xref: /freebsd/usr.bin/unexpand/unexpand.c (revision 9336e0699bda8a301cd2bfa37106b6ec5e32012e)
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. All advertising materials mentioning features or use of this software
14  *    must display the following acknowledgement:
15  *	This product includes software developed by the University of
16  *	California, Berkeley and its contributors.
17  * 4. Neither the name of the University nor the names of its contributors
18  *    may be used to endorse or promote products derived from this software
19  *    without specific prior written permission.
20  *
21  * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
22  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
23  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
24  * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
25  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
26  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
27  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
28  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
29  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
30  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
31  * SUCH DAMAGE.
32  */
33 
34 #include <sys/cdefs.h>
35 
36 __FBSDID("$FreeBSD$");
37 
38 #ifndef lint
39 static const char copyright[] =
40 "@(#) Copyright (c) 1980, 1993\n\
41 	The Regents of the University of California.  All rights reserved.\n";
42 #endif
43 
44 #ifndef lint
45 static const char sccsid[] = "@(#)unexpand.c	8.1 (Berkeley) 6/6/93";
46 #endif
47 
48 /*
49  * unexpand - put tabs into a file replacing blanks
50  */
51 #include <ctype.h>
52 #include <err.h>
53 #include <limits.h>
54 #include <locale.h>
55 #include <stdio.h>
56 #include <stdlib.h>
57 #include <string.h>
58 #include <unistd.h>
59 #include <wchar.h>
60 #include <wctype.h>
61 
62 int	all;
63 int	nstops;
64 int	tabstops[100];
65 
66 static void getstops(const char *);
67 static void usage(void);
68 static int tabify(const char *);
69 
70 int
71 main(int argc, char *argv[])
72 {
73 	int ch, failed;
74 	char *filename;
75 
76 	setlocale(LC_CTYPE, "");
77 
78 	nstops = 1;
79 	tabstops[0] = 8;
80 	while ((ch = getopt(argc, argv, "at:")) != -1) {
81 		switch (ch) {
82 		case 'a':	/* Un-expand all spaces, not just leading. */
83 			all = 1;
84 			break;
85 		case 't':	/* Specify tab list, implies -a. */
86 			getstops(optarg);
87 			all = 1;
88 			break;
89 		default:
90 			usage();
91 			/*NOTREACHED*/
92 		}
93 	}
94 	argc -= optind;
95 	argv += optind;
96 
97 	failed = 0;
98 	if (argc == 0)
99 		failed |= tabify("stdin");
100 	else {
101 		while ((filename = *argv++) != NULL) {
102 			if (freopen(filename, "r", stdin) == NULL) {
103 				warn("%s", filename);
104 				failed = 1;
105 			} else
106 				failed |= tabify(filename);
107 		}
108 	}
109 	exit(failed != 0);
110 }
111 
112 static void
113 usage(void)
114 {
115 	fprintf(stderr, "usage: unexpand [-a | -t tablist] [file ...]\n");
116 	exit(1);
117 }
118 
119 static int
120 tabify(const char *curfile)
121 {
122 	int dcol, doneline, limit, n, ocol, width;
123 	wint_t ch;
124 
125 	limit = nstops == 1 ? INT_MAX : tabstops[nstops - 1] - 1;
126 
127 	doneline = ocol = dcol = 0;
128 	while ((ch = getwchar()) != WEOF) {
129 		if (ch == ' ' && !doneline) {
130 			if (++dcol >= limit)
131 				doneline = 1;
132 			continue;
133 		} else if (ch == '\t') {
134 			if (nstops == 1) {
135 				dcol = (1 + dcol / tabstops[0]) *
136 				    tabstops[0];
137 				continue;
138 			} else {
139 				for (n = 0; tabstops[n] - 1 < dcol &&
140 				    n < nstops; n++)
141 					;
142 				if (n < nstops - 1 && tabstops[n] - 1 < limit) {
143 					dcol = tabstops[n];
144 					continue;
145 				}
146 				doneline = 1;
147 			}
148 		}
149 
150 		/* Output maximal number of tabs. */
151 		if (nstops == 1) {
152 			while (((ocol + tabstops[0]) / tabstops[0])
153 			    <= (dcol / tabstops[0])) {
154 				if (dcol - ocol < 2)
155 					break;
156 				putwchar('\t');
157 				ocol = (1 + ocol / tabstops[0]) *
158 				    tabstops[0];
159 			}
160 		} else {
161 			for (n = 0; tabstops[n] - 1 < ocol && n < nstops; n++)
162 				;
163 			while (ocol < dcol && n < nstops && ocol < limit) {
164 				putwchar('\t');
165 				ocol = tabstops[n++];
166 			}
167 		}
168 
169 		/* Then spaces. */
170 		while (ocol < dcol && ocol < limit) {
171 			putwchar(' ');
172 			ocol++;
173 		}
174 
175 		if (ch == '\b') {
176 			putwchar('\b');
177 			if (ocol > 0)
178 				ocol--, dcol--;
179 		} else if (ch == '\n') {
180 			putwchar('\n');
181 			doneline = ocol = dcol = 0;
182 			continue;
183 		} else if (ch != ' ' || dcol > limit) {
184 			putwchar(ch);
185 			if ((width = wcwidth(ch)) > 0)
186 				ocol += width, dcol += width;
187 		}
188 
189 		/*
190 		 * Only processing leading blanks or we've gone past the
191 		 * last tab stop. Emit remainder of this line unchanged.
192 		 */
193 		if (!all || dcol >= limit) {
194 			while ((ch = getwchar()) != '\n' && ch != WEOF)
195 				putwchar(ch);
196 			if (ch == '\n')
197 				putwchar('\n');
198 			doneline = ocol = dcol = 0;
199 		}
200 	}
201 	if (ferror(stdin)) {
202 		warn("%s", curfile);
203 		return (1);
204 	}
205 	return (0);
206 }
207 
208 static void
209 getstops(const char *cp)
210 {
211 	int i;
212 
213 	nstops = 0;
214 	for (;;) {
215 		i = 0;
216 		while (*cp >= '0' && *cp <= '9')
217 			i = i * 10 + *cp++ - '0';
218 		if (i <= 0)
219 			errx(1, "bad tab stop spec");
220 		if (nstops > 0 && i <= tabstops[nstops-1])
221 			errx(1, "bad tab stop spec");
222 		if (nstops == sizeof(tabstops) / sizeof(*tabstops))
223 			errx(1, "too many tab stops");
224 		tabstops[nstops++] = i;
225 		if (*cp == 0)
226 			break;
227 		if (*cp != ',' && !isblank((unsigned char)*cp))
228 			errx(1, "bad tab stop spec");
229 		cp++;
230 	}
231 }
232