busybox/coreutils/wc.c

215 lines
5.2 KiB
C

/* vi: set sw=4 ts=4: */
/*
* wc implementation for busybox
*
* Copyright (C) 2003 Manuel Novoa III <mjn3@codepoet.org>
*
* Licensed under GPLv2 or later, see file LICENSE in this tarball for details.
*/
/* BB_AUDIT SUSv3 _NOT_ compliant -- option -m is not currently supported. */
/* http://www.opengroup.org/onlinepubs/007904975/utilities/wc.html */
/* Mar 16, 2003 Manuel Novoa III (mjn3@codepoet.org)
*
* Rewritten to fix a number of problems and do some size optimizations.
* Problems in the previous busybox implementation (besides bloat) included:
* 1) broken 'wc -c' optimization (read note below)
* 2) broken handling of '-' args
* 3) no checking of ferror on EOF returns
* 4) isprint() wasn't considered when word counting.
*
* TODO:
*
* When locale support is enabled, count multibyte chars in the '-m' case.
*
* NOTES:
*
* The previous busybox wc attempted an optimization using stat for the
* case of counting chars only. I omitted that because it was broken.
* It didn't take into account the possibility of input coming from a
* pipe, or input from a file with file pointer not at the beginning.
*
* To implement such a speed optimization correctly, not only do you
* need the size, but also the file position. Note also that the
* file position may be past the end of file. Consider the example
* (adapted from example in gnu wc.c)
*
* echo hello > /tmp/testfile &&
* (dd ibs=1k skip=1 count=0 &> /dev/null ; wc -c) < /tmp/testfile
*
* for which 'wc -c' should output '0'.
*/
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <unistd.h>
#include "busybox.h"
#ifdef CONFIG_LOCALE_SUPPORT
#include <locale.h>
#include <ctype.h>
#define isspace_given_isprint(c) isspace(c)
#else
#undef isspace
#undef isprint
#define isspace(c) ((((c) == ' ') || (((unsigned int)((c) - 9)) <= (13 - 9))))
#define isprint(c) (((unsigned int)((c) - 0x20)) <= (0x7e - 0x20))
#define isspace_given_isprint(c) ((c) == ' ')
#endif
enum {
WC_LINES = 0,
WC_WORDS = 1,
WC_CHARS = 2,
WC_LENGTH = 3
};
/* Note: If this changes, remember to change the initialization of
* 'name' in wc_main. It needs to point to the terminating nul. */
static const char wc_opts[] = "lwcL"; /* READ THE WARNING ABOVE! */
enum {
OP_INC_LINE = 1, /* OP_INC_LINE must be 1. */
OP_SPACE = 2,
OP_NEWLINE = 4,
OP_TAB = 8,
OP_NUL = 16,
};
/* Note: If fmt_str changes, the offsets to 's' in the OUTPUT section
* will need to be updated. */
static const char fmt_str[] = " %7u\0 %s\n";
static const char total_str[] = "total";
int wc_main(int argc, char **argv)
{
FILE *fp;
const char *s;
unsigned int *pcounts;
unsigned int counts[4];
unsigned int totals[4];
unsigned int linepos;
unsigned int u;
int num_files = 0;
int c;
char status = EXIT_SUCCESS;
char in_word;
char print_type;
print_type = bb_getopt_ulflags(argc, argv, wc_opts);
if (print_type == 0) {
print_type = (1 << WC_LINES) | (1 << WC_WORDS) | (1 << WC_CHARS);
}
argv += optind;
if (!*argv) {
*--argv = (char *) bb_msg_standard_input;
}
memset(totals, 0, sizeof(totals));
pcounts = counts;
do {
++num_files;
if (!(fp = bb_wfopen_input(*argv))) {
status = EXIT_FAILURE;
continue;
}
memset(counts, 0, sizeof(counts));
linepos = 0;
in_word = 0;
do {
++counts[WC_CHARS];
c = getc(fp);
if (isprint(c)) {
++linepos;
if (!isspace_given_isprint(c)) {
in_word = 1;
continue;
}
} else if (((unsigned int)(c - 9)) <= 4) {
/* \t 9
* \n 10
* \v 11
* \f 12
* \r 13
*/
if (c == '\t') {
linepos = (linepos | 7) + 1;
} else { /* '\n', '\r', '\f', or '\v' */
DO_EOF:
if (linepos > counts[WC_LENGTH]) {
counts[WC_LENGTH] = linepos;
}
if (c == '\n') {
++counts[WC_LINES];
}
if (c != '\v') {
linepos = 0;
}
}
} else if (c == EOF) {
if (ferror(fp)) {
bb_perror_msg("%s", *argv);
status = EXIT_FAILURE;
}
--counts[WC_CHARS];
goto DO_EOF; /* Treat an EOF as '\r'. */
} else {
continue;
}
counts[WC_WORDS] += in_word;
in_word = 0;
if (c == EOF) {
break;
}
} while (1);
if (totals[WC_LENGTH] < counts[WC_LENGTH]) {
totals[WC_LENGTH] = counts[WC_LENGTH];
}
totals[WC_LENGTH] -= counts[WC_LENGTH];
bb_fclose_nonstdin(fp);
OUTPUT:
s = fmt_str + 1; /* Skip the leading space on 1st pass. */
u = 0;
do {
if (print_type & (1 << u)) {
bb_printf(s, pcounts[u]);
s = fmt_str; /* Ok... restore the leading space. */
}
totals[u] += pcounts[u];
} while (++u < 4);
s += 8; /* Set the format to the empty string. */
if (*argv != bb_msg_standard_input) {
s -= 3; /* We have a name, so do %s conversion. */
}
bb_printf(s, *argv);
} while (*++argv);
/* If more than one file was processed, we want the totals. To save some
* space, we set the pcounts ptr to the totals array. This has the side
* effect of trashing the totals array after outputting it, but that's
* irrelavent since we no longer need it. */
if (num_files > 1) {
num_files = 0; /* Make sure we don't get here again. */
*--argv = (char *) total_str;
pcounts = totals;
goto OUTPUT;
}
bb_fflush_stdout_and_exit(status);
}