xref: /freebsd/usr.bin/split/split.c (revision a6dd1c93f49cf4b8e1d7ed6ed4100aef37d0ceab)
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 #include <sys/cdefs.h>
35 __FBSDID("$FreeBSD$");
36 
37 #ifndef lint
38 static const char copyright[] =
39 "@(#) Copyright (c) 1987, 1993, 1994\n\
40 	The Regents of the University of California.  All rights reserved.\n";
41 #endif
42 
43 #ifndef lint
44 static const char sccsid[] = "@(#)split.c	8.2 (Berkeley) 4/16/94";
45 #endif
46 
47 #include <sys/param.h>
48 
49 #include <ctype.h>
50 #include <err.h>
51 #include <errno.h>
52 #include <fcntl.h>
53 #include <inttypes.h>
54 #include <limits.h>
55 #include <locale.h>
56 #include <stdint.h>
57 #include <stdio.h>
58 #include <stdlib.h>
59 #include <string.h>
60 #include <unistd.h>
61 #include <regex.h>
62 #include <sysexits.h>
63 
64 #define DEFLINE	1000			/* Default num lines per file. */
65 
66 off_t	 bytecnt;			/* Byte count to split on. */
67 long	 numlines;			/* Line count to split on. */
68 int	 file_open;			/* If a file open. */
69 int	 ifd = -1, ofd = -1;		/* Input/output file descriptors. */
70 char	 bfr[MAXBSIZE];			/* I/O buffer. */
71 char	 fname[MAXPATHLEN];		/* File name prefix. */
72 regex_t	 rgx;
73 int	 pflag;
74 long	 sufflen = 2;			/* File name suffix length. */
75 
76 void newfile(void);
77 void split1(void);
78 void split2(void);
79 static void usage(void);
80 
81 int
82 main(int argc, char **argv)
83 {
84 	intmax_t bytecnti;
85 	long scale;
86 	int ch;
87 	char *ep, *p;
88 
89 	setlocale(LC_ALL, "");
90 
91 	while ((ch = getopt(argc, argv, "0123456789a:b:l:p:")) != -1)
92 		switch (ch) {
93 		case '0': case '1': case '2': case '3': case '4':
94 		case '5': case '6': case '7': case '8': case '9':
95 			/*
96 			 * Undocumented kludge: split was originally designed
97 			 * to take a number after a dash.
98 			 */
99 			if (numlines == 0) {
100 				p = argv[optind - 1];
101 				if (p[0] == '-' && p[1] == ch && !p[2])
102 					numlines = strtol(++p, &ep, 10);
103 				else
104 					numlines =
105 					    strtol(argv[optind] + 1, &ep, 10);
106 				if (numlines <= 0 || *ep)
107 					errx(EX_USAGE,
108 					    "%s: illegal line count", optarg);
109 			}
110 			break;
111 		case 'a':		/* Suffix length */
112 			if ((sufflen = strtol(optarg, &ep, 10)) <= 0 || *ep)
113 				errx(EX_USAGE,
114 				    "%s: illegal suffix length", optarg);
115 			break;
116 		case 'b':		/* Byte count. */
117 			errno = 0;
118 			if ((bytecnti = strtoimax(optarg, &ep, 10)) <= 0 ||
119 			    strchr("kKmMgG", *ep) == NULL || errno != 0)
120 				errx(EX_USAGE,
121 				    "%s: illegal byte count", optarg);
122 			if (*ep == 'k' || *ep == 'K')
123 				scale = 1024;
124 			else if (*ep == 'm' || *ep == 'M')
125 				scale = 1024 * 1024;
126 			else if (*ep == 'g' || *ep == 'G')
127 				scale = 1024 * 1024 * 1024;
128 			else
129 				scale = 1;
130 			if (bytecnti > OFF_MAX / scale)
131 				errx(EX_USAGE, "%s: offset too large", optarg);
132 			bytecnt = (off_t)(bytecnti * scale);
133 			break;
134 		case 'l':		/* Line count. */
135 			if (numlines != 0)
136 				usage();
137 			if ((numlines = strtol(optarg, &ep, 10)) <= 0 || *ep)
138 				errx(EX_USAGE,
139 				    "%s: illegal line count", optarg);
140 			break;
141 		case 'p':		/* pattern matching. */
142 			if (regcomp(&rgx, optarg, REG_EXTENDED|REG_NOSUB) != 0)
143 				errx(EX_USAGE, "%s: illegal regexp", optarg);
144 			pflag = 1;
145 			break;
146 		default:
147 			usage();
148 		}
149 	argv += optind;
150 	argc -= optind;
151 
152 	if (*argv != NULL) {			/* Input file. */
153 		if (strcmp(*argv, "-") == 0)
154 			ifd = STDIN_FILENO;
155 		else if ((ifd = open(*argv, O_RDONLY, 0)) < 0)
156 			err(EX_NOINPUT, "%s", *argv);
157 		++argv;
158 	}
159 	if (*argv != NULL)			/* File name prefix. */
160 		if (strlcpy(fname, *argv++, sizeof(fname)) >= sizeof(fname))
161 			errx(EX_USAGE, "file name prefix is too long");
162 	if (*argv != NULL)
163 		usage();
164 
165 	if (strlen(fname) + (unsigned long)sufflen >= sizeof(fname))
166 		errx(EX_USAGE, "suffix is too long");
167 	if (pflag && (numlines != 0 || bytecnt != 0))
168 		usage();
169 
170 	if (numlines == 0)
171 		numlines = DEFLINE;
172 	else if (bytecnt != 0)
173 		usage();
174 
175 	if (ifd == -1)				/* Stdin by default. */
176 		ifd = 0;
177 
178 	if (bytecnt) {
179 		split1();
180 		exit (0);
181 	}
182 	split2();
183 	if (pflag)
184 		regfree(&rgx);
185 	exit(0);
186 }
187 
188 /*
189  * split1 --
190  *	Split the input by bytes.
191  */
192 void
193 split1(void)
194 {
195 	off_t bcnt;
196 	char *C;
197 	ssize_t dist, len;
198 
199 	for (bcnt = 0;;)
200 		switch ((len = read(ifd, bfr, MAXBSIZE))) {
201 		case 0:
202 			exit(0);
203 		case -1:
204 			err(EX_IOERR, "read");
205 			/* NOTREACHED */
206 		default:
207 			if (!file_open)
208 				newfile();
209 			if (bcnt + len >= bytecnt) {
210 				dist = bytecnt - bcnt;
211 				if (write(ofd, bfr, dist) != dist)
212 					err(EX_IOERR, "write");
213 				len -= dist;
214 				for (C = bfr + dist; len >= bytecnt;
215 				    len -= bytecnt, C += bytecnt) {
216 					newfile();
217 					if (write(ofd,
218 					    C, bytecnt) != bytecnt)
219 						err(EX_IOERR, "write");
220 				}
221 				if (len != 0) {
222 					newfile();
223 					if (write(ofd, C, len) != len)
224 						err(EX_IOERR, "write");
225 				} else
226 					file_open = 0;
227 				bcnt = len;
228 			} else {
229 				bcnt += len;
230 				if (write(ofd, bfr, len) != len)
231 					err(EX_IOERR, "write");
232 			}
233 		}
234 }
235 
236 /*
237  * split2 --
238  *	Split the input by lines.
239  */
240 void
241 split2(void)
242 {
243 	long lcnt = 0;
244 	FILE *infp;
245 
246 	/* Stick a stream on top of input file descriptor */
247 	if ((infp = fdopen(ifd, "r")) == NULL)
248 		err(EX_NOINPUT, "fdopen");
249 
250 	/* Process input one line at a time */
251 	while (fgets(bfr, sizeof(bfr), infp) != NULL) {
252 		const int len = strlen(bfr);
253 
254 		/* If line is too long to deal with, just write it out */
255 		if (bfr[len - 1] != '\n')
256 			goto writeit;
257 
258 		/* Check if we need to start a new file */
259 		if (pflag) {
260 			regmatch_t pmatch;
261 
262 			pmatch.rm_so = 0;
263 			pmatch.rm_eo = len - 1;
264 			if (regexec(&rgx, bfr, 0, &pmatch, REG_STARTEND) == 0)
265 				newfile();
266 		} else if (lcnt++ == numlines) {
267 			newfile();
268 			lcnt = 1;
269 		}
270 
271 writeit:
272 		/* Open output file if needed */
273 		if (!file_open)
274 			newfile();
275 
276 		/* Write out line */
277 		if (write(ofd, bfr, len) != len)
278 			err(EX_IOERR, "write");
279 	}
280 
281 	/* EOF or error? */
282 	if (ferror(infp))
283 		err(EX_IOERR, "read");
284 	else
285 		exit(0);
286 }
287 
288 /*
289  * newfile --
290  *	Open a new output file.
291  */
292 void
293 newfile(void)
294 {
295 	long i, maxfiles, tfnum;
296 	static long fnum;
297 	static int defname;
298 	static char *fpnt;
299 
300 	if (ofd == -1) {
301 		if (fname[0] == '\0') {
302 			fname[0] = 'x';
303 			fpnt = fname + 1;
304 			defname = 1;
305 		} else {
306 			fpnt = fname + strlen(fname);
307 			defname = 0;
308 		}
309 		ofd = fileno(stdout);
310 	}
311 
312 	/* maxfiles = 26^sufflen, but don't use libm. */
313 	for (maxfiles = 1, i = 0; i < sufflen; i++)
314 		if ((maxfiles *= 26) <= 0)
315 			errx(EX_USAGE, "suffix is too long (max %ld)", i);
316 
317 	if (fnum == maxfiles)
318 		errx(EX_DATAERR, "too many files");
319 
320 	/* Generate suffix of sufflen letters */
321 	tfnum = fnum;
322 	i = sufflen - 1;
323 	do {
324 		fpnt[i] = tfnum % 26 + 'a';
325 		tfnum /= 26;
326 	} while (i-- > 0);
327 	fpnt[sufflen] = '\0';
328 
329 	++fnum;
330 	if (!freopen(fname, "w", stdout))
331 		err(EX_IOERR, "%s", fname);
332 	file_open = 1;
333 }
334 
335 static void
336 usage(void)
337 {
338 	(void)fprintf(stderr,
339 "usage: split [-l line_count] [-a suffix_length] [file [prefix]]\n"
340 "       split -b byte_count[K|k|M|m|G|g] [-a suffix_length] [file [prefix]]\n"
341 "       split -p pattern [-a suffix_length] [file [prefix]]\n");
342 	exit(EX_USAGE);
343 }
344