usr.bin/wc/wc.c

   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. Neither the name of the University nor the names of its contributors
  14  *    may be used to endorse or promote products derived from this software
  15  *    without specific prior written permission.
  16  *
  17  * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
  18  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
  19  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
  20  * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
  21  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
  22  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
  23  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
  24  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
  25  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
  26  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
  27  * SUCH DAMAGE.
  28  *
  29  * @(#) Copyright (c) 1980, 1987, 1991, 1993 The Regents of the University of California.  All rights reserved.
  30  * @(#)wc.c     8.1 (Berkeley) 6/6/93
  31  * $FreeBSD: head/usr.bin/wc/wc.c 281617 2015-04-16 21:44:35Z bdrewery $
  32  */
  33
  34 #include <sys/param.h>
  35 #include <sys/stat.h>
  36
  37 #include <ctype.h>
  38 #include <err.h>
  39 #include <errno.h>
  40 #include <fcntl.h>
  41 #include <locale.h>
  42 #include <stdint.h>
  43 #include <stdio.h>
  44 #include <stdlib.h>
  45 #include <string.h>
  46 #include <unistd.h>
  47 #include <wchar.h>
  48 #include <wctype.h>
  49
  50 static uintmax_t tlinect, twordct, tcharct, tlongline;
  51 static int doline, doword, dochar, domulti, dolongline;
  52 static volatile sig_atomic_t siginfo;
  53
  54 static void     show_cnt(const char *file, uintmax_t linect, uintmax_t wordct,
  55                     uintmax_t charct, uintmax_t llct);
  56 static int      cnt(const char *);
  57 static void     usage(void);
  58
  59 #ifdef SIGINFO
  60 static void
  61 siginfo_handler(int sig __unused)
  62 {
  63         siginfo = 1;
  64 }
  65 #endif
  66
  67 static void
  68 reset_siginfo(void)
  69 {
  70 #ifdef SIGINFO
  71         signal(SIGINFO, SIG_DFL);
  72 #endif
  73         siginfo = 0;
  74 }
  75
  76 int
  77 main(int argc, char *argv[])
  78 {
  79         int ch, errors, total;
  80
  81         setlocale(LC_CTYPE, "");
  82
  83         while ((ch = getopt(argc, argv, "clmwL")) != -1) {
  84                 switch (ch) {
  85                 case 'l':
  86                         doline = 1;
  87                         break;
  88                 case 'w':
  89                         doword = 1;
  90                         break;
  91                 case 'c':
  92                         dochar = 1;
  93                         domulti = 0;
  94                         break;
  95                 case 'L':
  96                         dolongline = 1;
  97                         break;
  98                 case 'm':
  99                         domulti = 1;
 100                         dochar = 0;
 101                         break;
 102                 case '?':
 103                 default:
 104                         usage();
 105                 }
 106         }
 107         argv += optind;
 108         argc -= optind;
 109
 110 #ifdef SIGINFO
 111         signal(SIGINFO, siginfo_handler);
 112 #endif
 113
 114         /* Wc's flags are on by default. */
 115         if (doline + doword + dochar + domulti + dolongline == 0)
 116                 doline = doword = dochar = 1;
 117
 118         errors = 0;
 119         total = 0;
 120         if (*argv == NULL) {
 121                 if (cnt(NULL) != 0)
 122                         ++errors;
 123         } else {
 124                 do {
 125                         if (cnt(*argv) != 0)
 126                                 ++errors;
 127                         ++total;
 128                 } while(*++argv);
 129         }
 130
 131         if (total > 1)
 132                 show_cnt("total", tlinect, twordct, tcharct, tlongline);
 133         exit(errors == 0 ? 0 : 1);
 134 }
 135
 136 static void
 137 show_cnt(const char *file, uintmax_t linect, uintmax_t wordct,
 138     uintmax_t charct, uintmax_t llct)
 139 {
 140         FILE *out;
 141
 142         if (!siginfo)
 143                 out = stdout;
 144         else {
 145                 out = stderr;
 146                 siginfo = 0;
 147         }
 148
 149         if (doline)
 150                 fprintf(out, " %7ju", linect);
 151         if (doword)
 152                 fprintf(out, " %7ju", wordct);
 153         if (dochar || domulti)
 154                 fprintf(out, " %7ju", charct);
 155         if (dolongline)
 156                 fprintf(out, " %7ju", llct);
 157         if (file != NULL)
 158                 fprintf(out, " %s\n", file);
 159         else
 160                 fprintf(out, "\n");
 161 }
 162
 163 static int
 164 cnt(const char *file)
 165 {
 166         struct stat sb;
 167         uintmax_t linect, wordct, charct, llct, tmpll;
 168         int fd, len, warned;
 169         size_t clen;
 170         short gotsp;
 171         u_char *p;
 172         u_char buf[MAXBSIZE];
 173         wchar_t wch;
 174         mbstate_t mbs;
 175
 176         linect = wordct = charct = llct = tmpll = 0;
 177         if (file == NULL) {
 178                 fd = STDIN_FILENO;
 179         } else {
 180                 if ((fd = open(file, O_RDONLY)) < 0) {
 181                         warn("%s: open", file);
 182                         return (1);
 183                 }
 184                 if (doword || (domulti && MB_CUR_MAX != 1))
 185                         goto word;
 186                 /*
 187                  * Line counting is split out because it's a lot faster to get
 188                  * lines than to get words, since the word count requires some
 189                  * logic.
 190                  */
 191                 if (doline) {
 192                         while ((len = read(fd, buf, MAXBSIZE))) {
 193                                 if (len == -1) {
 194                                         warn("%s: read", file);
 195                                         close(fd);
 196                                         return (1);
 197                                 }
 198                                 if (siginfo) {
 199                                         show_cnt(file, linect, wordct, charct,
 200                                             llct);
 201                                 }
 202                                 charct += len;
 203                                 for (p = buf; len--; ++p) {
 204                                         if (*p == '\n') {
 205                                                 if (tmpll > llct)
 206                                                         llct = tmpll;
 207                                                 tmpll = 0;
 208                                                 ++linect;
 209                                         } else
 210                                                 tmpll++;
 211                                 }
 212                         }
 213                         reset_siginfo();
 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                         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                                 close(fd);
 233                                 return (1);
 234                         }
 235                         /* Pseudo-filesystems advertize a zero size. */
 236                         if (S_ISREG(sb.st_mode) && sb.st_size > 0) {
 237                                 reset_siginfo();
 238                                 charct = sb.st_size;
 239                                 show_cnt(file, linect, wordct, charct, llct);
 240                                 tcharct += charct;
 241                                 close(fd);
 242                                 return (0);
 243                         }
 244                 }
 245         }
 246
 247         /* Do it the hard way... */
 248 word:   gotsp = 1;
 249         warned = 0;
 250         memset(&mbs, 0, sizeof(mbs));
 251         while ((len = read(fd, buf, MAXBSIZE)) != 0) {
 252                 if (len == -1) {
 253                         warn("%s: read", file != NULL ? file : "stdin");
 254                         close(fd);
 255                         return (1);
 256                 }
 257                 p = buf;
 258                 while (len > 0) {
 259                         if (siginfo)
 260                                 show_cnt(file, linect, wordct, charct, llct);
 261                         if (!domulti || MB_CUR_MAX == 1) {
 262                                 clen = 1;
 263                                 wch = (unsigned char)*p;
 264                         } else if ((clen = mbrtowc(&wch, p, len, &mbs)) ==
 265                             (size_t)-1) {
 266                                 if (!warned) {
 267                                         errno = EILSEQ;
 268                                         warn("%s",
 269                                             file != NULL ? file : "stdin");
 270                                         warned = 1;
 271                                 }
 272                                 memset(&mbs, 0, sizeof(mbs));
 273                                 clen = 1;
 274                                 wch = (unsigned char)*p;
 275                         } else if (clen == (size_t)-2)
 276                                 break;
 277                         else if (clen == 0)
 278                                 clen = 1;
 279                         charct++;
 280                         if (wch != L'\n')
 281                                 tmpll++;
 282                         len -= clen;
 283                         p += clen;
 284                         if (wch == L'\n') {
 285                                 if (tmpll > llct)
 286                                         llct = tmpll;
 287                                 tmpll = 0;
 288                                 ++linect;
 289                         }
 290                         if (iswspace(wch))
 291                                 gotsp = 1;
 292                         else if (gotsp) {
 293                                 gotsp = 0;
 294                                 ++wordct;
 295                         }
 296                 }
 297         }
 298         reset_siginfo();
 299         if (domulti && MB_CUR_MAX > 1)
 300                 if (mbrtowc(NULL, NULL, 0, &mbs) == (size_t)-1 && !warned)
 301                         warn("%s", file != NULL ? file : "stdin");
 302         if (doline)
 303                 tlinect += linect;
 304         if (doword)
 305                 twordct += wordct;
 306         if (dochar || domulti)
 307                 tcharct += charct;
 308         if (dolongline) {
 309                 if (llct > tlongline)
 310                         tlongline = llct;
 311         }
 312         show_cnt(file, linect, wordct, charct, llct);
 313         close(fd);
 314         return (0);
 315 }
 316
 317 static void
 318 usage(void)
 319 {
 320         fprintf(stderr, "usage: wc [-Lclmw] [file ...]\n");
 321         exit(1);
 322 }