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 * 4. Neither the name of the University nor the names of its contributors
17 * may be used to endorse or promote products derived from this software
18 * without specific prior written permission.
20 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
21 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
22 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
23 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
24 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
25 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
26 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
27 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
28 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
29 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
34 static const char copyright[] =
35 "@(#) Copyright (c) 1990, 1993, 1994\n\
36 The Regents of the University of California. All rights reserved.\n";
41 static char sccsid[] = "@(#)col.c 8.5 (Berkeley) 5/4/95";
45 #include <sys/cdefs.h>
46 __FBSDID("$FreeBSD$");
57 #define BS '\b' /* backspace */
58 #define TAB '\t' /* tab */
59 #define SPACE ' ' /* space */
60 #define NL '\n' /* newline */
61 #define CR '\r' /* carriage return */
62 #define ESC '\033' /* escape */
63 #define SI '\017' /* shift in to normal character set */
64 #define SO '\016' /* shift out to alternate character set */
65 #define VT '\013' /* vertical tab (aka reverse line feed) */
66 #define RLF '\007' /* ESC-07 reverse line feed */
67 #define RHLF '\010' /* ESC-010 reverse half-line feed */
68 #define FHLF '\011' /* ESC-011 forward half-line feed */
70 /* build up at least this many lines before flushing them out */
71 #define BUFFER_MARGIN 32
75 typedef struct char_str {
77 #define CS_ALTERNATE 2
78 short c_column; /* column character is in */
79 CSET c_set; /* character set (currently only 2) */
80 wchar_t c_char; /* character in question */
81 int c_width; /* character width */
84 typedef struct line_str LINE;
86 CHAR *l_line; /* characters on the line */
87 LINE *l_prev; /* previous line */
88 LINE *l_next; /* next line */
89 int l_lsize; /* allocated sizeof l_line */
90 int l_line_len; /* strlen(l_line) */
91 int l_needs_sort; /* set if chars went in out of order */
92 int l_max_col; /* max column in the line */
95 LINE *alloc_line(void);
97 void flush_line(LINE *);
98 void flush_lines(int);
99 void flush_blanks(void);
100 void free_line(LINE *);
103 CSET last_set; /* char_set of last char printed */
105 int compress_spaces; /* if doing space -> tab conversion */
106 int fine; /* if `fine' resolution (half lines) */
107 int max_bufd_lines; /* max # lines to keep in memory */
108 int nblank_lines; /* # blanks after last flushed line */
109 int no_backspaces; /* if not to output any backspaces */
110 int pass_unknown_seqs; /* pass unknown control sequences */
114 if (putwchar(ch) == WEOF) \
115 errx(1, "write error"); \
119 main(int argc, char **argv)
123 CSET cur_set; /* current character set */
124 LINE *l; /* current line */
125 int extra_lines; /* # of lines above first line */
126 int cur_col; /* current column */
127 int cur_line; /* line number of current position */
128 int max_line; /* max value of cur_line */
129 int this_line; /* line l points to */
130 int nflushd_lines; /* number of lines that were flushed */
131 int adjust, opt, warned, width;
133 (void)setlocale(LC_CTYPE, "");
135 max_bufd_lines = 128;
136 compress_spaces = 1; /* compress spaces into tabs */
137 while ((opt = getopt(argc, argv, "bfhl:px")) != -1)
139 case 'b': /* do not output backspaces */
142 case 'f': /* allow half forward line feeds */
145 case 'h': /* compress spaces into tabs */
148 case 'l': /* buffered line count */
149 if ((max_bufd_lines = atoi(optarg)) <= 0)
150 errx(1, "bad -l argument %s", optarg);
152 case 'p': /* pass unknown control sequences */
153 pass_unknown_seqs = 1;
155 case 'x': /* do not compress spaces into tabs */
166 /* this value is in half lines */
169 adjust = cur_col = extra_lines = warned = 0;
170 cur_line = max_line = nflushd_lines = this_line = 0;
171 cur_set = last_set = CS_NORMAL;
172 lines = l = alloc_line();
174 while ((ch = getwchar()) != WEOF) {
177 case BS: /* can't go back further */
185 case ESC: /* just ignore EOF */
195 if (cur_line > max_line)
201 if (cur_line > max_line)
212 cur_set = CS_ALTERNATE;
214 case TAB: /* adjust column */
223 if ((width = wcwidth(ch)) > 0)
227 if (!pass_unknown_seqs)
231 /* Must stuff ch in a line - are we at the right one? */
232 if (cur_line != this_line - adjust) {
237 nmove = cur_line - this_line;
239 /* round up to next line */
246 for (; nmove < 0 && l->l_prev; nmove++)
249 if (nflushd_lines == 0) {
251 * Allow backup past first
252 * line if nothing has been
255 for (; nmove < 0; nmove++) {
269 /* may need to allocate here */
270 for (; nmove > 0 && l->l_next; nmove--)
272 for (; nmove > 0; nmove--) {
279 this_line = cur_line + adjust;
280 nmove = this_line - nflushd_lines;
281 if (nmove >= max_bufd_lines + BUFFER_MARGIN) {
282 nflushd_lines += nmove - max_bufd_lines;
283 flush_lines(nmove - max_bufd_lines);
286 /* grow line's buffer? */
287 if (l->l_line_len + 1 >= l->l_lsize) {
290 need = l->l_lsize ? l->l_lsize * 2 : 90;
291 if ((l->l_line = realloc(l->l_line,
292 (unsigned)need * sizeof(CHAR))) == NULL)
293 err(1, (char *)NULL);
296 c = &l->l_line[l->l_line_len++];
299 c->c_column = cur_col;
300 c->c_width = wcwidth(ch);
302 * If things are put in out of order, they will need sorting
303 * when it is flushed.
305 if (cur_col < l->l_max_col)
308 l->l_max_col = cur_col;
310 cur_col += c->c_width;
315 exit(0); /* no lines, so just exit */
317 /* goto the last line that had a character on it */
318 for (; l->l_next; l = l->l_next)
320 flush_lines(this_line - nflushd_lines + extra_lines + 1);
322 /* make sure we leave things in a sane state */
323 if (last_set != CS_NORMAL)
326 /* flush out the last few blank lines */
327 nblank_lines = max_line - this_line;
330 else if (!nblank_lines)
331 /* missing a \n on the last line? */
338 flush_lines(int nflush)
342 while (--nflush >= 0) {
351 (void)free(l->l_line);
355 lines->l_prev = NULL;
359 * Print a number of newline/half newlines. If fine flag is set, nblank_lines
360 * is the number of half line feeds, otherwise it is the number of whole line
377 for (i = nb; --i >= 0;)
389 * Write a line to stdout taking care of space to tab conversion (-h flag)
390 * and character set shifts.
396 int i, j, nchars, last_col, save, this_col, tot;
399 nchars = l->l_line_len;
401 if (l->l_needs_sort) {
403 static int count_size, *count, sorted_size;
406 * Do an O(n) sort on l->l_line by column being careful to
407 * preserve the order of characters in the same column.
409 if (l->l_lsize > sorted_size) {
410 sorted_size = l->l_lsize;
411 if ((sorted = realloc(sorted,
412 (unsigned)sizeof(CHAR) * sorted_size)) == NULL)
413 err(1, (char *)NULL);
415 if (l->l_max_col >= count_size) {
416 count_size = l->l_max_col + 1;
417 if ((count = realloc(count,
418 (unsigned)sizeof(int) * count_size)) == NULL)
419 err(1, (char *)NULL);
421 memset(count, 0, sizeof(int) * l->l_max_col + 1);
422 for (i = nchars, c = l->l_line; --i >= 0; c++)
423 count[c->c_column]++;
426 * calculate running total (shifted down by 1) to use as
427 * indices into new line.
429 for (tot = 0, i = 0; i <= l->l_max_col; i++) {
435 for (i = nchars, c = l->l_line; --i >= 0; c++)
436 sorted[count[c->c_column]++] = *c;
441 this_col = c->c_column;
445 } while (--nchars > 0 && this_col == endc->c_column);
447 /* if -b only print last character */
451 this_col + c->c_width > endc->c_column)
455 if (this_col > last_col) {
456 int nspace = this_col - last_col;
458 if (compress_spaces && nspace > 1) {
460 int tab_col, tab_size;
462 tab_col = (last_col + 8) & ~7;
463 if (tab_col > this_col)
465 tab_size = tab_col - last_col;
474 while (--nspace >= 0)
480 if (c->c_set != last_set) {
492 for (j = 0; j < c->c_width; j++)
497 last_col += (c - 1)->c_width;
503 static LINE *line_freelist;
511 if (!line_freelist) {
512 if ((l = realloc(NULL, sizeof(LINE) * NALLOC)) == NULL)
513 err(1, (char *)NULL);
515 for (i = 1; i < NALLOC; i++, l++)
520 line_freelist = l->l_next;
522 memset(l, 0, sizeof(LINE));
530 l->l_next = line_freelist;
538 (void)fprintf(stderr, "usage: col [-bfhpx] [-l nline]\n");
546 warnx("warning: can't back up %s",
547 line < 0 ? "past first line" : "-- line already flushed");