtab);
219: fldtab[0].flags &= ~F;
220: }
221: settables(fldtab[0].flags);
222: num_init();
223: fldtab->weights = gweights;
1.4 millert 224:
1.3 deraadt 225: if (optind == argc) {
226: static char *names[2];
227:
228: names[0] = devstdin;
229: names[1] = NULL;
230: filelist.names = names;
231: optind--;
232: } else
233: filelist.names = argv+optind;
1.4 millert 234:
1.1 millert 235: if (SINGL_FLD)
236: get = makeline;
237: else
238: get = makekey;
1.4 millert 239:
1.1 millert 240: if (cflag) {
241: order(filelist, get, fldtab);
242: /* NOT REACHED */
243: }
1.4 millert 244:
1.1 millert 245: if (!outpath) {
246: (void)snprintf(toutpath,
247: sizeof(toutpath), "%sstdout", _PATH_DEV);
248: outfile = outpath = toutpath;
249: } else if (!(ch = access(outpath, 0)) &&
250: strncmp(_PATH_DEV, outpath, 5)) {
251: struct sigaction act = {0, SIG_BLOCK, 6};
252: int sigtable[] = {SIGHUP, SIGINT, SIGPIPE, SIGXCPU, SIGXFSZ,
253: SIGVTALRM, SIGPROF, 0};
254: int outfd;
1.4 millert 255:
1.1 millert 256: errno = 0;
1.4 millert 257:
1.1 millert 258: if (access(outpath, W_OK))
1.2 millert 259: err(2, outpath);
1.1 millert 260: act.sa_handler = onsig;
1.4 millert 261: (void)snprintf(toutpath, sizeof(toutpath), "%sXXXXXXXXXX",
262: outpath);
1.1 millert 263: if ((outfd = mkstemp(toutpath)) < 0 ||
264: (outfp = fdopen(outfd, "w")) == 0)
265: err(2, toutpath);
266: outfile = toutpath;
1.4 millert 267:
1.1 millert 268: (void)atexit(cleanup);
269: for (i = 0; sigtable[i]; ++i) /* always unlink toutpath */
270: sigaction(sigtable[i], &act, 0);
271: } else
272: outfile = outpath;
273: if (outfp == NULL && (outfp = fopen(outfile, "w")) == NULL)
274: err(2, outfile);
275: if (mflag)
276: fmerge(-1, filelist, argc-optind, get, outfp, putline, fldtab);
277: else
278: fsort(-1, 0, filelist, argc-optind, outfp, fldtab);
279: if (outfile != outpath) {
280: if (access(outfile, 0))
1.2 millert 281: err(2, outfile);
1.1 millert 282: (void)unlink(outpath);
283: if (link(outfile, outpath))
284: err(2, "cannot link %s: output left in %s",
285: outpath, outfile);
286: (void)unlink(outfile);
287: }
288: exit(0);
289: }
290:
291: static void
292: onsig(s)
293: int s;
294: {
1.4 millert 295:
1.1 millert 296: cleanup();
297: exit(2); /* return 2 on error/interrupt */
298: }
299:
300: static void
301: cleanup()
302: {
1.4 millert 303:
1.1 millert 304: if (toutpath[0])
305: (void)unlink(toutpath);
306: }
307:
308: static void
309: usage(msg)
310: char *msg;
311: {
1.8 deraadt 312: extern char *__progname;
1.4 millert 313:
1.1 millert 314: if (msg)
1.10 mickey 315: warnx(msg);
1.11 ! deraadt 316: (void)fprintf(stderr, "usage: %s [-T dir] [-o output] [-cmubdfinrH] "
1.9 deraadt 317: "[-t char] [-R char] [-k keydef] ... [files]\n", __progname);
1.1 millert 318: exit(2);
319: }