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
38 static const char copyright[] =
39 "@(#) Copyright (c) 1990, 1993, 1994\n\
40 The Regents of the University of California. All rights reserved.\n";
45 static char sccsid[] = "@(#)col.c 8.5 (Berkeley) 5/4/95";
49 #include <sys/cdefs.h>
50 __FBSDID("$FreeBSD$");
61 #define BS '\b' /* backspace */
62 #define TAB '\t' /* tab */
63 #define SPACE ' ' /* space */
64 #define NL '\n' /* newline */
65 #define CR '\r' /* carriage return */
66 #define ESC '\033' /* escape */
67 #define SI '\017' /* shift in to normal character set */
68 #define SO '\016' /* shift out to alternate character set */
69 #define VT '\013' /* vertical tab (aka reverse line feed) */
70 #define RLF '\007' /* ESC-07 reverse line feed */
71 #define RHLF '\010' /* ESC-010 reverse half-line feed */
72 #define FHLF '\011' /* ESC-011 forward half-line feed */
74 /* build up at least this many lines before flushing them out */
75 #define BUFFER_MARGIN 32
79 typedef struct char_str {
81 #define CS_ALTERNATE 2
82 short c_column; /* column character is in */
83 CSET c_set; /* character set (currently only 2) */
84 wchar_t c_char; /* character in question */
85 int c_width; /* character width */
88 typedef struct line_str LINE;
90 CHAR *l_line; /* characters on the line */
91 LINE *l_prev; /* previous line */
92 LINE *l_next; /* next line */
93 int l_lsize; /* allocated sizeof l_line */
94 int l_line_len; /* strlen(l_line) */
95 int l_needs_sort; /* set if chars went in out of order */
96 int l_max_col; /* max column in the line */
99 LINE *alloc_line(void);
101 void flush_line(LINE *);
102 void flush_lines(int);
103 void flush_blanks(void);
104 void free_line(LINE *);
107 CSET last_set; /* char_set of last char printed */
109 int compress_spaces; /* if doing space -> tab conversion */
110 int fine; /* if `fine' resolution (half lines) */
111 int max_bufd_lines; /* max # lines to keep in memory */
112 int nblank_lines; /* # blanks after last flushed line */
113 int no_backspaces; /* if not to output any backspaces */
114 int pass_unknown_seqs; /* pass unknown control sequences */
118 if (putwchar(ch) == WEOF) \
119 errx(1, "write error"); \
123 main(int argc, char **argv)
127 CSET cur_set; /* current character set */
128 LINE *l; /* current line */
129 int extra_lines; /* # of lines above first line */
130 int cur_col; /* current column */
131 int cur_line; /* line number of current position */
132 int max_line; /* max value of cur_line */
133 int this_line; /* line l points to */
134 int nflushd_lines; /* number of lines that were flushed */
135 int adjust, opt, warned, width;
137 (void)setlocale(LC_CTYPE, "");
139 max_bufd_lines = 128;
140 compress_spaces = 1; /* compress spaces into tabs */
141 while ((opt = getopt(argc, argv, "bfhl:px")) != -1)
143 case 'b': /* do not output backspaces */
146 case 'f': /* allow half forward line feeds */
149 case 'h': /* compress spaces into tabs */
152 case 'l': /* buffered line count */
153 if ((max_bufd_lines = atoi(optarg)) <= 0)
154 errx(1, "bad -l argument %s", optarg);
156 case 'p': /* pass unknown control sequences */
157 pass_unknown_seqs = 1;
159 case 'x': /* do not compress spaces into tabs */
170 /* this value is in half lines */
173 adjust = cur_col = extra_lines = warned = 0;
174 cur_line = max_line = nflushd_lines = this_line = 0;
175 cur_set = last_set = CS_NORMAL;
176 lines = l = alloc_line();
178 while ((ch = getwchar()) != WEOF) {
181 case BS: /* can't go back further */
189 case ESC: /* just ignore EOF */
199 if (cur_line > max_line)
205 if (cur_line > max_line)
216 cur_set = CS_ALTERNATE;
218 case TAB: /* adjust column */
227 if ((width = wcwidth(ch)) > 0)
231 if (!pass_unknown_seqs)
235 /* Must stuff ch in a line - are we at the right one? */
236 if (cur_line != this_line - adjust) {
241 nmove = cur_line - this_line;
243 /* round up to next line */
250 for (; nmove < 0 && l->l_prev; nmove++)
253 if (nflushd_lines == 0) {
255 * Allow backup past first
256 * line if nothing has been
259 for (; nmove < 0; nmove++) {
273 /* may need to allocate here */
274 for (; nmove > 0 && l->l_next; nmove--)
276 for (; nmove > 0; nmove--) {
283 this_line = cur_line + adjust;
284 nmove = this_line - nflushd_lines;
285 if (nmove >= max_bufd_lines + BUFFER_MARGIN) {
286 nflushd_lines += nmove - max_bufd_lines;
287 flush_lines(nmove - max_bufd_lines);
290 /* grow line's buffer? */
291 if (l->l_line_len + 1 >= l->l_lsize) {
294 need = l->l_lsize ? l->l_lsize * 2 : 90;
295 if ((l->l_line = realloc(l->l_line,
296 (unsigned)need * sizeof(CHAR))) == NULL)
297 err(1, (char *)NULL);
300 c = &l->l_line[l->l_line_len++];
303 c->c_column = cur_col;
304 c->c_width = wcwidth(ch);
306 * If things are put in out of order, they will need sorting
307 * when it is flushed.
309 if (cur_col < l->l_max_col)
312 l->l_max_col = cur_col;
314 cur_col += c->c_width;
319 exit(0); /* no lines, so just exit */
321 /* goto the last line that had a character on it */
322 for (; l->l_next; l = l->l_next)
324 flush_lines(this_line - nflushd_lines + extra_lines + 1);
326 /* make sure we leave things in a sane state */
327 if (last_set != CS_NORMAL)
330 /* flush out the last few blank lines */
331 nblank_lines = max_line - this_line;
334 else if (!nblank_lines)
335 /* missing a \n on the last line? */
342 flush_lines(int nflush)
346 while (--nflush >= 0) {
355 (void)free(l->l_line);
359 lines->l_prev = NULL;
363 * Print a number of newline/half newlines. If fine flag is set, nblank_lines
364 * is the number of half line feeds, otherwise it is the number of whole line
381 for (i = nb; --i >= 0;)
393 * Write a line to stdout taking care of space to tab conversion (-h flag)
394 * and character set shifts.
400 int i, j, nchars, last_col, save, this_col, tot;
403 nchars = l->l_line_len;
405 if (l->l_needs_sort) {
407 static int count_size, *count, sorted_size;
410 * Do an O(n) sort on l->l_line by column being careful to
411 * preserve the order of characters in the same column.
413 if (l->l_lsize > sorted_size) {
414 sorted_size = l->l_lsize;
415 if ((sorted = realloc(sorted,
416 (unsigned)sizeof(CHAR) * sorted_size)) == NULL)
417 err(1, (char *)NULL);
419 if (l->l_max_col >= count_size) {
420 count_size = l->l_max_col + 1;
421 if ((count = realloc(count,
422 (unsigned)sizeof(int) * count_size)) == NULL)
423 err(1, (char *)NULL);
425 memset(count, 0, sizeof(int) * l->l_max_col + 1);
426 for (i = nchars, c = l->l_line; --i >= 0; c++)
427 count[c->c_column]++;
430 * calculate running total (shifted down by 1) to use as
431 * indices into new line.
433 for (tot = 0, i = 0; i <= l->l_max_col; i++) {
439 for (i = nchars, c = l->l_line; --i >= 0; c++)
440 sorted[count[c->c_column]++] = *c;
445 this_col = c->c_column;
449 } while (--nchars > 0 && this_col == endc->c_column);
451 /* if -b only print last character */
455 this_col + c->c_width > endc->c_column)
459 if (this_col > last_col) {
460 int nspace = this_col - last_col;
462 if (compress_spaces && nspace > 1) {
464 int tab_col, tab_size;;
466 tab_col = (last_col + 8) & ~7;
467 if (tab_col > this_col)
469 tab_size = tab_col - last_col;
478 while (--nspace >= 0)
484 if (c->c_set != last_set) {
496 for (j = 0; j < c->c_width; j++)
501 last_col += (c - 1)->c_width;
507 static LINE *line_freelist;
515 if (!line_freelist) {
516 if ((l = realloc(NULL, sizeof(LINE) * NALLOC)) == NULL)
517 err(1, (char *)NULL);
519 for (i = 1; i < NALLOC; i++, l++)
524 line_freelist = l->l_next;
526 memset(l, 0, sizeof(LINE));
534 l->l_next = line_freelist;
542 (void)fprintf(stderr, "usage: col [-bfhpx] [-l nline]\n");
550 warnx("warning: can't back up %s",
551 line < 0 ? "past first line" : "-- line already flushed");