2 * Copyright (c) 1990, 1993, 1994
3 * The Regents of the University of California. All rights reserved.
5 * This code is derived from software contributed to Berkeley by
6 * Michael Rendell of the Memorial University of Newfoundland.
8 * Redistribution and use in source and binary forms, with or without
9 * modification, are permitted provided that the following conditions
11 * 1. Redistributions of source code must retain the above copyright
12 * notice, this list of conditions and the following disclaimer.
13 * 2. Redistributions in binary form must reproduce the above copyright
14 * notice, this list of conditions and the following disclaimer in the
15 * documentation and/or other materials provided with the distribution.
16 * 3. All advertising materials mentioning features or use of this software
17 * must display the following acknowledgement:
18 * This product includes software developed by the University of
19 * California, Berkeley and its contributors.
20 * 4. Neither the name of the University nor the names of its contributors
21 * may be used to endorse or promote products derived from this software
22 * without specific prior written permission.
24 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
25 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
26 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
27 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
28 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
29 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
30 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
31 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
32 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
33 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
36 * @(#) Copyright (c) 1990, 1993, 1994 The Regents of the University of California. All rights reserved.
37 * @(#)col.c 8.5 (Berkeley) 5/4/95
38 * $FreeBSD: src/usr.bin/col/col.c,v 1.6.6.4 2001/08/02 01:27:12 obrien Exp $
39 * $DragonFly: src/usr.bin/col/col.c,v 1.2 2003/06/17 04:29:25 dillon Exp $
50 #define BS '\b' /* backspace */
51 #define TAB '\t' /* tab */
52 #define SPACE ' ' /* space */
53 #define NL '\n' /* newline */
54 #define CR '\r' /* carriage return */
55 #define ESC '\033' /* escape */
56 #define SI '\017' /* shift in to normal character set */
57 #define SO '\016' /* shift out to alternate character set */
58 #define VT '\013' /* vertical tab (aka reverse line feed) */
59 #define RLF '\007' /* ESC-07 reverse line feed */
60 #define RHLF '\010' /* ESC-010 reverse half-line feed */
61 #define FHLF '\011' /* ESC-011 forward half-line feed */
63 /* build up at least this many lines before flushing them out */
64 #define BUFFER_MARGIN 32
68 typedef struct char_str {
70 #define CS_ALTERNATE 2
71 short c_column; /* column character is in */
72 CSET c_set; /* character set (currently only 2) */
73 char c_char; /* character in question */
76 typedef struct line_str LINE;
78 CHAR *l_line; /* characters on the line */
79 LINE *l_prev; /* previous line */
80 LINE *l_next; /* next line */
81 int l_lsize; /* allocated sizeof l_line */
82 int l_line_len; /* strlen(l_line) */
83 int l_needs_sort; /* set if chars went in out of order */
84 int l_max_col; /* max column in the line */
87 LINE *alloc_line __P((void));
88 void dowarn __P((int));
89 void flush_line __P((LINE *));
90 void flush_lines __P((int));
91 void flush_blanks __P((void));
92 void free_line __P((LINE *));
93 int main __P((int, char **));
94 void usage __P((void));
96 CSET last_set; /* char_set of last char printed */
98 int compress_spaces; /* if doing space -> tab conversion */
99 int fine; /* if `fine' resolution (half lines) */
100 int max_bufd_lines; /* max # lines to keep in memory */
101 int nblank_lines; /* # blanks after last flushed line */
102 int no_backspaces; /* if not to output any backspaces */
103 int pass_unknown_seqs; /* pass unknown control sequences */
107 if (putchar(ch) == EOF) \
108 errx(1, "write error"); \
118 CSET cur_set; /* current character set */
119 LINE *l; /* current line */
120 int extra_lines; /* # of lines above first line */
121 int cur_col; /* current column */
122 int cur_line; /* line number of current position */
123 int max_line; /* max value of cur_line */
124 int this_line; /* line l points to */
125 int nflushd_lines; /* number of lines that were flushed */
126 int adjust, opt, warned;
128 (void)setlocale(LC_CTYPE, "");
130 max_bufd_lines = 128;
131 compress_spaces = 1; /* compress spaces into tabs */
132 while ((opt = getopt(argc, argv, "bfhl:px")) != -1)
134 case 'b': /* do not output backspaces */
137 case 'f': /* allow half forward line feeds */
140 case 'h': /* compress spaces into tabs */
143 case 'l': /* buffered line count */
144 if ((max_bufd_lines = atoi(optarg)) <= 0)
145 errx(1, "bad -l argument %s", optarg);
147 case 'p': /* pass unknown control sequences */
148 pass_unknown_seqs = 1;
150 case 'x': /* do not compress spaces into tabs */
161 /* this value is in half lines */
164 adjust = cur_col = extra_lines = warned = 0;
165 cur_line = max_line = nflushd_lines = this_line = 0;
166 cur_set = last_set = CS_NORMAL;
167 lines = l = alloc_line();
169 while ((ch = getchar()) != EOF) {
172 case BS: /* can't go back further */
180 case ESC: /* just ignore EOF */
190 if (cur_line > max_line)
196 if (cur_line > max_line)
207 cur_set = CS_ALTERNATE;
209 case TAB: /* adjust column */
217 if (!pass_unknown_seqs)
221 /* Must stuff ch in a line - are we at the right one? */
222 if (cur_line != this_line - adjust) {
227 nmove = cur_line - this_line;
229 /* round up to next line */
236 for (; nmove < 0 && l->l_prev; nmove++)
239 if (nflushd_lines == 0) {
241 * Allow backup past first
242 * line if nothing has been
245 for (; nmove < 0; nmove++) {
259 /* may need to allocate here */
260 for (; nmove > 0 && l->l_next; nmove--)
262 for (; nmove > 0; nmove--) {
269 this_line = cur_line + adjust;
270 nmove = this_line - nflushd_lines;
271 if (nmove >= max_bufd_lines + BUFFER_MARGIN) {
272 nflushd_lines += nmove - max_bufd_lines;
273 flush_lines(nmove - max_bufd_lines);
276 /* grow line's buffer? */
277 if (l->l_line_len + 1 >= l->l_lsize) {
280 need = l->l_lsize ? l->l_lsize * 2 : 90;
281 if ((l->l_line = realloc(l->l_line,
282 (unsigned)need * sizeof(CHAR))) == NULL)
283 err(1, (char *)NULL);
286 c = &l->l_line[l->l_line_len++];
289 c->c_column = cur_col;
291 * If things are put in out of order, they will need sorting
292 * when it is flushed.
294 if (cur_col < l->l_max_col)
297 l->l_max_col = cur_col;
301 exit(0); /* no lines, so just exit */
303 /* goto the last line that had a character on it */
304 for (; l->l_next; l = l->l_next)
306 flush_lines(this_line - nflushd_lines + extra_lines + 1);
308 /* make sure we leave things in a sane state */
309 if (last_set != CS_NORMAL)
312 /* flush out the last few blank lines */
313 nblank_lines = max_line - this_line;
316 else if (!nblank_lines)
317 /* missing a \n on the last line? */
329 while (--nflush >= 0) {
338 (void)free(l->l_line);
342 lines->l_prev = NULL;
346 * Print a number of newline/half newlines. If fine flag is set, nblank_lines
347 * is the number of half line feeds, otherwise it is the number of whole line
364 for (i = nb; --i >= 0;)
376 * Write a line to stdout taking care of space to tab conversion (-h flag)
377 * and character set shifts.
384 int nchars, last_col, this_col;
387 nchars = l->l_line_len;
389 if (l->l_needs_sort) {
391 static int count_size, *count, i, save, sorted_size, tot;
394 * Do an O(n) sort on l->l_line by column being careful to
395 * preserve the order of characters in the same column.
397 if (l->l_lsize > sorted_size) {
398 sorted_size = l->l_lsize;
399 if ((sorted = realloc(sorted,
400 (unsigned)sizeof(CHAR) * sorted_size)) == NULL)
401 err(1, (char *)NULL);
403 if (l->l_max_col >= count_size) {
404 count_size = l->l_max_col + 1;
405 if ((count = realloc(count,
406 (unsigned)sizeof(int) * count_size)) == NULL)
407 err(1, (char *)NULL);
409 memset(count, 0, sizeof(int) * l->l_max_col + 1);
410 for (i = nchars, c = l->l_line; --i >= 0; c++)
411 count[c->c_column]++;
414 * calculate running total (shifted down by 1) to use as
415 * indices into new line.
417 for (tot = 0, i = 0; i <= l->l_max_col; i++) {
423 for (i = nchars, c = l->l_line; --i >= 0; c++)
424 sorted[count[c->c_column]++] = *c;
429 this_col = c->c_column;
433 } while (--nchars > 0 && this_col == endc->c_column);
435 /* if -b only print last character */
439 if (this_col > last_col) {
440 int nspace = this_col - last_col;
442 if (compress_spaces && nspace > 1) {
444 int tab_col, tab_size;;
446 tab_col = (last_col + 8) & ~7;
447 if (tab_col > this_col)
449 tab_size = tab_col - last_col;
458 while (--nspace >= 0)
465 if (c->c_set != last_set) {
485 static LINE *line_freelist;
493 if (!line_freelist) {
494 if ((l = realloc(NULL, sizeof(LINE) * NALLOC)) == NULL)
495 err(1, (char *)NULL);
497 for (i = 1; i < NALLOC; i++, l++)
502 line_freelist = l->l_next;
504 memset(l, 0, sizeof(LINE));
513 l->l_next = line_freelist;
521 (void)fprintf(stderr, "usage: col [-bfhpx] [-l nline]\n");
530 warnx("warning: can't back up %s",
531 line < 0 ? "past first line" : "-- line already flushed");