1 /* unexpand - convert spaces to tabs
2 Copyright (C) 89, 91, 1995-2003 Free Software Foundation, Inc.
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 2, or (at your option)
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.
14 You should have received a copy of the GNU General Public License
15 along with this program; if not, write to the Free Software Foundation,
16 Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. */
18 /* By default, convert only maximal strings of initial blanks and tabs
20 Preserves backspace characters in the output; they decrement the
21 column count for tab calculations.
22 The default action is equivalent to -8.
25 --tabs=tab1[,tab2[,...]]
27 -tab1[,tab2[,...]] If only one tab stop is given, set the tabs tab1
28 spaces apart instead of the default 8. Otherwise,
29 set the tabs at columns tab1, tab2, etc. (numbered from
30 0); replace any tabs beyond the tabstops given with
33 -a Use tabs wherever they would replace 2 or more spaces,
34 not just at the beginnings of lines.
36 David MacKenzie <djm@gnu.ai.mit.edu> */
42 #include <sys/types.h>
47 /* The official name of this program (e.g., no `g' prefix). */
48 #define PROGRAM_NAME "unexpand"
50 #define AUTHORS "David MacKenzie"
52 /* The number of bytes added at a time to the amount of memory
53 allocated for the output line. */
54 #define OUTPUT_BLOCK 256
56 /* The number of bytes added at a time to the amount of memory
57 allocated for the list of tabstops. */
58 #define TABLIST_BLOCK 256
60 /* A sentinel value that's placed at the end of the list of tab stops.
61 This value must be a large number, but not so large that adding the
62 length of a line to it would cause the column variable to overflow. */
63 #define TAB_STOP_SENTINEL INT_MAX
65 /* The name this program was run with. */
68 /* If nonzero, convert blanks even after nonblank characters have been
70 static int convert_entire_line;
72 /* If nonzero, the size of all tab stops. If zero, use `tab_list' instead. */
75 /* Array of the explicit column numbers of the tab stops;
76 after `tab_list' is exhausted, the rest of the line is printed
77 unchanged. The first column is column 0. */
80 /* The index of the first invalid element of `tab_list',
81 where the next element can be added. */
82 static int first_free_tab;
84 /* Null-terminated array of input filenames. */
85 static char **file_list;
87 /* Default for `file_list' if no files are given on the command line. */
88 static char *stdin_argv[] =
93 /* Nonzero if we have ever read standard input. */
94 static int have_read_stdin;
96 /* Status to return to the system. */
97 static int exit_status;
99 /* For long options that have no equivalent short option, use a
100 non-character as a pseudo short option, starting with CHAR_MAX + 1. */
103 CONVERT_FIRST_ONLY_OPTION = CHAR_MAX + 1
106 static struct option const longopts[] =
108 {"tabs", required_argument, NULL, 't'},
109 {"all", no_argument, NULL, 'a'},
110 {"first-only", no_argument, NULL, CONVERT_FIRST_ONLY_OPTION},
111 {GETOPT_HELP_OPTION_DECL},
112 {GETOPT_VERSION_OPTION_DECL},
116 /* Add tab stop TABVAL to the end of `tab_list', except
117 if TABVAL is -1, do nothing. */
120 add_tabstop (int tabval)
124 if (first_free_tab % TABLIST_BLOCK == 0)
125 tab_list = xrealloc (tab_list, first_free_tab + TABLIST_BLOCK);
126 tab_list[first_free_tab++] = tabval;
129 /* Add the comma or blank separated list of tabstops STOPS
130 to the list of tabstops. */
133 parse_tabstops (const char *stops)
137 for (; *stops; stops++)
139 if (*stops == ',' || ISBLANK (*stops))
141 add_tabstop (tabval);
144 else if (ISDIGIT (*stops))
148 tabval = tabval * 10 + *stops - '0';
151 error (EXIT_FAILURE, 0, _("tab size contains an invalid character"));
154 add_tabstop (tabval);
157 /* Check that the list of tabstops TABS, with ENTRIES entries,
158 contains only nonzero, ascending values. */
161 validate_tabstops (const int *tabs, int entries)
166 for (i = 0; i < entries; i++)
169 error (EXIT_FAILURE, 0, _("tab size cannot be 0"));
170 if (tabs[i] <= prev_tab)
171 error (EXIT_FAILURE, 0, _("tab sizes must be ascending"));
176 /* Close the old stream pointer FP if it is non-NULL,
177 and return a new one opened to read the next input file.
178 Open a filename of `-' as the standard input.
179 Return NULL if there are no more input files. */
184 static char *prev_file;
191 error (0, errno, "%s", prev_file);
195 clearerr (fp); /* Also clear EOF. */
196 else if (fclose (fp) == EOF)
198 error (0, errno, "%s", prev_file);
203 while ((file = *file_list++) != NULL)
205 if (file[0] == '-' && file[1] == '\0')
211 fp = fopen (file, "r");
217 error (0, errno, "%s", file);
223 /* Change spaces to tabs, writing to stdout.
224 Read each file in `file_list', in order. */
229 FILE *fp; /* Input stream. */
230 int c; /* Each input character. */
231 /* Index in `tab_list' of next tabstop: */
232 int tab_index = 0; /* For calculating width of pending tabs. */
233 int print_tab_index = 0; /* For printing as many tabs as possible. */
234 unsigned int column = 0; /* Column on screen of next char. */
235 int next_tab_column; /* Column the next tab stop is on. */
236 int convert = 1; /* If nonzero, perform translations. */
237 unsigned int pending = 0; /* Pending columns of blanks. */
240 fp = next_file ((FILE *) NULL);
244 /* Binary I/O will preserve the original EOL style (DOS/Unix) of files. */
245 SET_BINARY2 (fileno (fp), STDOUT_FILENO);
252 if (c == ' ' && convert && column < TAB_STOP_SENTINEL)
257 else if (c == '\t' && convert)
261 /* Do not let tab_index == first_free_tab;
262 stop when it is 1 less. */
263 while (tab_index < first_free_tab - 1
264 && column >= tab_list[tab_index])
266 next_tab_column = tab_list[tab_index];
267 if (tab_index < first_free_tab - 1)
269 if (column >= next_tab_column)
271 convert = 0; /* Ran out of tab stops. */
277 next_tab_column = column + tab_size - column % tab_size;
279 pending += next_tab_column - column;
280 column = next_tab_column;
285 /* Flush pending spaces. Print as many tabs as possible,
286 then print the rest as spaces. */
297 /* Do not let print_tab_index == first_free_tab;
298 stop when it is 1 less. */
299 while (print_tab_index < first_free_tab - 1
300 && column >= tab_list[print_tab_index])
302 next_tab_column = tab_list[print_tab_index];
303 if (print_tab_index < first_free_tab - 1)
308 next_tab_column = column + tab_size - column % tab_size;
310 if (next_tab_column - column <= pending)
313 pending -= next_tab_column - column;
314 column = next_tab_column;
333 break; /* No more files. */
336 SET_BINARY2 (fileno (fp), STDOUT_FILENO);
351 if (convert_entire_line == 0)
360 tab_index = print_tab_index = 0;
361 column = pending = 0;
372 fprintf (stderr, _("Try `%s --help' for more information.\n"),
377 Usage: %s [OPTION]... [FILE]...\n\
381 Convert spaces in each FILE to tabs, writing to standard output.\n\
382 With no FILE, or when FILE is -, read standard input.\n\
386 Mandatory arguments to long options are mandatory for short options too.\n\
389 -a, --all convert all whitespace, instead of just initial whitespace\n\
390 --first-only convert only leading sequences of whitespace (overrides -a)\n\
391 -t, --tabs=N have tabs N characters apart instead of 8 (enables -a)\n\
392 -t, --tabs=LIST use comma separated LIST of tab positions (enables -a)\n\
394 fputs (HELP_OPTION_DESCRIPTION, stdout);
395 fputs (VERSION_OPTION_DESCRIPTION, stdout);
396 printf (_("\nReport bugs to <%s>.\n"), PACKAGE_BUGREPORT);
398 exit (status == 0 ? EXIT_SUCCESS : EXIT_FAILURE);
402 main (int argc, char **argv)
404 int tabval = -1; /* Value of tabstop being read, or -1. */
405 int c; /* Option character. */
407 /* If nonzero, cancel the effect of any -a (explicit or implicit in -t),
408 so that only leading white space will be considered. */
409 int convert_first_only = 0;
411 bool obsolete_tablist = false;
413 initialize_main (&argc, &argv);
414 program_name = argv[0];
415 setlocale (LC_ALL, "");
416 bindtextdomain (PACKAGE, LOCALEDIR);
417 textdomain (PACKAGE);
419 atexit (close_stdout);
423 convert_entire_line = 0;
427 while ((c = getopt_long (argc, argv, ",0123456789at:", longopts, NULL))
436 usage (EXIT_FAILURE);
438 convert_entire_line = 1;
441 convert_entire_line = 1;
442 parse_tabstops (optarg);
444 case CONVERT_FIRST_ONLY_OPTION:
445 convert_first_only = 1;
448 add_tabstop (tabval);
450 obsolete_tablist = true;
452 case_GETOPT_HELP_CHAR;
453 case_GETOPT_VERSION_CHAR (PROGRAM_NAME, AUTHORS);
457 tabval = tabval * 10 + c - '0';
458 obsolete_tablist = true;
463 if (obsolete_tablist && 200112 <= posix2_version ())
466 _("`-LIST' option is obsolete; use `--first-only -t LIST'"));
467 usage (EXIT_FAILURE);
470 if (convert_first_only)
471 convert_entire_line = 0;
473 add_tabstop (tabval);
475 validate_tabstops (tab_list, first_free_tab);
477 if (first_free_tab == 0)
479 else if (first_free_tab == 1)
480 tab_size = tab_list[0];
483 /* Append a sentinel to the list of tab stop indices. */
484 add_tabstop (TAB_STOP_SENTINEL);
488 file_list = (optind < argc ? &argv[optind] : stdin_argv);
492 if (have_read_stdin && fclose (stdin) == EOF)
493 error (EXIT_FAILURE, errno, "-");
494 exit (exit_status == 0 ? EXIT_SUCCESS : EXIT_FAILURE);