freebsd-skq/usr.bin/split/split.c
Pedro F. Giffuni 8a16b7a18f General further adoption of SPDX licensing ID tags.
Mainly focus on files that use BSD 3-Clause license.

The Software Package Data Exchange (SPDX) group provides a specification
to make it easier for automated tools to detect and summarize well known
opensource licenses. We are gradually adopting the specification, noting
that the tags are considered only advisory and do not, in any way,
superceed or replace the license texts.

Special thanks to Wind River for providing access to "The Duke of
Highlander" tool: an older (2014) run over FreeBSD tree was useful as a
starting point.
2017-11-20 19:49:47 +00:00

405 lines
9.3 KiB
C

/*-
* SPDX-License-Identifier: BSD-3-Clause
*
* Copyright (c) 1987, 1993, 1994
* The Regents of the University of California. All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
* 3. Neither the name of the University nor the names of its contributors
* may be used to endorse or promote products derived from this software
* without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*/
#include <sys/cdefs.h>
__FBSDID("$FreeBSD$");
#ifndef lint
static const char copyright[] =
"@(#) Copyright (c) 1987, 1993, 1994\n\
The Regents of the University of California. All rights reserved.\n";
#endif
#ifndef lint
static const char sccsid[] = "@(#)split.c 8.2 (Berkeley) 4/16/94";
#endif
#include <sys/param.h>
#include <sys/types.h>
#include <sys/stat.h>
#include <ctype.h>
#include <err.h>
#include <errno.h>
#include <fcntl.h>
#include <inttypes.h>
#include <libutil.h>
#include <limits.h>
#include <locale.h>
#include <stdbool.h>
#include <stdint.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <unistd.h>
#include <regex.h>
#include <sysexits.h>
#define DEFLINE 1000 /* Default num lines per file. */
static off_t bytecnt; /* Byte count to split on. */
static off_t chunks = 0; /* Chunks count to split into. */
static long numlines; /* Line count to split on. */
static int file_open; /* If a file open. */
static int ifd = -1, ofd = -1; /* Input/output file descriptors. */
static char bfr[MAXBSIZE]; /* I/O buffer. */
static char fname[MAXPATHLEN]; /* File name prefix. */
static regex_t rgx;
static int pflag;
static bool dflag;
static long sufflen = 2; /* File name suffix length. */
static void newfile(void);
static void split1(void);
static void split2(void);
static void split3(void);
static void usage(void);
int
main(int argc, char **argv)
{
int ch;
int error;
char *ep, *p;
setlocale(LC_ALL, "");
dflag = false;
while ((ch = getopt(argc, argv, "0123456789a:b:dl:n:p:")) != -1)
switch (ch) {
case '0': case '1': case '2': case '3': case '4':
case '5': case '6': case '7': case '8': case '9':
/*
* Undocumented kludge: split was originally designed
* to take a number after a dash.
*/
if (numlines == 0) {
p = argv[optind - 1];
if (p[0] == '-' && p[1] == ch && !p[2])
numlines = strtol(++p, &ep, 10);
else
numlines =
strtol(argv[optind] + 1, &ep, 10);
if (numlines <= 0 || *ep)
errx(EX_USAGE,
"%s: illegal line count", optarg);
}
break;
case 'a': /* Suffix length */
if ((sufflen = strtol(optarg, &ep, 10)) <= 0 || *ep)
errx(EX_USAGE,
"%s: illegal suffix length", optarg);
break;
case 'b': /* Byte count. */
errno = 0;
error = expand_number(optarg, &bytecnt);
if (error == -1)
errx(EX_USAGE, "%s: offset too large", optarg);
break;
case 'd': /* Decimal suffix */
dflag = true;
break;
case 'l': /* Line count. */
if (numlines != 0)
usage();
if ((numlines = strtol(optarg, &ep, 10)) <= 0 || *ep)
errx(EX_USAGE,
"%s: illegal line count", optarg);
break;
case 'n': /* Chunks. */
if (!isdigit((unsigned char)optarg[0]) ||
(chunks = (size_t)strtoul(optarg, &ep, 10)) == 0 ||
*ep != '\0') {
errx(EX_USAGE, "%s: illegal number of chunks",
optarg);
}
break;
case 'p': /* pattern matching. */
if (regcomp(&rgx, optarg, REG_EXTENDED|REG_NOSUB) != 0)
errx(EX_USAGE, "%s: illegal regexp", optarg);
pflag = 1;
break;
default:
usage();
}
argv += optind;
argc -= optind;
if (*argv != NULL) { /* Input file. */
if (strcmp(*argv, "-") == 0)
ifd = STDIN_FILENO;
else if ((ifd = open(*argv, O_RDONLY, 0)) < 0)
err(EX_NOINPUT, "%s", *argv);
++argv;
}
if (*argv != NULL) /* File name prefix. */
if (strlcpy(fname, *argv++, sizeof(fname)) >= sizeof(fname))
errx(EX_USAGE, "file name prefix is too long");
if (*argv != NULL)
usage();
if (strlen(fname) + (unsigned long)sufflen >= sizeof(fname))
errx(EX_USAGE, "suffix is too long");
if (pflag && (numlines != 0 || bytecnt != 0 || chunks != 0))
usage();
if (numlines == 0)
numlines = DEFLINE;
else if (bytecnt != 0 || chunks != 0)
usage();
if (bytecnt && chunks)
usage();
if (ifd == -1) /* Stdin by default. */
ifd = 0;
if (bytecnt) {
split1();
exit (0);
} else if (chunks) {
split3();
exit (0);
}
split2();
if (pflag)
regfree(&rgx);
exit(0);
}
/*
* split1 --
* Split the input by bytes.
*/
static void
split1(void)
{
off_t bcnt;
char *C;
ssize_t dist, len;
int nfiles;
nfiles = 0;
for (bcnt = 0;;)
switch ((len = read(ifd, bfr, MAXBSIZE))) {
case 0:
exit(0);
case -1:
err(EX_IOERR, "read");
/* NOTREACHED */
default:
if (!file_open) {
if (!chunks || (nfiles < chunks)) {
newfile();
nfiles++;
}
}
if (bcnt + len >= bytecnt) {
dist = bytecnt - bcnt;
if (write(ofd, bfr, dist) != dist)
err(EX_IOERR, "write");
len -= dist;
for (C = bfr + dist; len >= bytecnt;
len -= bytecnt, C += bytecnt) {
if (!chunks || (nfiles < chunks)) {
newfile();
nfiles++;
}
if (write(ofd,
C, bytecnt) != bytecnt)
err(EX_IOERR, "write");
}
if (len != 0) {
if (!chunks || (nfiles < chunks)) {
newfile();
nfiles++;
}
if (write(ofd, C, len) != len)
err(EX_IOERR, "write");
} else
file_open = 0;
bcnt = len;
} else {
bcnt += len;
if (write(ofd, bfr, len) != len)
err(EX_IOERR, "write");
}
}
}
/*
* split2 --
* Split the input by lines.
*/
static void
split2(void)
{
long lcnt = 0;
FILE *infp;
/* Stick a stream on top of input file descriptor */
if ((infp = fdopen(ifd, "r")) == NULL)
err(EX_NOINPUT, "fdopen");
/* Process input one line at a time */
while (fgets(bfr, sizeof(bfr), infp) != NULL) {
const int len = strlen(bfr);
/* If line is too long to deal with, just write it out */
if (bfr[len - 1] != '\n')
goto writeit;
/* Check if we need to start a new file */
if (pflag) {
regmatch_t pmatch;
pmatch.rm_so = 0;
pmatch.rm_eo = len - 1;
if (regexec(&rgx, bfr, 0, &pmatch, REG_STARTEND) == 0)
newfile();
} else if (lcnt++ == numlines) {
newfile();
lcnt = 1;
}
writeit:
/* Open output file if needed */
if (!file_open)
newfile();
/* Write out line */
if (write(ofd, bfr, len) != len)
err(EX_IOERR, "write");
}
/* EOF or error? */
if (ferror(infp))
err(EX_IOERR, "read");
else
exit(0);
}
/*
* split3 --
* Split the input into specified number of chunks
*/
static void
split3(void)
{
struct stat sb;
if (fstat(ifd, &sb) == -1) {
err(1, "stat");
/* NOTREACHED */
}
if (chunks > sb.st_size) {
errx(1, "can't split into more than %d files",
(int)sb.st_size);
/* NOTREACHED */
}
bytecnt = sb.st_size / chunks;
split1();
}
/*
* newfile --
* Open a new output file.
*/
static void
newfile(void)
{
long i, maxfiles, tfnum;
static long fnum;
static char *fpnt;
char beg, end;
int pattlen;
if (ofd == -1) {
if (fname[0] == '\0') {
fname[0] = 'x';
fpnt = fname + 1;
} else {
fpnt = fname + strlen(fname);
}
ofd = fileno(stdout);
}
if (dflag) {
beg = '0';
end = '9';
}
else {
beg = 'a';
end = 'z';
}
pattlen = end - beg + 1;
/* maxfiles = pattlen^sufflen, but don't use libm. */
for (maxfiles = 1, i = 0; i < sufflen; i++)
if (LONG_MAX / pattlen < maxfiles)
errx(EX_USAGE, "suffix is too long (max %ld)", i);
else
maxfiles *= pattlen;
if (fnum == maxfiles)
errx(EX_DATAERR, "too many files");
/* Generate suffix of sufflen letters */
tfnum = fnum;
i = sufflen - 1;
do {
fpnt[i] = tfnum % pattlen + beg;
tfnum /= pattlen;
} while (i-- > 0);
fpnt[sufflen] = '\0';
++fnum;
if (!freopen(fname, "w", stdout))
err(EX_IOERR, "%s", fname);
file_open = 1;
}
static void
usage(void)
{
(void)fprintf(stderr,
"usage: split [-l line_count] [-a suffix_length] [file [prefix]]\n"
" split -b byte_count[K|k|M|m|G|g] [-a suffix_length] [file [prefix]]\n"
" split -n chunk_count [-a suffix_length] [file [prefix]]\n"
" split -p pattern [-a suffix_length] [file [prefix]]\n");
exit(EX_USAGE);
}