[821] | 1 | /* vi: set sw=4 ts=4: */
|
---|
| 2 | /*
|
---|
| 3 | * wc implementation for busybox
|
---|
| 4 | *
|
---|
| 5 | * Copyright (C) 2003 Manuel Novoa III <mjn3@codepoet.org>
|
---|
| 6 | *
|
---|
| 7 | * This program is free software; you can redistribute it and/or modify
|
---|
| 8 | * it under the terms of the GNU General Public License as published by
|
---|
| 9 | * the Free Software Foundation; either version 2 of the License, or
|
---|
| 10 | * (at your option) any later version.
|
---|
| 11 | *
|
---|
| 12 | * This program is distributed in the hope that it will be useful,
|
---|
| 13 | * but WITHOUT ANY WARRANTY; without even the implied warranty of
|
---|
| 14 | * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
|
---|
| 15 | * General Public License for more details.
|
---|
| 16 | *
|
---|
| 17 | * You should have received a copy of the GNU General Public License
|
---|
| 18 | * along with this program; if not, write to the Free Software
|
---|
| 19 | * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
|
---|
| 20 | *
|
---|
| 21 | */
|
---|
| 22 |
|
---|
| 23 | /* BB_AUDIT SUSv3 _NOT_ compliant -- option -m is not currently supported. */
|
---|
| 24 | /* http://www.opengroup.org/onlinepubs/007904975/utilities/wc.html */
|
---|
| 25 |
|
---|
| 26 | /* Mar 16, 2003 Manuel Novoa III (mjn3@codepoet.org)
|
---|
| 27 | *
|
---|
| 28 | * Rewritten to fix a number of problems and do some size optimizations.
|
---|
| 29 | * Problems in the previous busybox implementation (besides bloat) included:
|
---|
| 30 | * 1) broken 'wc -c' optimization (read note below)
|
---|
| 31 | * 2) broken handling of '-' args
|
---|
| 32 | * 3) no checking of ferror on EOF returns
|
---|
| 33 | * 4) isprint() wasn't considered when word counting.
|
---|
| 34 | *
|
---|
| 35 | * TODO:
|
---|
| 36 | *
|
---|
| 37 | * When locale support is enabled, count multibyte chars in the '-m' case.
|
---|
| 38 | *
|
---|
| 39 | * NOTES:
|
---|
| 40 | *
|
---|
| 41 | * The previous busybox wc attempted an optimization using stat for the
|
---|
| 42 | * case of counting chars only. I omitted that because it was broken.
|
---|
| 43 | * It didn't take into account the possibility of input coming from a
|
---|
| 44 | * pipe, or input from a file with file pointer not at the beginning.
|
---|
| 45 | *
|
---|
| 46 | * To implement such a speed optimization correctly, not only do you
|
---|
| 47 | * need the size, but also the file position. Note also that the
|
---|
| 48 | * file position may be past the end of file. Consider the example
|
---|
| 49 | * (adapted from example in gnu wc.c)
|
---|
| 50 | *
|
---|
| 51 | * echo hello > /tmp/testfile &&
|
---|
| 52 | * (dd ibs=1k skip=1 count=0 &> /dev/null ; wc -c) < /tmp/testfile
|
---|
| 53 | *
|
---|
| 54 | * for which 'wc -c' should output '0'.
|
---|
| 55 | */
|
---|
| 56 |
|
---|
| 57 | #include <stdio.h>
|
---|
| 58 | #include <stdlib.h>
|
---|
| 59 | #include <string.h>
|
---|
| 60 | #include <unistd.h>
|
---|
| 61 | #include "busybox.h"
|
---|
| 62 |
|
---|
| 63 | #ifdef CONFIG_LOCALE_SUPPORT
|
---|
| 64 | #include <locale.h>
|
---|
| 65 | #include <ctype.h>
|
---|
| 66 | #define isspace_given_isprint(c) isspace(c)
|
---|
| 67 | #else
|
---|
| 68 | #undef isspace
|
---|
| 69 | #undef isprint
|
---|
| 70 | #define isspace(c) ((((c) == ' ') || (((unsigned int)((c) - 9)) <= (13 - 9))))
|
---|
| 71 | #define isprint(c) (((unsigned int)((c) - 0x20)) <= (0x7e - 0x20))
|
---|
| 72 | #define isspace_given_isprint(c) ((c) == ' ')
|
---|
| 73 | #endif
|
---|
| 74 |
|
---|
| 75 | enum {
|
---|
| 76 | WC_LINES = 0,
|
---|
| 77 | WC_WORDS = 1,
|
---|
| 78 | WC_CHARS = 2,
|
---|
| 79 | WC_LENGTH = 3
|
---|
| 80 | };
|
---|
| 81 |
|
---|
| 82 | /* Note: If this changes, remember to change the initialization of
|
---|
| 83 | * 'name' in wc_main. It needs to point to the terminating nul. */
|
---|
| 84 | static const char wc_opts[] = "lwcL"; /* READ THE WARNING ABOVE! */
|
---|
| 85 |
|
---|
| 86 | enum {
|
---|
| 87 | OP_INC_LINE = 1, /* OP_INC_LINE must be 1. */
|
---|
| 88 | OP_SPACE = 2,
|
---|
| 89 | OP_NEWLINE = 4,
|
---|
| 90 | OP_TAB = 8,
|
---|
| 91 | OP_NUL = 16,
|
---|
| 92 | };
|
---|
| 93 |
|
---|
| 94 | /* Note: If fmt_str changes, the offsets to 's' in the OUTPUT section
|
---|
| 95 | * will need to be updated. */
|
---|
| 96 | static const char fmt_str[] = " %7u\0 %s\n";
|
---|
| 97 | static const char total_str[] = "total";
|
---|
| 98 |
|
---|
| 99 | int wc_main(int argc, char **argv)
|
---|
| 100 | {
|
---|
| 101 | FILE *fp;
|
---|
| 102 | const char *s;
|
---|
| 103 | unsigned int *pcounts;
|
---|
| 104 | unsigned int counts[4];
|
---|
| 105 | unsigned int totals[4];
|
---|
| 106 | unsigned int linepos;
|
---|
| 107 | unsigned int u;
|
---|
| 108 | int num_files = 0;
|
---|
| 109 | int c;
|
---|
| 110 | char status = EXIT_SUCCESS;
|
---|
| 111 | char in_word;
|
---|
| 112 | char print_type;
|
---|
| 113 |
|
---|
| 114 | print_type = bb_getopt_ulflags(argc, argv, wc_opts);
|
---|
| 115 |
|
---|
| 116 | if (print_type == 0) {
|
---|
| 117 | print_type = (1 << WC_LINES) | (1 << WC_WORDS) | (1 << WC_CHARS);
|
---|
| 118 | }
|
---|
| 119 |
|
---|
| 120 | argv += optind;
|
---|
| 121 | if (!*argv) {
|
---|
| 122 | *--argv = (char *) bb_msg_standard_input;
|
---|
| 123 | }
|
---|
| 124 |
|
---|
| 125 | memset(totals, 0, sizeof(totals));
|
---|
| 126 |
|
---|
| 127 | pcounts = counts;
|
---|
| 128 |
|
---|
| 129 | do {
|
---|
| 130 | ++num_files;
|
---|
| 131 | if (!(fp = bb_wfopen_input(*argv))) {
|
---|
| 132 | status = EXIT_FAILURE;
|
---|
| 133 | continue;
|
---|
| 134 | }
|
---|
| 135 |
|
---|
| 136 | memset(counts, 0, sizeof(counts));
|
---|
| 137 | linepos = 0;
|
---|
| 138 | in_word = 0;
|
---|
| 139 |
|
---|
| 140 | do {
|
---|
| 141 | ++counts[WC_CHARS];
|
---|
| 142 | c = getc(fp);
|
---|
| 143 | if (isprint(c)) {
|
---|
| 144 | ++linepos;
|
---|
| 145 | if (!isspace_given_isprint(c)) {
|
---|
| 146 | in_word = 1;
|
---|
| 147 | continue;
|
---|
| 148 | }
|
---|
| 149 | } else if (((unsigned int)(c - 9)) <= 4) {
|
---|
| 150 | /* \t 9
|
---|
| 151 | * \n 10
|
---|
| 152 | * \v 11
|
---|
| 153 | * \f 12
|
---|
| 154 | * \r 13
|
---|
| 155 | */
|
---|
| 156 | if (c == '\t') {
|
---|
| 157 | linepos = (linepos | 7) + 1;
|
---|
| 158 | } else { /* '\n', '\r', '\f', or '\v' */
|
---|
| 159 | DO_EOF:
|
---|
| 160 | if (linepos > counts[WC_LENGTH]) {
|
---|
| 161 | counts[WC_LENGTH] = linepos;
|
---|
| 162 | }
|
---|
| 163 | if (c == '\n') {
|
---|
| 164 | ++counts[WC_LINES];
|
---|
| 165 | }
|
---|
| 166 | if (c != '\v') {
|
---|
| 167 | linepos = 0;
|
---|
| 168 | }
|
---|
| 169 | }
|
---|
| 170 | } else if (c == EOF) {
|
---|
| 171 | if (ferror(fp)) {
|
---|
| 172 | bb_perror_msg("%s", *argv);
|
---|
| 173 | status = EXIT_FAILURE;
|
---|
| 174 | }
|
---|
| 175 | --counts[WC_CHARS];
|
---|
| 176 | goto DO_EOF; /* Treat an EOF as '\r'. */
|
---|
| 177 | } else {
|
---|
| 178 | continue;
|
---|
| 179 | }
|
---|
| 180 |
|
---|
| 181 | counts[WC_WORDS] += in_word;
|
---|
| 182 | in_word = 0;
|
---|
| 183 | if (c == EOF) {
|
---|
| 184 | break;
|
---|
| 185 | }
|
---|
| 186 | } while (1);
|
---|
| 187 |
|
---|
| 188 | if (totals[WC_LENGTH] < counts[WC_LENGTH]) {
|
---|
| 189 | totals[WC_LENGTH] = counts[WC_LENGTH];
|
---|
| 190 | }
|
---|
| 191 | totals[WC_LENGTH] -= counts[WC_LENGTH];
|
---|
| 192 |
|
---|
| 193 | bb_fclose_nonstdin(fp);
|
---|
| 194 |
|
---|
| 195 | OUTPUT:
|
---|
| 196 | s = fmt_str + 1; /* Skip the leading space on 1st pass. */
|
---|
| 197 | u = 0;
|
---|
| 198 | do {
|
---|
| 199 | if (print_type & (1 << u)) {
|
---|
| 200 | bb_printf(s, pcounts[u]);
|
---|
| 201 | s = fmt_str; /* Ok... restore the leading space. */
|
---|
| 202 | }
|
---|
| 203 | totals[u] += pcounts[u];
|
---|
| 204 | } while (++u < 4);
|
---|
| 205 |
|
---|
| 206 | s += 8; /* Set the format to the empty string. */
|
---|
| 207 |
|
---|
| 208 | if (*argv != bb_msg_standard_input) {
|
---|
| 209 | s -= 3; /* We have a name, so do %s conversion. */
|
---|
| 210 | }
|
---|
| 211 | bb_printf(s, *argv);
|
---|
| 212 |
|
---|
| 213 | } while (*++argv);
|
---|
| 214 |
|
---|
| 215 | /* If more than one file was processed, we want the totals. To save some
|
---|
| 216 | * space, we set the pcounts ptr to the totals array. This has the side
|
---|
| 217 | * effect of trashing the totals array after outputting it, but that's
|
---|
| 218 | * irrelavent since we no longer need it. */
|
---|
| 219 | if (num_files > 1) {
|
---|
| 220 | num_files = 0; /* Make sure we don't get here again. */
|
---|
| 221 | *--argv = (char *) total_str;
|
---|
| 222 | pcounts = totals;
|
---|
| 223 | goto OUTPUT;
|
---|
| 224 | }
|
---|
| 225 |
|
---|
| 226 | bb_fflush_stdout_and_exit(status);
|
---|
| 227 | }
|
---|