src/fold.c

   1 /* fold -- wrap each input line to fit in specified width.
   2    Copyright (C) 1991-2023 Free Software Foundation, Inc.
   3
   4    This program is free software: you can redistribute it and/or modify
   5    it under the terms of the GNU General Public License as published by
   6    the Free Software Foundation, either version 3 of the License, or
   7    (at your option) any later version.
   8
   9    This program is distributed in the hope that it will be useful,
  10    but WITHOUT ANY WARRANTY; without even the implied warranty of
  11    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  12    GNU General Public License for more details.
  13
  14    You should have received a copy of the GNU General Public License
  15    along with this program.  If not, see <https://www.gnu.org/licenses/>.  */
  16
  17 /* Written by David MacKenzie, djm@gnu.ai.mit.edu. */
  18
  19 #include <config.h>
  20
  21 #include <ctype.h>
  22 #include <stdio.h>
  23 #include <getopt.h>
  24 #include <sys/types.h>
  25
  26 #include "system.h"
  27 #include "fadvise.h"
  28 #include "xdectoint.h"
  29
  30 #define TAB_WIDTH 8
  31
  32 /* The official name of this program (e.g., no 'g' prefix).  */
  33 #define PROGRAM_NAME "fold"
  34
  35 #define AUTHORS proper_name ("David MacKenzie")
  36
  37 /* If nonzero, try to break on whitespace. */
  38 static bool break_spaces;
  39
  40 /* If nonzero, count bytes, not column positions. */
  41 static bool count_bytes;
  42
  43 /* If nonzero, at least one of the files we read was standard input. */
  44 static bool have_read_stdin;
  45
  46 static char const shortopts[] = "bsw:0::1::2::3::4::5::6::7::8::9::";
  47
  48 static struct option const longopts[] =
  49 {
  50   {"bytes", no_argument, nullptr, 'b'},
  51   {"spaces", no_argument, nullptr, 's'},
  52   {"width", required_argument, nullptr, 'w'},
  53   {GETOPT_HELP_OPTION_DECL},
  54   {GETOPT_VERSION_OPTION_DECL},
  55   {nullptr, 0, nullptr, 0}
  56 };
  57
  58 void
  59 usage (int status)
  60 {
  61   if (status != EXIT_SUCCESS)
  62     emit_try_help ();
  63   else
  64     {
  65       printf (_("\
  66 Usage: %s [OPTION]... [FILE]...\n\
  67 "),
  68               program_name);
  69       fputs (_("\
  70 Wrap input lines in each FILE, writing to standard output.\n\
  71 "), stdout);
  72
  73       emit_stdin_note ();
  74       emit_mandatory_arg_note ();
  75
  76       fputs (_("\
  77   -b, --bytes         count bytes rather than columns\n\
  78   -s, --spaces        break at spaces\n\
  79   -w, --width=WIDTH   use WIDTH columns instead of 80\n\
  80 "), stdout);
  81       fputs (HELP_OPTION_DESCRIPTION, stdout);
  82       fputs (VERSION_OPTION_DESCRIPTION, stdout);
  83       emit_ancillary_info (PROGRAM_NAME);
  84     }
  85   exit (status);
  86 }
  87
  88 /* Assuming the current column is COLUMN, return the column that
  89    printing C will move the cursor to.
  90    The first column is 0. */
  91
  92 static size_t
  93 adjust_column (size_t column, char c)
  94 {
  95   if (!count_bytes)
  96     {
  97       if (c == '\b')
  98         {
  99           if (column > 0)
 100             column--;
 101         }
 102       else if (c == '\r')
 103         column = 0;
 104       else if (c == '\t')
 105         column += TAB_WIDTH - column % TAB_WIDTH;
 106       else /* if (isprint (c)) */
 107         column++;
 108     }
 109   else
 110     column++;
 111   return column;
 112 }
 113
 114 /* Fold file FILENAME, or standard input if FILENAME is "-",
 115    to stdout, with maximum line length WIDTH.
 116    Return true if successful.  */
 117
 118 static bool
 119 fold_file (char const *filename, size_t width)
 120 {
 121   FILE *istream;
 122   int c;
 123   size_t column = 0;            /* Screen column where next char will go. */
 124   size_t offset_out = 0;        /* Index in 'line_out' for next char. */
 125   static char *line_out = nullptr;
 126   static size_t allocated_out = 0;
 127   int saved_errno;
 128
 129   if (STREQ (filename, "-"))
 130     {
 131       istream = stdin;
 132       have_read_stdin = true;
 133     }
 134   else
 135     istream = fopen (filename, "r");
 136
 137   if (istream == nullptr)
 138     {
 139       error (0, errno, "%s", quotef (filename));
 140       return false;
 141     }
 142
 143   fadvise (istream, FADVISE_SEQUENTIAL);
 144
 145   while ((c = getc (istream)) != EOF)
 146     {
 147       if (offset_out + 1 >= allocated_out)
 148         line_out = X2REALLOC (line_out, &allocated_out);
 149
 150       if (c == '\n')
 151         {
 152           line_out[offset_out++] = c;
 153           fwrite (line_out, sizeof (char), offset_out, stdout);
 154           column = offset_out = 0;
 155           continue;
 156         }
 157
 158     rescan:
 159       column = adjust_column (column, c);
 160
 161       if (column > width)
 162         {
 163           /* This character would make the line too long.
 164              Print the line plus a newline, and make this character
 165              start the next line. */
 166           if (break_spaces)
 167             {
 168               bool found_blank = false;
 169               size_t logical_end = offset_out;
 170
 171               /* Look for the last blank. */
 172               while (logical_end)
 173                 {
 174                   --logical_end;
 175                   if (isblank (to_uchar (line_out[logical_end])))
 176                     {
 177                       found_blank = true;
 178                       break;
 179                     }
 180                 }
 181
 182               if (found_blank)
 183                 {
 184                   size_t i;
 185
 186                   /* Found a blank.  Don't output the part after it. */
 187                   logical_end++;
 188                   fwrite (line_out, sizeof (char), (size_t) logical_end,
 189                           stdout);
 190                   putchar ('\n');
 191                   /* Move the remainder to the beginning of the next line.
 192                      The areas being copied here might overlap. */
 193                   memmove (line_out, line_out + logical_end,
 194                            offset_out - logical_end);
 195                   offset_out -= logical_end;
 196                   for (column = i = 0; i < offset_out; i++)
 197                     column = adjust_column (column, line_out[i]);
 198                   goto rescan;
 199                 }
 200             }
 201
 202           if (offset_out == 0)
 203             {
 204               line_out[offset_out++] = c;
 205               continue;
 206             }
 207
 208           line_out[offset_out++] = '\n';
 209           fwrite (line_out, sizeof (char), (size_t) offset_out, stdout);
 210           column = offset_out = 0;
 211           goto rescan;
 212         }
 213
 214       line_out[offset_out++] = c;
 215     }
 216
 217   saved_errno = errno;
 218   if (!ferror (istream))
 219     saved_errno = 0;
 220
 221   if (offset_out)
 222     fwrite (line_out, sizeof (char), (size_t) offset_out, stdout);
 223
 224   if (STREQ (filename, "-"))
 225     clearerr (istream);
 226   else if (fclose (istream) != 0 && !saved_errno)
 227     saved_errno = errno;
 228
 229   if (saved_errno)
 230     {
 231       error (0, saved_errno, "%s", quotef (filename));
 232       return false;
 233     }
 234
 235   return true;
 236 }
 237
 238 int
 239 main (int argc, char **argv)
 240 {
 241   size_t width = 80;
 242   int i;
 243   int optc;
 244   bool ok;
 245
 246   initialize_main (&argc, &argv);
 247   set_program_name (argv[0]);
 248   setlocale (LC_ALL, "");
 249   bindtextdomain (PACKAGE, LOCALEDIR);
 250   textdomain (PACKAGE);
 251
 252   atexit (close_stdout);
 253
 254   break_spaces = count_bytes = have_read_stdin = false;
 255
 256   while ((optc = getopt_long (argc, argv, shortopts, longopts, nullptr)) != -1)
 257     {
 258       char optargbuf[2];
 259
 260       switch (optc)
 261         {
 262         case 'b':               /* Count bytes rather than columns. */
 263           count_bytes = true;
 264           break;
 265
 266         case 's':               /* Break at word boundaries. */
 267           break_spaces = true;
 268           break;
 269
 270         case '0': case '1': case '2': case '3': case '4':
 271         case '5': case '6': case '7': case '8': case '9':
 272           if (optarg)
 273             optarg--;
 274           else
 275             {
 276               optargbuf[0] = optc;
 277               optargbuf[1] = '\0';
 278               optarg = optargbuf;
 279             }
 280           FALLTHROUGH;
 281         case 'w':               /* Line width. */
 282           width = xdectoumax (optarg, 1, SIZE_MAX - TAB_WIDTH - 1, "",
 283                               _("invalid number of columns"), 0);
 284           break;
 285
 286         case_GETOPT_HELP_CHAR;
 287
 288         case_GETOPT_VERSION_CHAR (PROGRAM_NAME, AUTHORS);
 289
 290         default:
 291           usage (EXIT_FAILURE);
 292         }
 293     }
 294
 295   if (argc == optind)
 296     ok = fold_file ("-", width);
 297   else
 298     {
 299       ok = true;
 300       for (i = optind; i < argc; i++)
 301         ok &= fold_file (argv[i], width);
 302     }
 303
 304   if (have_read_stdin && fclose (stdin) == EOF)
 305     error (EXIT_FAILURE, errno, "-");
 306
 307   return ok ? EXIT_SUCCESS : EXIT_FAILURE;
 308 }