2 * SPDX-License-Identifier: BSD-3-Clause
4 * Copyright (c) 1987, 1993, 1994
5 * The Regents of the University of California. All rights reserved.
7 * Redistribution and use in source and binary forms, with or without
8 * modification, are permitted provided that the following conditions
10 * 1. Redistributions of source code must retain the above copyright
11 * notice, this list of conditions and the following disclaimer.
12 * 2. Redistributions in binary form must reproduce the above copyright
13 * notice, this list of conditions and the following disclaimer in the
14 * documentation and/or other materials provided with the distribution.
15 * 3. Neither the name of the University nor the names of its contributors
16 * may be used to endorse or promote products derived from this software
17 * without specific prior written permission.
19 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
20 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
21 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
22 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
23 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
24 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
25 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
26 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
27 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
28 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
32 #include <sys/cdefs.h>
34 static const char copyright[] =
35 "@(#) Copyright (c) 1987, 1993, 1994\n\
36 The Regents of the University of California. All rights reserved.\n";
40 static const char sccsid[] = "@(#)split.c 8.2 (Berkeley) 4/16/94";
43 #include <sys/param.h>
44 #include <sys/types.h>
64 #define DEFLINE 1000 /* Default num lines per file. */
66 static off_t bytecnt; /* Byte count to split on. */
67 static off_t chunks = 0; /* Chunks count to split into. */
68 static bool clobber = true; /* Whether to overwrite existing output files. */
69 static long numlines; /* Line count to split on. */
70 static int file_open; /* If a file open. */
71 static int ifd = -1, ofd = -1; /* Input/output file descriptors. */
72 static char fname[MAXPATHLEN]; /* File name prefix. */
76 static long sufflen = 2; /* File name suffix length. */
77 static int autosfx = 1; /* Whether to auto-extend the suffix length. */
79 static void newfile(void);
80 static void split1(void);
81 static void split2(void);
82 static void split3(void);
83 static void usage(void) __dead2;
86 main(int argc, char **argv)
92 setlocale(LC_ALL, "");
95 while ((ch = getopt(argc, argv, "0123456789a:b:cdl:n:p:")) != -1)
97 case '0': case '1': case '2': case '3': case '4':
98 case '5': case '6': case '7': case '8': case '9':
100 * Undocumented kludge: split was originally designed
101 * to take a number after a dash.
104 p = argv[optind - 1];
105 if (p[0] == '-' && p[1] == ch && !p[2])
106 numlines = strtol(++p, &ep, 10);
109 strtol(argv[optind] + 1, &ep, 10);
110 if (numlines <= 0 || *ep)
112 "%s: illegal line count", optarg);
115 case 'a': /* Suffix length */
116 if ((sufflen = strtol(optarg, &ep, 10)) <= 0 || *ep)
118 "%s: illegal suffix length", optarg);
121 case 'b': /* Byte count. */
123 error = expand_number(optarg, &bytecnt);
125 errx(EX_USAGE, "%s: offset too large", optarg);
127 case 'c': /* Continue, don't overwrite output files. */
130 case 'd': /* Decimal suffix */
133 case 'l': /* Line count. */
136 if ((numlines = strtol(optarg, &ep, 10)) <= 0 || *ep)
138 "%s: illegal line count", optarg);
140 case 'n': /* Chunks. */
141 if (!isdigit((unsigned char)optarg[0]) ||
142 (chunks = (size_t)strtoul(optarg, &ep, 10)) == 0 ||
144 errx(EX_USAGE, "%s: illegal number of chunks",
149 case 'p': /* pattern matching. */
150 if (regcomp(&rgx, optarg, REG_EXTENDED|REG_NOSUB) != 0)
151 errx(EX_USAGE, "%s: illegal regexp", optarg);
160 if (*argv != NULL) { /* Input file. */
161 if (strcmp(*argv, "-") == 0)
163 else if ((ifd = open(*argv, O_RDONLY, 0)) < 0)
164 err(EX_NOINPUT, "%s", *argv);
167 if (*argv != NULL) /* File name prefix. */
168 if (strlcpy(fname, *argv++, sizeof(fname)) >= sizeof(fname))
169 errx(EX_USAGE, "file name prefix is too long");
173 if (strlen(fname) + (unsigned long)sufflen >= sizeof(fname))
174 errx(EX_USAGE, "suffix is too long");
175 if (pflag && (numlines != 0 || bytecnt != 0 || chunks != 0))
180 else if (bytecnt != 0 || chunks != 0)
183 if (bytecnt && chunks)
186 if (ifd == -1) /* Stdin by default. */
204 * Split the input by bytes.
209 static char bfr[MAXBSIZE];
218 switch ((len = read(ifd, bfr, sizeof(bfr)))) {
222 err(EX_IOERR, "read");
226 if (!chunks || (nfiles < chunks)) {
231 if (bcnt + len >= bytecnt) {
232 dist = bytecnt - bcnt;
233 if (write(ofd, bfr, dist) != dist)
234 err(EX_IOERR, "write");
236 for (C = bfr + dist; len >= bytecnt;
237 len -= bytecnt, C += bytecnt) {
238 if (!chunks || (nfiles < chunks)) {
243 C, bytecnt) != bytecnt)
244 err(EX_IOERR, "write");
247 if (!chunks || (nfiles < chunks)) {
251 if (write(ofd, C, len) != len)
252 err(EX_IOERR, "write");
258 if (write(ofd, bfr, len) != len)
259 err(EX_IOERR, "write");
266 * Split the input by lines.
280 /* Stick a stream on top of input file descriptor */
281 if ((infp = fdopen(ifd, "r")) == NULL)
282 err(EX_NOINPUT, "fdopen");
284 /* Process input one line at a time */
285 while ((errno = 0, len = getline(&buf, &bufsize, infp)) > 0) {
286 /* Check if we need to start a new file */
291 pmatch.rm_eo = len - 1;
292 if (regexec(&rgx, buf, 0, &pmatch, REG_STARTEND) == 0)
294 } else if (lcnt++ == numlines) {
299 /* Open output file if needed */
304 if (write(ofd, buf, len) != len)
305 err(EX_IOERR, "write");
309 if ((len == -1 && errno != 0) || ferror(infp))
310 err(EX_IOERR, "read");
317 * Split the input into specified number of chunks
324 if (fstat(ifd, &sb) == -1) {
329 if (chunks > sb.st_size) {
330 errx(1, "can't split into more than %d files",
335 bytecnt = sb.st_size / chunks;
342 * Open a new output file.
347 long i, maxfiles, tfnum;
352 int flags = O_WRONLY | O_CREAT | O_TRUNC;
358 if (fname[0] == '\0') {
362 fpnt = fname + strlen(fname);
364 } else if (close(ofd) != 0)
376 pattlen = end - beg + 1;
379 * If '-a' is not specified, then we automatically expand the
380 * suffix length to accomodate splitting all input. We do this
381 * by moving the suffix pointer (fpnt) forward and incrementing
382 * sufflen by one, thereby yielding an additional two characters
383 * and allowing all output files to sort such that 'cat *' yields
384 * the input in order. I.e., the order is '... xyy xyz xzaaa
385 * xzaab ... xzyzy, xzyzz, xzzaaaa, xzzaaab' and so on.
387 if (!dflag && autosfx && (fpnt[0] == 'y') &&
388 strspn(fpnt+1, "z") == strlen(fpnt+1)) {
389 fpnt = fname + strlen(fname) - sufflen;
390 fpnt[sufflen + 2] = '\0';
402 /* Reset so we start back at all 'a's in our extended suffix. */
407 /* maxfiles = pattlen^sufflen, but don't use libm. */
408 for (maxfiles = 1, i = 0; i < sufflen; i++)
409 if (LONG_MAX / pattlen < maxfiles)
410 errx(EX_USAGE, "suffix is too long (max %ld)", i);
414 if (fnum == maxfiles)
415 errx(EX_DATAERR, "too many files");
417 /* Generate suffix of sufflen letters */
421 fpnt[i] = tfnum % pattlen + beg;
424 fpnt[sufflen] = '\0';
427 if ((ofd = open(fname, flags, DEFFILEMODE)) < 0) {
428 if (!clobber && errno == EEXIST)
430 err(EX_IOERR, "%s", fname);
438 (void)fprintf(stderr,
439 "usage: split [-cd] [-l line_count] [-a suffix_length] [file [prefix]]\n"
440 " split [-cd] -b byte_count[K|k|M|m|G|g] [-a suffix_length] [file [prefix]]\n"
441 " split [-cd] -n chunk_count [-a suffix_length] [file [prefix]]\n"
442 " split [-cd] -p pattern [-a suffix_length] [file [prefix]]\n");