fixed bug with -d leaving seperators; From: Chris Torek <torek@BSDI.COM>
[unix-history] / usr / src / usr.bin / cut / cut.c
/*
* Copyright (c) 1989, 1993
* The Regents of the University of California. All rights reserved.
*
* This code is derived from software contributed to Berkeley by
* Adam S. Moskowitz of Menlo Consulting and Marciano Pitargue.
*
* %sccs.include.redist.c%
*/
#ifndef lint
static char copyright[] =
"@(#) Copyright (c) 1989, 1993\n\
The Regents of the University of California. All rights reserved.\n";
#endif /* not lint */
#ifndef lint
static char sccsid[] = "@(#)cut.c 8.2 (Berkeley) %G%";
#endif /* not lint */
#include <ctype.h>
#include <errno.h>
#include <limits.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
int cflag;
char dchar;
int dflag;
int fflag;
int sflag;
void c_cut __P((FILE *, char *));
void err __P((const char *, ...));
void f_cut __P((FILE *, char *));
void get_list __P((char *));
void usage __P((void));
int
main(argc, argv)
int argc;
char *argv[];
{
FILE *fp;
void (*fcn) __P((FILE *, char *));
int ch;
dchar = '\t'; /* default delimiter is \t */
while ((ch = getopt(argc, argv, "c:d:f:s")) != EOF)
switch(ch) {
case 'c':
fcn = c_cut;
get_list(optarg);
cflag = 1;
break;
case 'd':
dchar = *optarg;
dflag = 1;
break;
case 'f':
get_list(optarg);
fcn = f_cut;
fflag = 1;
break;
case 's':
sflag = 1;
break;
case '?':
default:
usage();
}
argc -= optind;
argv += optind;
if (fflag) {
if (cflag)
usage();
} else if (!cflag || dflag || sflag)
usage();
if (*argv)
for (; *argv; ++argv) {
if (!(fp = fopen(*argv, "r")))
err("%s: %s\n", *argv, strerror(errno));
fcn(fp, *argv);
(void)fclose(fp);
}
else
fcn(stdin, "stdin");
exit(0);
}
int autostart, autostop, maxval;
char positions[_POSIX2_LINE_MAX + 1];
void
get_list(list)
char *list;
{
register int setautostart, start, stop;
register char *pos;
char *p;
/*
* set a byte in the positions array to indicate if a field or
* column is to be selected; use +1, it's 1-based, not 0-based.
* This parser is less restrictive than the Draft 9 POSIX spec.
* POSIX doesn't allow lists that aren't in increasing order or
* overlapping lists. We also handle "-3-5" although there's no
* real reason too.
*/
for (; p = strtok(list, ", \t"); list = NULL) {
setautostart = start = stop = 0;
if (*p == '-') {
++p;
setautostart = 1;
}
if (isdigit(*p)) {
start = stop = strtol(p, &p, 10);
if (setautostart && start > autostart)
autostart = start;
}
if (*p == '-') {
if (isdigit(p[1]))
stop = strtol(p + 1, &p, 10);
if (*p == '-') {
++p;
if (!autostop || autostop > stop)
autostop = stop;
}
}
if (*p)
err("[-cf] list: illegal list value\n");
if (!stop || !start)
err("[-cf] list: values may not include zero\n");
if (stop > _POSIX2_LINE_MAX)
err("[-cf] list: %d too large (max %d)\n",
stop, _POSIX2_LINE_MAX);
if (maxval < stop)
maxval = stop;
for (pos = positions + start; start++ <= stop; *pos++ = 1);
}
/* overlapping ranges */
if (autostop && maxval > autostop)
maxval = autostop;
/* set autostart */
if (autostart)
memset(positions + 1, '1', autostart);
}
/* ARGSUSED */
void
c_cut(fp, fname)
FILE *fp;
char *fname;
{
register int ch, col;
register char *pos;
for (;;) {
pos = positions + 1;
for (col = maxval; col; --col) {
if ((ch = getc(fp)) == EOF)
return;
if (ch == '\n')
break;
if (*pos++)
(void)putchar(ch);
}
if (ch != '\n')
if (autostop)
while ((ch = getc(fp)) != EOF && ch != '\n')
(void)putchar(ch);
else
while ((ch = getc(fp)) != EOF && ch != '\n');
(void)putchar('\n');
}
}
void
f_cut(fp, fname)
FILE *fp;
char *fname;
{
register int ch, field, isdelim;
register char *pos, *p, sep;
int output;
char lbuf[_POSIX2_LINE_MAX + 1];
for (sep = dchar; fgets(lbuf, sizeof(lbuf), fp);) {
output = 0;
for (isdelim = 0, p = lbuf;; ++p) {
if (!(ch = *p))
err("%s: line too long.\n", fname);
/* this should work if newline is delimiter */
if (ch == sep)
isdelim = 1;
if (ch == '\n') {
if (!isdelim && !sflag)
(void)printf("%s", lbuf);
break;
}
}
if (!isdelim)
continue;
pos = positions + 1;
for (field = maxval, p = lbuf; field; --field, ++pos) {
if (*pos) {
if (output++)
(void)putchar(sep);
while ((ch = *p++) != '\n' && ch != sep)
(void)putchar(ch);
} else
while ((ch = *p++) != '\n' && ch != sep);
if (ch == '\n')
break;
}
if (ch != '\n')
if (autostop) {
if (output)
(void)putchar(sep);
for (; (ch = *p) != '\n'; ++p)
(void)putchar(ch);
} else
for (; (ch = *p) != '\n'; ++p);
(void)putchar('\n');
}
}
void
usage()
{
(void)fprintf(stderr,
"usage:\tcut -c list [file1 ...]\n\tcut -f list [-s] [-d delim] [file ...]\n");
exit(1);
}
#if __STDC__
#include <stdarg.h>
#else
#include <varargs.h>
#endif
void
#if __STDC__
err(const char *fmt, ...)
#else
err(fmt, va_alist)
char *fmt;
va_dcl
#endif
{
va_list ap;
#if __STDC__
va_start(ap, fmt);
#else
va_start(ap);
#endif
(void)fprintf(stderr, "cut: ");
(void)vfprintf(stderr, fmt, ap);
va_end(ap);
(void)fprintf(stderr, "\n");
exit(1);
/* NOTREACHED */
}