usr.bin/vis/vis.c

   1 /*      $NetBSD: vis.c,v 1.22 2013/02/20 17:04:45 christos Exp $        */
   2
   3 /*-
   4  * Copyright (c) 1989, 1993
   5  *      The Regents of the University of California.  All rights reserved.
   6  *
   7  * Redistribution and use in source and binary forms, with or without
   8  * modification, are permitted provided that the following conditions
   9  * are met:
  10  * 1. Redistributions of source code must retain the above copyright
  11  *    notice, this list of conditions and the following disclaimer.
  12  * 2. Redistributions in binary form must reproduce the above copyright
  13  *    notice, this list of conditions and the following disclaimer in the
  14  *    documentation and/or other materials provided with the distribution.
  15  * 3. Neither the name of the University nor the names of its contributors
  16  *    may be used to endorse or promote products derived from this software
  17  *    without specific prior written permission.
  18  *
  19  * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
  20  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
  21  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
  22  * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
  23  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
  24  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
  25  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
  26  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
  27  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
  28  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
  29  * SUCH DAMAGE.
  30  */
  31
  32 #include <sys/cdefs.h>
  33 #ifndef lint
  34 __COPYRIGHT("@(#) Copyright (c) 1989, 1993\
  35  The Regents of the University of California.  All rights reserved.");
  36 #endif /* not lint */
  37
  38 #ifndef lint
  39 #if 0
  40 static char sccsid[] = "@(#)vis.c       8.1 (Berkeley) 6/6/93";
  41 #endif
  42 __RCSID("$NetBSD: vis.c,v 1.22 2013/02/20 17:04:45 christos Exp $");
  43 #endif /* not lint */
  44
  45 #include <stdio.h>
  46 #include <string.h>
  47 #include <stdlib.h>
  48 #include <string.h>
  49 #include <errno.h>
  50 #include <wchar.h>
  51 #include <limits.h>
  52 #include <unistd.h>
  53 #include <err.h>
  54 #include <vis.h>
  55
  56 #include "extern.h"
  57
  58 static int eflags, fold, foldwidth = 80, none, markeol;
  59 #ifdef DEBUG
  60 int debug;
  61 #endif
  62 static const char *extra = "";
  63
  64 static void process(FILE *);
  65
  66 int
  67 main(int argc, char *argv[])
  68 {
  69         FILE *fp;
  70         int ch;
  71         int rval;
  72
  73         while ((ch = getopt(argc, argv, "bcde:F:fhlmnostw")) != -1)
  74                 switch((char)ch) {
  75                 case 'b':
  76                         eflags |= VIS_NOSLASH;
  77                         break;
  78                 case 'c':
  79                         eflags |= VIS_CSTYLE;
  80                         break;
  81 #ifdef DEBUG
  82                 case 'd':
  83                         debug++;
  84                         break;
  85 #endif
  86                 case 'e':
  87                         extra = optarg;
  88                         break;
  89                 case 'F':
  90                         if ((foldwidth = atoi(optarg)) < 5) {
  91                                 errx(1, "can't fold lines to less than 5 cols");
  92                                 /* NOTREACHED */
  93                         }
  94                         markeol++;
  95                         break;
  96                 case 'f':
  97                         fold++;         /* fold output lines to 80 cols */
  98                         break;          /* using hidden newline */
  99                 case 'h':
 100                         eflags |= VIS_HTTPSTYLE;
 101                         break;
 102                 case 'l':
 103                         markeol++;      /* mark end of line with \$ */
 104                         break;
 105                 case 'm':
 106                         eflags |= VIS_MIMESTYLE;
 107                         if (foldwidth == 80)
 108                                 foldwidth = 76;
 109                         break;
 110                 case 'n':
 111                         none++;
 112                         break;
 113                 case 'o':
 114                         eflags |= VIS_OCTAL;
 115                         break;
 116                 case 's':
 117                         eflags |= VIS_SAFE;
 118                         break;
 119                 case 't':
 120                         eflags |= VIS_TAB;
 121                         break;
 122                 case 'w':
 123                         eflags |= VIS_WHITE;
 124                         break;
 125                 case '?':
 126                 default:
 127                         (void)fprintf(stderr,
 128                             "Usage: %s [-bcfhlmnostw] [-e extra]"
 129                             " [-F foldwidth] [file ...]\n", getprogname());
 130                         return 1;
 131                 }
 132
 133         if ((eflags & (VIS_HTTPSTYLE|VIS_MIMESTYLE)) ==
 134             (VIS_HTTPSTYLE|VIS_MIMESTYLE))
 135                 errx(1, "Can't specify -m and -h at the same time");
 136
 137         argc -= optind;
 138         argv += optind;
 139
 140         rval = 0;
 141
 142         if (*argv)
 143                 while (*argv) {
 144                         if ((fp = fopen(*argv, "r")) != NULL) {
 145                                 process(fp);
 146                                 (void)fclose(fp);
 147                         } else {
 148                                 warn("%s", *argv);
 149                                 rval = 1;
 150                         }
 151                         argv++;
 152                 }
 153         else
 154                 process(stdin);
 155         return rval;
 156 }
 157
 158 static void
 159 process(FILE *fp)
 160 {
 161         static int col = 0;
 162         static char nul[] = "\0";
 163         char *cp = nul + 1;     /* so *(cp-1) starts out != '\n' */
 164         wint_t c, c1, rachar;
 165         char mbibuff[2 * MB_LEN_MAX + 1]; /* max space for 2 wchars */
 166         char buff[4 * MB_LEN_MAX + 1]; /* max encoding length for one char */
 167         int mbilen, cerr = 0, raerr = 0;
 168
 169         /*
 170          * The input stream is considered to be multibyte characters.
 171          * The input loop will read this data inputing one character,
 172          * possibly multiple bytes, at a time and converting each to
 173          * a wide character wchar_t.
 174          *
 175          * The vis(3) functions, however, require single either bytes
 176          * or a multibyte string as their arguments.  So we convert
 177          * our input wchar_t and the following look-ahead wchar_t to
 178          * a multibyte string for processing by vis(3).
 179          */
 180
 181         /* Read one multibyte character, store as wchar_t */
 182         c = getwc(fp);
 183         if (c == WEOF && errno == EILSEQ) {
 184                 /* Error in multibyte data.  Read one byte. */
 185                 c = (wint_t)getc(fp);
 186                 cerr = 1;
 187         }
 188         while (c != WEOF) {
 189                 /* Clear multibyte input buffer. */
 190                 memset(mbibuff, 0, sizeof(mbibuff));
 191                 /* Read-ahead next multibyte character. */
 192                 if (!cerr)
 193                         rachar = getwc(fp);
 194                 if (cerr || (rachar == WEOF && errno == EILSEQ)) {
 195                         /* Error in multibyte data.  Read one byte. */
 196                         rachar = (wint_t)getc(fp);
 197                         raerr = 1;
 198                 }
 199                 if (none) {
 200                         /* Handle -n flag. */
 201                         cp = buff;
 202                         *cp++ = c;
 203                         if (c == '\\')
 204                                 *cp++ = '\\';
 205                         *cp = '\0';
 206                 } else if (markeol && c == '\n') {
 207                         /* Handle -l flag. */
 208                         cp = buff;
 209                         if ((eflags & VIS_NOSLASH) == 0)
 210                                 *cp++ = '\\';
 211                         *cp++ = '$';
 212                         *cp++ = '\n';
 213                         *cp = '\0';
 214                 } else {
 215                         /*
 216                          * Convert character using vis(3) library.
 217                          * At this point we will process one character.
 218                          * But we must pass the vis(3) library this
 219                          * character plus the next one because the next
 220                          * one is used as a look-ahead to decide how to
 221                          * encode this one under certain circumstances.
 222                          *
 223                          * Since our characters may be multibyte, e.g.,
 224                          * in the UTF-8 locale, we cannot use vis() and
 225                          * svis() which require byte input, so we must
 226                          * create a multibyte string and use strvisx().
 227                          */
 228                         /* Treat EOF as a NUL char. */
 229                         c1 = rachar;
 230                         if (c1 == WEOF)
 231                                 c1 = L'\0';
 232                         /*
 233                          * If we hit a multibyte conversion error above,
 234                          * insert byte directly into string buff because
 235                          * wctomb() will fail.  Else convert wchar_t to
 236                          * multibyte using wctomb().
 237                          */
 238                         if (cerr) {
 239                                 *mbibuff = (char)c;
 240                                 mbilen = 1;
 241                         } else
 242                                 mbilen = wctomb(mbibuff, c);
 243                         /* Same for look-ahead character. */
 244                         if (raerr)
 245                                 mbibuff[mbilen] = (char)c1;
 246                         else
 247                                 wctomb(mbibuff + mbilen, c1);
 248                         /* Perform encoding on just first character. */
 249                         (void) strsenvisx(buff, 4 * MB_LEN_MAX, mbibuff,
 250                             1, eflags, extra, &cerr);
 251                 }
 252
 253                 cp = buff;
 254                 if (fold) {
 255 #ifdef DEBUG
 256                         if (debug)
 257                                 (void)printf("<%02d,", col);
 258 #endif
 259                         col = foldit(cp, col, foldwidth, eflags);
 260 #ifdef DEBUG
 261                         if (debug)
 262                                 (void)printf("%02d>", col);
 263 #endif
 264                 }
 265                 do {
 266                         (void)putchar(*cp);
 267                 } while (*++cp);
 268                 c = rachar;
 269                 cerr = raerr;
 270         }
 271         /*
 272          * terminate partial line with a hidden newline
 273          */
 274         if (fold && *(cp - 1) != '\n')
 275                 (void)printf(eflags & VIS_MIMESTYLE ? "=\n" : "\\\n");
 276 }