wc.c revision 208210
1/* 2 * Copyright (c) 1980, 1987, 1991, 1993 3 * The Regents of the University of California. All rights reserved. 4 * 5 * Redistribution and use in source and binary forms, with or without 6 * modification, are permitted provided that the following conditions 7 * are met: 8 * 1. Redistributions of source code must retain the above copyright 9 * notice, this list of conditions and the following disclaimer. 10 * 2. Redistributions in binary form must reproduce the above copyright 11 * notice, this list of conditions and the following disclaimer in the 12 * documentation and/or other materials provided with the distribution. 13 * 3. All advertising materials mentioning features or use of this software 14 * must display the following acknowledgement: 15 * This product includes software developed by the University of 16 * California, Berkeley and its contributors. 17 * 4. Neither the name of the University nor the names of its contributors 18 * may be used to endorse or promote products derived from this software 19 * without specific prior written permission. 20 * 21 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND 22 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 23 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 24 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE 25 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 26 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 27 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 28 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 29 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 30 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 31 * SUCH DAMAGE. 32 */ 33 34#ifndef lint 35static const char copyright[] = 36"@(#) Copyright (c) 1980, 1987, 1991, 1993\n\ 37 The Regents of the University of California. All rights reserved.\n"; 38#endif /* not lint */ 39 40#if 0 41#ifndef lint 42static char sccsid[] = "@(#)wc.c 8.1 (Berkeley) 6/6/93"; 43#endif /* not lint */ 44#endif 45 46#include <sys/cdefs.h> 47__FBSDID("$FreeBSD: head/usr.bin/wc/wc.c 208210 2010-05-17 19:13:49Z pjd $"); 48 49#include <sys/param.h> 50#include <sys/stat.h> 51 52#include <ctype.h> 53#include <err.h> 54#include <errno.h> 55#include <fcntl.h> 56#include <locale.h> 57#include <stdint.h> 58#include <stdio.h> 59#include <stdlib.h> 60#include <string.h> 61#include <unistd.h> 62#include <wchar.h> 63#include <wctype.h> 64 65uintmax_t tlinect, twordct, tcharct, tlongline; 66int doline, doword, dochar, domulti, dolongline; 67static volatile sig_atomic_t siginfo; 68 69static void show_cnt(const char *file, uintmax_t linect, uintmax_t wordct, 70 uintmax_t charct, uintmax_t llct); 71static int cnt(const char *); 72static void usage(void); 73 74static void 75siginfo_handler(int sig __unused) 76{ 77 78 siginfo = 1; 79} 80 81int 82main(int argc, char *argv[]) 83{ 84 int ch, errors, total; 85 86 (void) setlocale(LC_CTYPE, ""); 87 88 while ((ch = getopt(argc, argv, "clmwL")) != -1) 89 switch((char)ch) { 90 case 'l': 91 doline = 1; 92 break; 93 case 'w': 94 doword = 1; 95 break; 96 case 'c': 97 dochar = 1; 98 domulti = 0; 99 break; 100 case 'L': 101 dolongline = 1; 102 break; 103 case 'm': 104 domulti = 1; 105 dochar = 0; 106 break; 107 case '?': 108 default: 109 usage(); 110 } 111 argv += optind; 112 argc -= optind; 113 114 (void)signal(SIGINFO, siginfo_handler); 115 116 /* Wc's flags are on by default. */ 117 if (doline + doword + dochar + domulti + dolongline == 0) 118 doline = doword = dochar = 1; 119 120 errors = 0; 121 total = 0; 122 if (!*argv) { 123 if (cnt((char *)NULL) != 0) 124 ++errors; 125 } else { 126 do { 127 if (cnt(*argv) != 0) 128 ++errors; 129 ++total; 130 } while(*++argv); 131 } 132 133 if (total > 1) 134 show_cnt("total", tlinect, twordct, tcharct, tlongline); 135 exit(errors == 0 ? 0 : 1); 136} 137 138static void 139show_cnt(const char *file, uintmax_t linect, uintmax_t wordct, 140 uintmax_t charct, uintmax_t llct) 141{ 142 FILE *out; 143 144 if (!siginfo) 145 out = stdout; 146 else { 147 out = stderr; 148 siginfo = 0; 149 } 150 151 if (doline) 152 (void)fprintf(out, " %7ju", linect); 153 if (doword) 154 (void)fprintf(out, " %7ju", wordct); 155 if (dochar || domulti) 156 (void)fprintf(out, " %7ju", charct); 157 if (dolongline) 158 (void)fprintf(out, " %7ju", llct); 159 if (file != NULL) 160 (void)fprintf(out, " %s\n", file); 161 else 162 (void)fprintf(out, "\n"); 163} 164 165static int 166cnt(const char *file) 167{ 168 struct stat sb; 169 uintmax_t linect, wordct, charct, llct, tmpll; 170 int fd, len, warned; 171 size_t clen; 172 short gotsp; 173 u_char *p; 174 u_char buf[MAXBSIZE]; 175 wchar_t wch; 176 mbstate_t mbs; 177 178 linect = wordct = charct = llct = tmpll = 0; 179 if (file == NULL) 180 fd = STDIN_FILENO; 181 else { 182 if ((fd = open(file, O_RDONLY, 0)) < 0) { 183 warn("%s: open", file); 184 return (1); 185 } 186 if (doword || (domulti && MB_CUR_MAX != 1)) 187 goto word; 188 /* 189 * Line counting is split out because it's a lot faster to get 190 * lines than to get words, since the word count requires some 191 * logic. 192 */ 193 if (doline) { 194 while ((len = read(fd, buf, MAXBSIZE))) { 195 if (len == -1) { 196 warn("%s: read", file); 197 (void)close(fd); 198 return (1); 199 } 200 if (siginfo) { 201 show_cnt(file, linect, wordct, charct, 202 llct); 203 } 204 charct += len; 205 for (p = buf; len--; ++p) 206 if (*p == '\n') { 207 if (tmpll > llct) 208 llct = tmpll; 209 tmpll = 0; 210 ++linect; 211 } else 212 tmpll++; 213 } 214 tlinect += linect; 215 if (dochar) 216 tcharct += charct; 217 if (dolongline) { 218 if (llct > tlongline) 219 tlongline = llct; 220 } 221 show_cnt(file, linect, wordct, charct, llct); 222 (void)close(fd); 223 return (0); 224 } 225 /* 226 * If all we need is the number of characters and it's a 227 * regular file, just stat the puppy. 228 */ 229 if (dochar || domulti) { 230 if (fstat(fd, &sb)) { 231 warn("%s: fstat", file); 232 (void)close(fd); 233 return (1); 234 } 235 if (S_ISREG(sb.st_mode)) { 236 charct = sb.st_size; 237 show_cnt(file, linect, wordct, charct, llct); 238 tcharct += charct; 239 (void)close(fd); 240 return (0); 241 } 242 } 243 } 244 245 /* Do it the hard way... */ 246word: gotsp = 1; 247 warned = 0; 248 memset(&mbs, 0, sizeof(mbs)); 249 while ((len = read(fd, buf, MAXBSIZE)) != 0) { 250 if (len == -1) { 251 warn("%s: read", file != NULL ? file : "stdin"); 252 (void)close(fd); 253 return (1); 254 } 255 p = buf; 256 while (len > 0) { 257 if (siginfo) 258 show_cnt(file, linect, wordct, charct, llct); 259 if (!domulti || MB_CUR_MAX == 1) { 260 clen = 1; 261 wch = (unsigned char)*p; 262 } else if ((clen = mbrtowc(&wch, p, len, &mbs)) == 263 (size_t)-1) { 264 if (!warned) { 265 errno = EILSEQ; 266 warn("%s", 267 file != NULL ? file : "stdin"); 268 warned = 1; 269 } 270 memset(&mbs, 0, sizeof(mbs)); 271 clen = 1; 272 wch = (unsigned char)*p; 273 } else if (clen == (size_t)-2) 274 break; 275 else if (clen == 0) 276 clen = 1; 277 charct++; 278 if (wch != L'\n') 279 tmpll++; 280 len -= clen; 281 p += clen; 282 if (wch == L'\n') { 283 if (tmpll > llct) 284 llct = tmpll; 285 tmpll = 0; 286 ++linect; 287 } 288 if (iswspace(wch)) 289 gotsp = 1; 290 else if (gotsp) { 291 gotsp = 0; 292 ++wordct; 293 } 294 } 295 } 296 if (domulti && MB_CUR_MAX > 1) 297 if (mbrtowc(NULL, NULL, 0, &mbs) == (size_t)-1 && !warned) 298 warn("%s", file != NULL ? file : "stdin"); 299 if (doline) 300 tlinect += linect; 301 if (doword) 302 twordct += wordct; 303 if (dochar || domulti) 304 tcharct += charct; 305 if (dolongline) { 306 if (llct > tlongline) 307 tlongline = llct; 308 } 309 show_cnt(file, linect, wordct, charct, llct); 310 (void)close(fd); 311 return (0); 312} 313 314static void 315usage(void) 316{ 317 (void)fprintf(stderr, "usage: wc [-Lclmw] [file ...]\n"); 318 exit(1); 319} 320