xref: /dragonfly/usr.bin/split/split.c (revision 8accc937)
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  * @(#) Copyright (c) 1987, 1993, 1994 The Regents of the University of California.  All rights reserved.
34  * @(#)split.c	8.2 (Berkeley) 4/16/94
35  * $FreeBSD: src/usr.bin/split/split.c,v 1.6.2.2 2002/07/25 12:46:36 tjr Exp $
36  */
37 
38 #include <sys/param.h>
39 
40 #include <ctype.h>
41 #include <err.h>
42 #include <errno.h>
43 #include <fcntl.h>
44 #include <limits.h>
45 #include <stdio.h>
46 #include <stdlib.h>
47 #include <string.h>
48 #include <unistd.h>
49 #include <regex.h>
50 #include <sysexits.h>
51 
52 #define DEFLINE	1000			/* Default num lines per file. */
53 
54 off_t	 bytecnt;			/* Byte count to split on. */
55 long	 numlines;			/* Line count to split on. */
56 int	 file_open;			/* If a file open. */
57 int	 ifd = -1, ofd = -1;		/* Input/output file descriptors. */
58 char	 *bfr;				/* I/O buffer. */
59 char	 fname[MAXPATHLEN];		/* File name prefix. */
60 regex_t	 rgx;
61 int	 pflag;
62 long	 sufflen = 2;			/* File name suffix length. */
63 
64 void newfile(void);
65 void split1(void);
66 void split2(void);
67 static void usage(void);
68 
69 int
70 main(int argc, char **argv)
71 {
72 	long long bytecnti;
73 	long scale;
74 	int ch;
75 	char *ep, *p;
76 
77 	while ((ch = getopt(argc, argv, "0123456789a:b:l:p:")) != -1)
78 		switch (ch) {
79 		case '0': case '1': case '2': case '3': case '4':
80 		case '5': case '6': case '7': case '8': case '9':
81 			/*
82 			 * Undocumented kludge: split was originally designed
83 			 * to take a number after a dash.
84 			 */
85 			if (numlines == 0) {
86 				p = argv[optind - 1];
87 				if (p[0] == '-' && p[1] == ch && !p[2])
88 					numlines = strtol(++p, &ep, 10);
89 				else
90 					numlines =
91 					    strtol(argv[optind] + 1, &ep, 10);
92 				if (numlines <= 0 || *ep)
93 					errx(EX_USAGE,
94 					    "%s: illegal line count", optarg);
95 			}
96 			break;
97 		case 'a':		/* Suffix length */
98 			if ((sufflen = strtol(optarg, &ep, 10)) <= 0 || *ep)
99 				errx(EX_USAGE,
100 				    "%s: illegal suffix length", optarg);
101 			break;
102 		case 'b':		/* Byte count. */
103 			errno = 0;
104 			if ((bytecnti = strtoll(optarg, &ep, 10)) <= 0 ||
105 			    (*ep != '\0' && *ep != 'k' && *ep != 'm') ||
106 			    errno != 0)
107 				errx(EX_USAGE,
108 				    "%s: illegal byte count", optarg);
109 			if (*ep == 'k')
110 				scale = 1024;
111 			else if (*ep == 'm')
112 				scale = 1024 * 1024;
113 			else
114 				scale = 1;
115 			bytecnt = (off_t)(bytecnti * scale);
116 			break;
117 		case 'p' :      /* pattern matching. */
118 			if (regcomp(&rgx, optarg, REG_EXTENDED|REG_NOSUB) != 0)
119 				errx(EX_USAGE, "%s: illegal regexp", optarg);
120 			pflag = 1;
121 			break;
122 		case 'l':		/* Line count. */
123 			if (numlines != 0)
124 				usage();
125 			if ((numlines = strtol(optarg, &ep, 10)) <= 0 || *ep)
126 				errx(EX_USAGE,
127 				    "%s: illegal line count", optarg);
128 			break;
129 		default:
130 			usage();
131 		}
132 	argv += optind;
133 	argc -= optind;
134 
135 	if (*argv != NULL) {
136 		if (strcmp(*argv, "-") == 0)
137 			ifd = STDIN_FILENO;
138 		else if ((ifd = open(*argv, O_RDONLY, 0)) < 0)
139 			err(EX_NOINPUT, "%s", *argv);
140 		++argv;
141 	}
142 
143 	if (*argv != NULL)			/* File name prefix. */
144 		if (strlcpy(fname, *argv++, sizeof(fname)) >= sizeof(fname))
145 			errx(EX_USAGE, "file name prefix is too long");
146 	if (*argv != NULL)
147 		usage();
148 
149 	if (strlen(fname) + (unsigned long)sufflen >= sizeof(fname))
150 		errx(EX_USAGE, "suffix is too long");
151 	if (pflag && (numlines != 0 || bytecnt != 0))
152 		usage();
153 
154 	if (numlines == 0)
155 		numlines = DEFLINE;
156 	else if (bytecnt != 0)
157 		usage();
158 
159 	if (ifd == -1)				/* Stdin by default. */
160 		ifd = 0;
161 
162 	if (bytecnt) {
163 		split1();
164 		exit (0);
165 	}
166 	split2();
167 	if (pflag)
168 		regfree(&rgx);
169 	exit(0);
170 }
171 
172 /*
173  * split1 --
174  *	Split the input by bytes.
175  */
176 void
177 split1(void)
178 {
179 	off_t bcnt;
180 	char *C;
181 	ssize_t dist, len;
182 
183 	if((bfr = (char *)malloc(bytecnt)) == NULL)
184 		err(EX_OSERR, "malloc");
185 
186 	for (bcnt = 0;;)
187 		switch ((len = read(ifd, bfr, bytecnt))) {
188 		case 0:
189 			free(bfr);
190 			exit(0);
191 		case -1:
192 			free(bfr);
193 			err(EX_IOERR, "read");
194 			/* NOTREACHED */
195 		default:
196 			if (!file_open)
197 				newfile();
198 			if (bcnt + len >= bytecnt) {
199 				dist = bytecnt - bcnt;
200 				if (write(ofd, bfr, dist) != dist)
201 					err(EX_IOERR, "write");
202 				len -= dist;
203 				for (C = bfr + dist; len >= bytecnt;
204 				    len -= bytecnt, C += bytecnt) {
205 					newfile();
206 					if (write(ofd,
207 					    C, (int)bytecnt) != bytecnt) {
208 						free(bfr);
209 						err(EX_IOERR, "write");
210 					}
211 				}
212 				if (len != 0) {
213 					newfile();
214 					if (write(ofd, C, len) != len) {
215 						free(bfr);
216 						err(EX_IOERR, "write");
217 					}
218 				} else
219 					file_open = 0;
220 				bcnt = len;
221 			} else {
222 				bcnt += len;
223 				if (write(ofd, bfr, len) != len) {
224 					free(bfr);
225 					err(EX_IOERR, "write");
226 				}
227 			}
228 		}
229 	free(bfr);
230 }
231 
232 /*
233  * split2 --
234  *	Split the input by lines.
235  */
236 void
237 split2(void)
238 {
239 	int startofline = 1;
240 	long lcnt = 0;
241 	FILE *infp;
242 
243 	/* Stick a stream on top of input file descriptor */
244 	if ((infp = fdopen(ifd, "r")) == NULL)
245 		err(EX_NOINPUT, "fdopen");
246 
247 	if((bfr = (char *)malloc(MAXBSIZE)) == NULL)
248 		err(EX_OSERR, "malloc");
249 
250 	/* Process input one line at a time */
251 	while (fgets(bfr, MAXBSIZE, infp) != NULL) {
252 		const int len = strlen(bfr);
253 
254 		/* Consider starting a new file only when at beginning of a line */
255 		if (startofline) {
256 			/* Check if we need to start a new file */
257 			if (pflag) {
258 				regmatch_t pmatch;
259 
260 				pmatch.rm_so = 0;
261 				pmatch.rm_eo = len - 1;
262 				if (regexec(&rgx, bfr, 0, &pmatch, REG_STARTEND) == 0)
263 					newfile();
264 			} else if (lcnt++ == numlines) {
265 				newfile();
266 				lcnt = 1;
267 			}
268 		}
269 
270 		if (bfr[len - 1] != '\n')
271 			startofline = 0;
272 		else
273 			startofline = 1;
274 
275 		/* Open output file if needed */
276 		if (!file_open)
277 			newfile();
278 
279 		/* Write out line */
280 		if (write(ofd, bfr, len) != len) {
281 			free(bfr);
282 			err(EX_IOERR, "write");
283 		}
284 	}
285 
286 	free(bfr);
287 
288 	/* EOF or error? */
289 	if (ferror(infp))
290 		err(EX_IOERR, "read");
291 	else
292 		exit(0);
293 }
294 
295 /*
296  * newfile --
297  *	Open a new output file.
298  */
299 void
300 newfile(void)
301 {
302 	long i, maxfiles, tfnum;
303 	static long fnum;
304 	static char *fpnt;
305 
306 	if (ofd == -1) {
307 		if (fname[0] == '\0') {
308 			fname[0] = 'x';
309 			fpnt = fname + 1;
310 		} else {
311 			fpnt = fname + strlen(fname);
312 		}
313 		ofd = fileno(stdout);
314 	}
315 
316 	/* maxfiles = 26^sufflen, but don't use libm. */
317 	for (maxfiles = 1, i = 0; i < sufflen; i++)
318 		if ((maxfiles *= 26) <= 0)
319 			errx(EX_USAGE, "suffix is too long (max %ld)", i);
320 
321 	if (fnum == maxfiles)
322 		errx(EX_DATAERR, "too many files");
323 
324 	/* Generate suffix of sufflen letters */
325 	tfnum = fnum;
326 	i = sufflen - 1;
327 	do {
328 		fpnt[i] = tfnum % 26 + 'a';
329 		tfnum /= 26;
330 	} while (i-- > 0);
331 	fpnt[sufflen] = '\0';
332 
333 	++fnum;
334 	if (!freopen(fname, "w", stdout))
335 		err(EX_IOERR, "%s", fname);
336 	file_open = 1;
337 }
338 
339 static void
340 usage(void)
341 {
342 	(void)fprintf(stderr,
343 "usage: split [-a sufflen] [-b byte_count] [-l line_count] [-p pattern]\n");
344 	(void)fprintf(stderr,
345 "             [file [prefix]]\n");
346 	exit(EX_USAGE);
347 }
348