xref: /dragonfly/usr.bin/unexpand/unexpand.c (revision 9348a738)
1 /*-
2  * Copyright (c) 1980, 1993
3  *	The Regents of the University of California.  All rights reserved.
4  *
5  * Redistribution and use in source and binary forms, with or without
6  * modification, are permitted provided that the following conditions
7  * are met:
8  * 1. Redistributions of source code must retain the above copyright
9  *    notice, this list of conditions and the following disclaimer.
10  * 2. Redistributions in binary form must reproduce the above copyright
11  *    notice, this list of conditions and the following disclaimer in the
12  *    documentation and/or other materials provided with the distribution.
13  * 3. Neither the name of the University nor the names of its contributors
14  *    may be used to endorse or promote products derived from this software
15  *    without specific prior written permission.
16  *
17  * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
18  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
19  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
20  * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
21  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
22  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
23  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
24  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
25  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
26  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
27  * SUCH DAMAGE.
28  *
29  * @(#) Copyright (c) 1980, 1993 The Regents of the University of California.  All rights reserved.
30  * @(#)unexpand.c	8.1 (Berkeley) 6/6/93
31  * $FreeBSD: head/usr.bin/unexpand/unexpand.c 227192 2011-11-06 08:18:05Z ed $
32  */
33 
34 /*
35  * unexpand - put tabs into a file replacing blanks
36  */
37 #include <ctype.h>
38 #include <err.h>
39 #include <limits.h>
40 #include <locale.h>
41 #include <stdio.h>
42 #include <stdlib.h>
43 #include <string.h>
44 #include <unistd.h>
45 #include <wchar.h>
46 #include <wctype.h>
47 
48 static int	all;
49 static int	nstops;
50 static int	tabstops[100];
51 
52 static void getstops(const char *);
53 static void usage(void);
54 static int tabify(const char *);
55 
56 int
57 main(int argc, char *argv[])
58 {
59 	int ch, failed;
60 	char *filename;
61 
62 	setlocale(LC_CTYPE, "");
63 
64 	nstops = 1;
65 	tabstops[0] = 8;
66 	while ((ch = getopt(argc, argv, "at:")) != -1) {
67 		switch (ch) {
68 		case 'a':	/* Un-expand all spaces, not just leading. */
69 			all = 1;
70 			break;
71 		case 't':	/* Specify tab list, implies -a. */
72 			getstops(optarg);
73 			all = 1;
74 			break;
75 		default:
76 			usage();
77 			/*NOTREACHED*/
78 		}
79 	}
80 	argc -= optind;
81 	argv += optind;
82 
83 	failed = 0;
84 	if (argc == 0)
85 		failed |= tabify("stdin");
86 	else {
87 		while ((filename = *argv++) != NULL) {
88 			if (freopen(filename, "r", stdin) == NULL) {
89 				warn("%s", filename);
90 				failed = 1;
91 			} else
92 				failed |= tabify(filename);
93 		}
94 	}
95 	exit(failed != 0);
96 }
97 
98 static void
99 usage(void)
100 {
101 	fprintf(stderr, "usage: unexpand [-a | -t tablist] [file ...]\n");
102 	exit(1);
103 }
104 
105 static int
106 tabify(const char *curfile)
107 {
108 	int dcol, doneline, limit, n, ocol, width;
109 	wint_t ch;
110 
111 	limit = nstops == 1 ? INT_MAX : tabstops[nstops - 1] - 1;
112 
113 	doneline = ocol = dcol = 0;
114 	while ((ch = getwchar()) != WEOF) {
115 		if (ch == ' ' && !doneline) {
116 			if (++dcol >= limit)
117 				doneline = 1;
118 			continue;
119 		} else if (ch == '\t') {
120 			if (nstops == 1) {
121 				dcol = (1 + dcol / tabstops[0]) *
122 				    tabstops[0];
123 				continue;
124 			} else {
125 				for (n = 0; tabstops[n] - 1 < dcol &&
126 				    n < nstops; n++)
127 					;
128 				if (n < nstops - 1 && tabstops[n] - 1 < limit) {
129 					dcol = tabstops[n];
130 					continue;
131 				}
132 				doneline = 1;
133 			}
134 		}
135 
136 		/* Output maximal number of tabs. */
137 		if (nstops == 1) {
138 			while (((ocol + tabstops[0]) / tabstops[0])
139 			    <= (dcol / tabstops[0])) {
140 				if (dcol - ocol < 2)
141 					break;
142 				putwchar('\t');
143 				ocol = (1 + ocol / tabstops[0]) *
144 				    tabstops[0];
145 			}
146 		} else {
147 			for (n = 0; tabstops[n] - 1 < ocol && n < nstops; n++)
148 				;
149 			while (ocol < dcol && n < nstops && ocol < limit) {
150 				putwchar('\t');
151 				ocol = tabstops[n++];
152 			}
153 		}
154 
155 		/* Then spaces. */
156 		while (ocol < dcol && ocol < limit) {
157 			putwchar(' ');
158 			ocol++;
159 		}
160 
161 		if (ch == '\b') {
162 			putwchar('\b');
163 			if (ocol > 0)
164 				ocol--, dcol--;
165 		} else if (ch == '\n') {
166 			putwchar('\n');
167 			doneline = ocol = dcol = 0;
168 			continue;
169 		} else if (ch != ' ' || dcol > limit) {
170 			putwchar(ch);
171 			if ((width = wcwidth(ch)) > 0)
172 				ocol += width, dcol += width;
173 		}
174 
175 		/*
176 		 * Only processing leading blanks or we've gone past the
177 		 * last tab stop. Emit remainder of this line unchanged.
178 		 */
179 		if (!all || dcol >= limit) {
180 			while ((ch = getwchar()) != '\n' && ch != WEOF)
181 				putwchar(ch);
182 			if (ch == '\n')
183 				putwchar('\n');
184 			doneline = ocol = dcol = 0;
185 		}
186 	}
187 	if (ferror(stdin)) {
188 		warn("%s", curfile);
189 		return (1);
190 	}
191 	return (0);
192 }
193 
194 static void
195 getstops(const char *cp)
196 {
197 	int i;
198 
199 	nstops = 0;
200 	for (;;) {
201 		i = 0;
202 		while (*cp >= '0' && *cp <= '9')
203 			i = i * 10 + *cp++ - '0';
204 		if (i <= 0)
205 			errx(1, "bad tab stop spec");
206 		if (nstops > 0 && i <= tabstops[nstops-1])
207 			errx(1, "bad tab stop spec");
208 		if (nstops == sizeof(tabstops) / sizeof(*tabstops))
209 			errx(1, "too many tab stops");
210 		tabstops[nstops++] = i;
211 		if (*cp == 0)
212 			break;
213 		if (*cp != ',' && !isblank((unsigned char)*cp))
214 			errx(1, "bad tab stop spec");
215 		cp++;
216 	}
217 }
218