xref: /dragonfly/usr.bin/split/split.c (revision 3170ffd7)
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  * 4. Neither the name of the University nor the names of its contributors
14  *    may be used to endorse or promote products derived from this software
15  *    without specific prior written permission.
16  *
17  * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
18  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
19  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
20  * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
21  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
22  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
23  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
24  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
25  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
26  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
27  * SUCH DAMAGE.
28  *
29  * @(#) Copyright (c) 1987, 1993, 1994 The Regents of the University of California.  All rights reserved.
30  * @(#)split.c	8.2 (Berkeley) 4/16/94
31  * $FreeBSD: src/usr.bin/split/split.c,v 1.6.2.2 2002/07/25 12:46:36 tjr Exp $
32  */
33 
34 #include <sys/param.h>
35 
36 #include <ctype.h>
37 #include <err.h>
38 #include <errno.h>
39 #include <fcntl.h>
40 #include <limits.h>
41 #include <stdio.h>
42 #include <stdlib.h>
43 #include <string.h>
44 #include <unistd.h>
45 #include <regex.h>
46 #include <sysexits.h>
47 
48 #define DEFLINE	1000			/* Default num lines per file. */
49 
50 off_t	 bytecnt;			/* Byte count to split on. */
51 long	 numlines;			/* Line count to split on. */
52 int	 file_open;			/* If a file open. */
53 int	 ifd = -1, ofd = -1;		/* Input/output file descriptors. */
54 char	 *bfr;				/* I/O buffer. */
55 char	 fname[MAXPATHLEN];		/* File name prefix. */
56 regex_t	 rgx;
57 int	 pflag;
58 long	 sufflen = 2;			/* File name suffix length. */
59 
60 void newfile(void);
61 void split1(void);
62 void split2(void);
63 static void usage(void);
64 
65 int
66 main(int argc, char **argv)
67 {
68 	long long bytecnti;
69 	long scale;
70 	int ch;
71 	char *ep, *p;
72 
73 	while ((ch = getopt(argc, argv, "0123456789a:b:l:p:")) != -1)
74 		switch (ch) {
75 		case '0': case '1': case '2': case '3': case '4':
76 		case '5': case '6': case '7': case '8': case '9':
77 			/*
78 			 * Undocumented kludge: split was originally designed
79 			 * to take a number after a dash.
80 			 */
81 			if (numlines == 0) {
82 				p = argv[optind - 1];
83 				if (p[0] == '-' && p[1] == ch && !p[2])
84 					numlines = strtol(++p, &ep, 10);
85 				else
86 					numlines =
87 					    strtol(argv[optind] + 1, &ep, 10);
88 				if (numlines <= 0 || *ep)
89 					errx(EX_USAGE,
90 					    "%s: illegal line count", optarg);
91 			}
92 			break;
93 		case 'a':		/* Suffix length */
94 			if ((sufflen = strtol(optarg, &ep, 10)) <= 0 || *ep)
95 				errx(EX_USAGE,
96 				    "%s: illegal suffix length", optarg);
97 			break;
98 		case 'b':		/* Byte count. */
99 			errno = 0;
100 			if ((bytecnti = strtoll(optarg, &ep, 10)) <= 0 ||
101 			    (*ep != '\0' && *ep != 'k' && *ep != 'm') ||
102 			    errno != 0)
103 				errx(EX_USAGE,
104 				    "%s: illegal byte count", optarg);
105 			if (*ep == 'k')
106 				scale = 1024;
107 			else if (*ep == 'm')
108 				scale = 1024 * 1024;
109 			else
110 				scale = 1;
111 			bytecnt = (off_t)(bytecnti * scale);
112 			break;
113 		case 'p' :      /* pattern matching. */
114 			if (regcomp(&rgx, optarg, REG_EXTENDED|REG_NOSUB) != 0)
115 				errx(EX_USAGE, "%s: illegal regexp", optarg);
116 			pflag = 1;
117 			break;
118 		case 'l':		/* Line count. */
119 			if (numlines != 0)
120 				usage();
121 			if ((numlines = strtol(optarg, &ep, 10)) <= 0 || *ep)
122 				errx(EX_USAGE,
123 				    "%s: illegal line count", optarg);
124 			break;
125 		default:
126 			usage();
127 		}
128 	argv += optind;
129 	argc -= optind;
130 
131 	if (*argv != NULL) {
132 		if (strcmp(*argv, "-") == 0)
133 			ifd = STDIN_FILENO;
134 		else if ((ifd = open(*argv, O_RDONLY, 0)) < 0)
135 			err(EX_NOINPUT, "%s", *argv);
136 		++argv;
137 	}
138 
139 	if (*argv != NULL)			/* File name prefix. */
140 		if (strlcpy(fname, *argv++, sizeof(fname)) >= sizeof(fname))
141 			errx(EX_USAGE, "file name prefix is too long");
142 	if (*argv != NULL)
143 		usage();
144 
145 	if (strlen(fname) + (unsigned long)sufflen >= sizeof(fname))
146 		errx(EX_USAGE, "suffix is too long");
147 	if (pflag && (numlines != 0 || bytecnt != 0))
148 		usage();
149 
150 	if (numlines == 0)
151 		numlines = DEFLINE;
152 	else if (bytecnt != 0)
153 		usage();
154 
155 	if (ifd == -1)				/* Stdin by default. */
156 		ifd = 0;
157 
158 	if (bytecnt) {
159 		split1();
160 		exit (0);
161 	}
162 	split2();
163 	if (pflag)
164 		regfree(&rgx);
165 	exit(0);
166 }
167 
168 /*
169  * split1 --
170  *	Split the input by bytes.
171  */
172 void
173 split1(void)
174 {
175 	off_t bcnt;
176 	char *C;
177 	ssize_t dist, len;
178 
179 	if((bfr = (char *)malloc(bytecnt)) == NULL)
180 		err(EX_OSERR, "malloc");
181 
182 	for (bcnt = 0;;)
183 		switch ((len = read(ifd, bfr, bytecnt))) {
184 		case 0:
185 			free(bfr);
186 			exit(0);
187 		case -1:
188 			free(bfr);
189 			err(EX_IOERR, "read");
190 			/* NOTREACHED */
191 		default:
192 			if (!file_open)
193 				newfile();
194 			if (bcnt + len >= bytecnt) {
195 				dist = bytecnt - bcnt;
196 				if (write(ofd, bfr, dist) != dist)
197 					err(EX_IOERR, "write");
198 				len -= dist;
199 				for (C = bfr + dist; len >= bytecnt;
200 				    len -= bytecnt, C += bytecnt) {
201 					newfile();
202 					if (write(ofd,
203 					    C, (int)bytecnt) != bytecnt) {
204 						free(bfr);
205 						err(EX_IOERR, "write");
206 					}
207 				}
208 				if (len != 0) {
209 					newfile();
210 					if (write(ofd, C, len) != len) {
211 						free(bfr);
212 						err(EX_IOERR, "write");
213 					}
214 				} else
215 					file_open = 0;
216 				bcnt = len;
217 			} else {
218 				bcnt += len;
219 				if (write(ofd, bfr, len) != len) {
220 					free(bfr);
221 					err(EX_IOERR, "write");
222 				}
223 			}
224 		}
225 	free(bfr);
226 }
227 
228 /*
229  * split2 --
230  *	Split the input by lines.
231  */
232 void
233 split2(void)
234 {
235 	int startofline = 1;
236 	long lcnt = 0;
237 	FILE *infp;
238 
239 	/* Stick a stream on top of input file descriptor */
240 	if ((infp = fdopen(ifd, "r")) == NULL)
241 		err(EX_NOINPUT, "fdopen");
242 
243 	if((bfr = (char *)malloc(MAXBSIZE)) == NULL)
244 		err(EX_OSERR, "malloc");
245 
246 	/* Process input one line at a time */
247 	while (fgets(bfr, MAXBSIZE, infp) != NULL) {
248 		const int len = strlen(bfr);
249 
250 		/* Consider starting a new file only when at beginning of a line */
251 		if (startofline) {
252 			/* Check if we need to start a new file */
253 			if (pflag) {
254 				regmatch_t pmatch;
255 
256 				pmatch.rm_so = 0;
257 				pmatch.rm_eo = len - 1;
258 				if (regexec(&rgx, bfr, 0, &pmatch, REG_STARTEND) == 0)
259 					newfile();
260 			} else if (lcnt++ == numlines) {
261 				newfile();
262 				lcnt = 1;
263 			}
264 		}
265 
266 		if (bfr[len - 1] != '\n')
267 			startofline = 0;
268 		else
269 			startofline = 1;
270 
271 		/* Open output file if needed */
272 		if (!file_open)
273 			newfile();
274 
275 		/* Write out line */
276 		if (write(ofd, bfr, len) != len) {
277 			free(bfr);
278 			err(EX_IOERR, "write");
279 		}
280 	}
281 
282 	free(bfr);
283 
284 	/* EOF or error? */
285 	if (ferror(infp))
286 		err(EX_IOERR, "read");
287 	else
288 		exit(0);
289 }
290 
291 /*
292  * newfile --
293  *	Open a new output file.
294  */
295 void
296 newfile(void)
297 {
298 	long i, maxfiles, tfnum;
299 	static long fnum;
300 	static char *fpnt;
301 
302 	if (ofd == -1) {
303 		if (fname[0] == '\0') {
304 			fname[0] = 'x';
305 			fpnt = fname + 1;
306 		} else {
307 			fpnt = fname + strlen(fname);
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 [-a sufflen] [-b byte_count] [-l line_count] [-p pattern]\n");
340 	(void)fprintf(stderr,
341 "             [file [prefix]]\n");
342 	exit(EX_USAGE);
343 }
344