1 /* unexpand - convert spaces to tabs
2 Copyright (C) 89, 91, 1995-2001 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
16 Foundation, 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>
48 /* The official name of this program (e.g., no `g' prefix). */
49 #define PROGRAM_NAME "unexpand"
51 #define AUTHORS "David MacKenzie"
53 /* The number of bytes added at a time to the amount of memory
54 allocated for the output line. */
55 #define OUTPUT_BLOCK 256
57 /* The number of bytes added at a time to the amount of memory
58 allocated for the list of tabstops. */
59 #define TABLIST_BLOCK 256
61 /* A sentinel value that's placed at the end of the list of tab stops.
62 This value must be a large number, but not so large that adding the
63 length of a line to it would cause the column variable to overflow. */
64 #define TAB_STOP_SENTINEL INT_MAX
66 /* The name this program was run with. */
69 /* If nonzero, convert blanks even after nonblank characters have been
71 static int convert_entire_line;
73 /* If nonzero, the size of all tab stops. If zero, use `tab_list' instead. */
76 /* Array of the explicit column numbers of the tab stops;
77 after `tab_list' is exhausted, the rest of the line is printed
78 unchanged. The first column is column 0. */
81 /* The index of the first invalid element of `tab_list',
82 where the next element can be added. */
83 static int first_free_tab;
85 /* Null-terminated array of input filenames. */
86 static char **file_list;
88 /* Default for `file_list' if no files are given on the command line. */
89 static char *stdin_argv[] =
94 /* Nonzero if we have ever read standard input. */
95 static int have_read_stdin;
97 /* Status to return to the system. */
98 static int exit_status;
100 /* For long options that have no equivalent short option, use a
101 non-character as a pseudo short option, starting with CHAR_MAX + 1. */
104 CONVERT_FIRST_ONLY_OPTION = CHAR_MAX + 1
107 static struct option const longopts[] =
109 {"tabs", required_argument, NULL, 't'},
110 {"all", no_argument, NULL, 'a'},
111 {"first-only", no_argument, NULL, CONVERT_FIRST_ONLY_OPTION},
112 {GETOPT_HELP_OPTION_DECL},
113 {GETOPT_VERSION_OPTION_DECL},
117 /* Add tab stop TABVAL to the end of `tab_list', except
118 if TABVAL is -1, do nothing. */
121 add_tabstop (int tabval)
125 if (first_free_tab % TABLIST_BLOCK == 0)
126 tab_list = (int *) xrealloc ((char *) tab_list,
127 first_free_tab + TABLIST_BLOCK);
128 tab_list[first_free_tab++] = tabval;
131 /* Add the comma or blank separated list of tabstops STOPS
132 to the list of tabstops. */
135 parse_tabstops (const char *stops)
139 for (; *stops; stops++)
141 if (*stops == ',' || ISBLANK (*stops))
143 add_tabstop (tabval);
146 else if (ISDIGIT (*stops))
150 tabval = tabval * 10 + *stops - '0';
153 error (EXIT_FAILURE, 0, _("tab size contains an invalid character"));
156 add_tabstop (tabval);
159 /* Check that the list of tabstops TABS, with ENTRIES entries,
160 contains only nonzero, ascending values. */
163 validate_tabstops (const int *tabs, int entries)
168 for (i = 0; i < entries; i++)
171 error (EXIT_FAILURE, 0, _("tab size cannot be 0"));
172 if (tabs[i] <= prev_tab)
173 error (EXIT_FAILURE, 0, _("tab sizes must be ascending"));
178 /* Close the old stream pointer FP if it is non-NULL,
179 and return a new one opened to read the next input file.
180 Open a filename of `-' as the standard input.
181 Return NULL if there are no more input files. */
186 static char *prev_file;
193 error (0, errno, "%s", prev_file);
197 clearerr (fp); /* Also clear EOF. */
198 else if (fclose (fp) == EOF)
200 error (0, errno, "%s", prev_file);
205 while ((file = *file_list++) != NULL)
207 if (file[0] == '-' && file[1] == '\0')
213 fp = fopen (file, "r");
219 error (0, errno, "%s", file);
225 /* Change spaces to tabs, writing to stdout.
226 Read each file in `file_list', in order. */
231 FILE *fp; /* Input stream. */
232 int c; /* Each input character. */
233 /* Index in `tab_list' of next tabstop: */
234 int tab_index = 0; /* For calculating width of pending tabs. */
235 int print_tab_index = 0; /* For printing as many tabs as possible. */
236 unsigned int column = 0; /* Column on screen of next char. */
237 int next_tab_column; /* Column the next tab stop is on. */
238 int convert = 1; /* If nonzero, perform translations. */
239 unsigned int pending = 0; /* Pending columns of blanks. */
241 fp = next_file ((FILE *) NULL);
245 /* Binary I/O will preserve the original EOL style (DOS/Unix) of files. */
246 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;
332 break; /* No more files. */
335 SET_BINARY2 (fileno (fp), STDOUT_FILENO);
350 if (convert_entire_line == 0)
359 tab_index = print_tab_index = 0;
360 column = pending = 0;
371 fprintf (stderr, _("Try `%s --help' for more information.\n"),
376 Usage: %s [OPTION]... [FILE]...\n\
380 Convert spaces in each FILE to tabs, writing to standard output.\n\
381 With no FILE, or when FILE is -, read standard input.\n\
383 Mandatory arguments to long options are mandatory for short options too.\n\
384 -a, --all convert all whitespace, instead of initial whitespace\n\
385 -t, --tabs=NUMBER have tabs NUMBER characters apart instead of 8\n\
386 -t, --tabs=LIST use comma separated list of explicit tab positions\n\
387 --help display this help and exit\n\
388 --version output version information and exit\n\
390 Instead of -t NUMBER or -t LIST, -NUMBER or -LIST may be used.\n\
392 puts (_("\nReport bugs to <bug-textutils@gnu.org>."));
394 exit (status == 0 ? EXIT_SUCCESS : EXIT_FAILURE);
398 main (int argc, char **argv)
400 int tabval = -1; /* Value of tabstop being read, or -1. */
401 int c; /* Option character. */
403 /* If nonzero, cancel the effect of any -a (explicit or implicit in -t),
404 so that only leading white space will be considered. */
405 int convert_first_only = 0;
407 program_name = argv[0];
408 setlocale (LC_ALL, "");
409 bindtextdomain (PACKAGE, LOCALEDIR);
410 textdomain (PACKAGE);
412 atexit (close_stdout);
416 convert_entire_line = 0;
420 while ((c = getopt_long (argc, argv, "at:,0123456789", longopts, NULL)) != -1)
430 convert_entire_line = 1;
433 convert_entire_line = 1;
434 parse_tabstops (optarg);
436 case CONVERT_FIRST_ONLY_OPTION:
437 convert_first_only = 1;
440 add_tabstop (tabval);
443 case_GETOPT_HELP_CHAR;
444 case_GETOPT_VERSION_CHAR (PROGRAM_NAME, AUTHORS);
448 tabval = tabval * 10 + c - '0';
453 if (convert_first_only)
454 convert_entire_line = 0;
456 add_tabstop (tabval);
458 validate_tabstops (tab_list, first_free_tab);
460 if (first_free_tab == 0)
462 else if (first_free_tab == 1)
463 tab_size = tab_list[0];
466 /* Append a sentinel to the list of tab stop indices. */
467 add_tabstop (TAB_STOP_SENTINEL);
472 file_list = stdin_argv;
474 file_list = &argv[optind];
478 if (have_read_stdin && fclose (stdin) == EOF)
479 error (EXIT_FAILURE, errno, "-");
480 exit (exit_status == 0 ? EXIT_SUCCESS : EXIT_FAILURE);