split.c revision 68887 
1/*
2 * Copyright (c) 1987, 1993, 1994
3 *	The Regents of the University of California.  All rights reserved.
4 *
5 * Redistribution and use in source and binary forms, with or without
6 * modification, are permitted provided that the following conditions
7 * are met:
8 * 1. Redistributions of source code must retain the above copyright
9 *    notice, this list of conditions and the following disclaimer.
10 * 2. Redistributions in binary form must reproduce the above copyright
11 *    notice, this list of conditions and the following disclaimer in the
12 *    documentation and/or other materials provided with the distribution.
13 * 3. All advertising materials mentioning features or use of this software
14 *    must display the following acknowledgement:
15 *	This product includes software developed by the University of
16 *	California, Berkeley and its contributors.
17 * 4. Neither the name of the University nor the names of its contributors
18 *    may be used to endorse or promote products derived from this software
19 *    without specific prior written permission.
20 *
21 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
22 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
23 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
24 * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
25 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
26 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
27 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
28 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
29 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
30 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
31 * SUCH DAMAGE.
32 */
33
34#ifndef lint
35static const char copyright[] =
36"@(#) Copyright (c) 1987, 1993, 1994\n\
37	The Regents of the University of California.  All rights reserved.\n";
38#endif /* not lint */
39
40#ifndef lint
41#if 0
42static char sccsid[] = "@(#)split.c	8.2 (Berkeley) 4/16/94";
43#else
44static const char rcsid[] =
45  "$FreeBSD: head/usr.bin/split/split.c 68887 2000-11-19 01:44:20Z jwd $";
46#endif
47#endif /* not lint */
48
49#include <sys/param.h>
50#include <sys/types.h>
51
52#include <ctype.h>
53#include <err.h>
54#include <fcntl.h>
55#include <stdio.h>
56#include <stdlib.h>
57#include <string.h>
58#include <unistd.h>
59#include <regex.h>
60#include <sysexits.h>
61
62#define DEFLINE	1000			/* Default num lines per file. */
63
64size_t	 bytecnt;			/* Byte count to split on. */
65long	 numlines;			/* Line count to split on. */
66int	 file_open;			/* If a file open. */
67int	 ifd = -1, ofd = -1;		/* Input/output file descriptors. */
68char	 bfr[MAXBSIZE];			/* I/O buffer. */
69char	 fname[MAXPATHLEN];		/* File name prefix. */
70regex_t	 rgx;
71int	 pflag;
72
73void newfile __P((void));
74void split1 __P((void));
75void split2 __P((void));
76static void usage __P((void));
77
78int
79main(argc, argv)
80	int argc;
81	char *argv[];
82{
83	int ch;
84	char *ep, *p;
85
86	while ((ch = getopt(argc, argv, "-0123456789b:l:p:")) != -1)
87		switch (ch) {
88		case '0': case '1': case '2': case '3': case '4':
89		case '5': case '6': case '7': case '8': case '9':
90			/*
91			 * Undocumented kludge: split was originally designed
92			 * to take a number after a dash.
93			 */
94			if (numlines == 0) {
95				p = argv[optind - 1];
96				if (p[0] == '-' && p[1] == ch && !p[2])
97					numlines = strtol(++p, &ep, 10);
98				else
99					numlines =
100					    strtol(argv[optind] + 1, &ep, 10);
101				if (numlines <= 0 || *ep)
102					errx(EX_USAGE,
103					    "%s: illegal line count", optarg);
104			}
105			break;
106		case '-':		/* Undocumented: historic stdin flag. */
107			if (ifd != -1)
108				usage();
109			ifd = 0;
110			break;
111		case 'b':		/* Byte count. */
112			if ((bytecnt = strtoq(optarg, &ep, 10)) <= 0 ||
113			    (*ep != '\0' && *ep != 'k' && *ep != 'm'))
114				errx(EX_USAGE,
115				    "%s: illegal byte count", optarg);
116			if (*ep == 'k')
117				bytecnt *= 1024;
118			else if (*ep == 'm')
119				bytecnt *= 1048576;
120			break;
121		case 'p' :      /* pattern matching. */
122			if (regcomp(&rgx, optarg, REG_EXTENDED|REG_NOSUB) != 0)
123				errx(EX_USAGE, "%s: illegal regexp", optarg);
124			pflag = 1;
125			break;
126		case 'l':		/* Line count. */
127			if (numlines != 0)
128				usage();
129			if ((numlines = strtol(optarg, &ep, 10)) <= 0 || *ep)
130				errx(EX_USAGE,
131				    "%s: illegal line count", optarg);
132			break;
133		default:
134			usage();
135		}
136	argv += optind;
137	argc -= optind;
138
139	if (*argv != NULL)
140		if (ifd == -1) {		/* Input file. */
141			if ((ifd = open(*argv, O_RDONLY, 0)) < 0)
142				err(EX_NOINPUT, "%s", *argv);
143			++argv;
144		}
145	if (*argv != NULL)			/* File name prefix. */
146		(void)strcpy(fname, *argv++);
147	if (*argv != NULL)
148		usage();
149
150	if (pflag && (numlines != 0 || bytecnt != 0))
151		usage();
152
153	if (numlines == 0)
154		numlines = DEFLINE;
155	else if (bytecnt != 0)
156		usage();
157
158	if (ifd == -1)				/* Stdin by default. */
159		ifd = 0;
160
161	if (bytecnt) {
162		split1();
163		exit (0);
164	}
165	split2();
166	if (pflag)
167		regfree(&rgx);
168	exit(0);
169}
170
171/*
172 * split1 --
173 *	Split the input by bytes.
174 */
175void
176split1()
177{
178	size_t bcnt, dist, len;
179	char *C;
180
181	for (bcnt = 0;;)
182		switch ((len = read(ifd, bfr, MAXBSIZE))) {
183		case 0:
184			exit(0);
185		case -1:
186			err(EX_IOERR, "read");
187			/* NOTREACHED */
188		default:
189			if (!file_open)
190				newfile();
191			if (bcnt + len >= bytecnt) {
192				dist = bytecnt - bcnt;
193				if (write(ofd, bfr, dist) != dist)
194					err(EX_IOERR, "write");
195				len -= dist;
196				for (C = bfr + dist; len >= bytecnt;
197				    len -= bytecnt, C += bytecnt) {
198					newfile();
199					if (write(ofd,
200					    C, (int)bytecnt) != bytecnt)
201						err(EX_IOERR, "write");
202				}
203				if (len != 0) {
204					newfile();
205					if (write(ofd, C, len) != len)
206						err(EX_IOERR, "write");
207				} else
208					file_open = 0;
209				bcnt = len;
210			} else {
211				bcnt += len;
212				if (write(ofd, bfr, len) != len)
213					err(EX_IOERR, "write");
214			}
215		}
216}
217
218/*
219 * split2 --
220 *	Split the input by lines.
221 */
222void
223split2()
224{
225	long lcnt = 0;
226	FILE *infp;
227
228	/* Stick a stream on top of input file descriptor */
229	if ((infp = fdopen(ifd, "r")) == NULL)
230		err(EX_NOINPUT, "fdopen");
231
232	/* Process input one line at a time */
233	while (fgets(bfr, sizeof(bfr), infp) != NULL) {
234		const int len = strlen(bfr);
235
236		/* If line is too long to deal with, just write it out */
237		if (bfr[len - 1] != '\n')
238			goto writeit;
239
240		/* Check if we need to start a new file */
241		if (pflag) {
242			regmatch_t pmatch;
243
244			pmatch.rm_so = 0;
245			pmatch.rm_eo = len - 1;
246			if (regexec(&rgx, bfr, 0, &pmatch, REG_STARTEND) == 0)
247				newfile();
248		} else if (lcnt++ == numlines) {
249			newfile();
250			lcnt = 1;
251		}
252
253writeit:
254		/* Open output file if needed */
255		if (!file_open)
256			newfile();
257
258		/* Write out line */
259		if (write(ofd, bfr, len) != len)
260			err(EX_IOERR, "write");
261	}
262
263	/* EOF or error? */
264	if (ferror(infp))
265		err(EX_IOERR, "read");
266	else
267		exit(0);
268}
269
270/*
271 * newfile --
272 *	Open a new output file.
273 */
274void
275newfile()
276{
277	static long fnum;
278	static int defname;
279	static char *fpnt;
280
281	if (ofd == -1) {
282		if (fname[0] == '\0') {
283			fname[0] = 'x';
284			fpnt = fname + 1;
285			defname = 1;
286		} else {
287			fpnt = fname + strlen(fname);
288			defname = 0;
289		}
290		ofd = fileno(stdout);
291	}
292	/*
293	 * Hack to increase max files; original code wandered through
294	 * magic characters.  Maximum files is 3 * 26 * 26 == 2028
295	 */
296#define MAXFILES	676
297	if (fnum == MAXFILES) {
298		if (!defname || fname[0] == 'z')
299			errx(EX_DATAERR, "too many files");
300		++fname[0];
301		fnum = 0;
302	}
303	fpnt[0] = fnum / 26 + 'a';
304	fpnt[1] = fnum % 26 + 'a';
305	++fnum;
306	if (!freopen(fname, "w", stdout))
307		err(EX_IOERR, "%s", fname);
308	file_open = 1;
309}
310
311static void
312usage()
313{
314	(void)fprintf(stderr,
315"usage: split [-b byte_count] [-l line_count] [-p pattern] [file [prefix]]\n");
316	exit(EX_USAGE);
317}
318