Annotation of src/usr.bin/sort/sort.c, Revision 1.2
1.2 ! millert 1: /* $OpenBSD: sort.c,v 1.1 1997/01/20 19:39:55 millert 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
1.2 ! millert 49: static char rcsid[] = "$OpenBSD: sort.c,v 1.1 1997/01/20 19:39:55 millert Exp $";
1.1 millert 50: #endif
51: #endif /* not lint */
52:
53: /* Sort sorts a file using an optional user-defined key.
54: * Sort uses radix sort for internal sorting, and allows
55: * a choice of merge sort and radix sort for external sorting.
56: */
57:
58: #include "sort.h"
59: #include "fsort.h"
60: #include "pathnames.h"
61:
62: #include <paths.h>
63: #include <signal.h>
64: #include <stdlib.h>
65: #include <string.h>
66: #include <unistd.h>
67:
68: int REC_D = '\n';
69: u_char d_mask[NBINS]; /* flags for rec_d, field_d, <blank> */
70: /*
71: * weight tables. Gweights is one of ascii, Rascii..
72: * modified to weight rec_d = 0 (or 255)
73: */
74: extern u_char gweights[NBINS];
75: u_char ascii[NBINS], Rascii[NBINS], RFtable[NBINS], Ftable[NBINS];
76: /*
77: * masks of ignored characters. Alltable is 256 ones
78: */
79: u_char dtable[NBINS], itable[NBINS], alltable[NBINS];
80: int SINGL_FLD = 0, SEP_FLAG = 0, UNIQUE = 0;
81: struct coldesc clist[(ND+1)*2];
82: int ncols = 0;
83: extern struct coldesc clist[(ND+1)*2];
84: extern int ncols;
85:
86: char devstdin[] = _PATH_STDIN;
87: char toutpath[_POSIX_PATH_MAX];
88: char *tmpdir = _PATH_VARTMP;
89:
90: static void cleanup __P((void));
91: static void onsig __P((int));
92: static void usage __P((char *));
93:
94: int
95: main(argc, argv)
96: int argc;
97: char *argv[];
98: {
99: extern int optind;
100: extern char *optarg;
101: int (*get)();
102: int ch, i, stdinflag = 0, tmp = 0;
103: char cflag = 0, mflag = 0, nflag = 0;
104: char *outfile, *outpath = 0;
105: struct field fldtab[ND+2], *ftpos;
106: union f_handle filelist;
107: FILE *outfp = NULL;
108: memset(fldtab, 0, (ND+2)*sizeof(struct field));
109: memset(d_mask, 0, NBINS);
110: d_mask[REC_D = '\n'] = REC_D_F;
111: SINGL_FLD = SEP_FLAG = 0;
112: d_mask['\t'] = d_mask[' '] = BLANK | FLD_D;
113: ftpos = fldtab;
114: fixit(&argc, argv);
115: if (!issetugid() && (outfile = getenv("TMPDIR")))
116: tmpdir = outfile;
117: while ((ch = getopt(argc, argv, "bcdfik:mHno:rR:t:T:uy:")) != -1) {
118: switch (ch) {
119: case 'b': fldtab->flags |= BI | BT;
120: break;
121: case 'd':
122: case 'i':
123: case 'f':
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 'n':
137: nflag = 1;
138: setfield("1n", ++ftpos, fldtab->flags&(~R));
139: break;
140: case 'k':
141: setfield(optarg, ++ftpos, fldtab->flags);
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 '?':
180: default: usage("");
181: }
182: }
183: if (cflag && argc > optind+1)
184: errx(2, "too many input files for -c option");
185: if (argc - 2 > optind && !strcmp(argv[argc-2], "-o")) {
186: outpath = argv[argc-1];
187: argc -= 2;
188: }
189: if (mflag && argc - optind > (MAXFCT - (16+1))*16)
190: errx(2, "too many input files for -m option");
191: for (i = optind; i < argc; i++) {
192: /* allow one occurrence of /dev/stdin */
193: if (!strcmp(argv[i], "-") || !strcmp(argv[i], devstdin)) {
194: if (stdinflag)
195: warnx("ignoring extra \"%s\" in file list",
196: argv[i]);
197: else {
198: stdinflag = 1;
199: argv[i] = devstdin;
200: }
201: } else if ((ch = access(argv[i], R_OK)))
1.2 ! millert 202: err(2, argv[i]);
1.1 millert 203: }
204: if (!(fldtab->flags & (I|D) || fldtab[1].icol.num)) {
205: SINGL_FLD = 1;
206: fldtab[0].icol.num = 1;
207: } else {
208: if (!fldtab[1].icol.num) {
209: fldtab[0].flags &= ~(BI|BT);
210: setfield("1", ++ftpos, fldtab->flags);
211: }
212: if (nflag)
213: fldtab[1].flags |= fldtab->flags;
214: fldreset(fldtab);
215: fldtab[0].flags &= ~F;
216: }
217: settables(fldtab[0].flags);
218: num_init();
219: fldtab->weights = gweights;
220: if (optind == argc)
221: argv[--optind] = devstdin;
222: filelist.names = argv+optind;
223: if (SINGL_FLD)
224: get = makeline;
225: else
226: get = makekey;
227: if (cflag) {
228: order(filelist, get, fldtab);
229: /* NOT REACHED */
230: }
231: if (!outpath) {
232: (void)snprintf(toutpath,
233: sizeof(toutpath), "%sstdout", _PATH_DEV);
234: outfile = outpath = toutpath;
235: } else if (!(ch = access(outpath, 0)) &&
236: strncmp(_PATH_DEV, outpath, 5)) {
237: struct sigaction act = {0, SIG_BLOCK, 6};
238: int sigtable[] = {SIGHUP, SIGINT, SIGPIPE, SIGXCPU, SIGXFSZ,
239: SIGVTALRM, SIGPROF, 0};
240: int outfd;
241: errno = 0;
242: if (access(outpath, W_OK))
1.2 ! millert 243: err(2, outpath);
1.1 millert 244: act.sa_handler = onsig;
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))
1.2 ! millert 263: err(2, outfile);
1.1 millert 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: onsig(s)
275: int s;
276: {
277: cleanup();
278: exit(2); /* return 2 on error/interrupt */
279: }
280:
281: static void
282: cleanup()
283: {
284: if (toutpath[0])
285: (void)unlink(toutpath);
286: }
287:
288: static void
289: usage(msg)
290: char *msg;
291: {
292: if (msg)
293: (void)fprintf(stderr, "sort: %s\n", msg);
1.2 ! millert 294: (void)fprintf(stderr, "usage: [-T dir] [-o output] [-cmubdfinr] ");
! 295: (void)fprintf(stderr, "[-t char] [-R char] [-k keydef] ... [files]\n");
1.1 millert 296: exit(2);
297: }