xref: /freebsd/lib/libfigpar/string_m.c (revision 8df8b2d3e51d1b816201d8a1fe8bc29fe192e562)
1 /*-
2  * Copyright (c) 2001-2014 Devin Teske <dteske@FreeBSD.org>
3  * 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  *
14  * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
15  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
16  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
17  * ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
18  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
19  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
20  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
21  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
22  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
23  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
24  * SUCH DAMAGE.
25  */
26 
27 #include <sys/cdefs.h>
28 __FBSDID("$FreeBSD$");
29 
30 #include <sys/types.h>
31 
32 #include <ctype.h>
33 #include <errno.h>
34 #include <stdio.h>
35 #include <stdlib.h>
36 #include <string.h>
37 
38 #include "string_m.h"
39 
40 /*
41  * Counts the number of occurrences of one string that appear in the source
42  * string. Return value is the total count.
43  *
44  * An example use would be if you need to know how large a block of memory
45  * needs to be for a replaceall() series.
46  */
47 unsigned int
48 strcount(const char *source, const char *find)
49 {
50 	const char *p = source;
51 	size_t flen;
52 	unsigned int n = 0;
53 
54 	/* Both parameters are required */
55 	if (source == NULL || find == NULL)
56 		return (0);
57 
58 	/* Cache the length of find element */
59 	flen = strlen(find);
60 	if (strlen(source) == 0 || flen == 0)
61 		return (0);
62 
63 	/* Loop until the end of the string */
64 	while (*p != '\0') {
65 		if (strncmp(p, find, flen) == 0) { /* found an instance */
66 			p += flen;
67 			n++;
68 		} else
69 			p++;
70 	}
71 
72 	return (n);
73 }
74 
75 /*
76  * Replaces all occurrences of `find' in `source' with `replace'.
77  *
78  * You should not pass a string constant as the first parameter, it needs to be
79  * a pointer to an allocated block of memory. The block of memory that source
80  * points to should be large enough to hold the result. If the length of the
81  * replacement string is greater than the length of the find string, the result
82  * will be larger than the original source string. To allocate enough space for
83  * the result, use the function strcount() declared above to determine the
84  * number of occurrences and how much larger the block size needs to be.
85  *
86  * If source is not large enough, the application will crash. The return value
87  * is the length (in bytes) of the result.
88  *
89  * When an error occurs, -1 is returned and the global variable errno is set
90  * accordingly. Returns zero on success.
91  */
92 int
93 replaceall(char *source, const char *find, const char *replace)
94 {
95 	char *p;
96 	char *t;
97 	char *temp;
98 	size_t flen;
99 	size_t rlen;
100 	size_t slen;
101 	uint32_t n = 0;
102 
103 	errno = 0; /* reset global error number */
104 
105 	/* Check that we have non-null parameters */
106 	if (source == NULL)
107 		return (0);
108 	if (find == NULL)
109 		return (strlen(source));
110 
111 	/* Cache the length of the strings */
112 	slen = strlen(source);
113 	flen = strlen(find);
114 	rlen = replace ? strlen(replace) : 0;
115 
116 	/* Cases where no replacements need to be made */
117 	if (slen == 0 || flen == 0 || slen < flen)
118 		return (slen);
119 
120 	/* If replace is longer than find, we'll need to create a temp copy */
121 	if (rlen > flen) {
122 		temp = malloc(slen + 1);
123 		if (temp == NULL) /* could not allocate memory */
124 			return (-1);
125 		memcpy(temp, source, slen + 1);
126 	} else
127 		temp = source;
128 
129 	/* Reconstruct the string with the replacements */
130 	p = source; t = temp; /* position elements */
131 
132 	while (*t != '\0') {
133 		if (strncmp(t, find, flen) == 0) {
134 			/* found an occurrence */
135 			for (n = 0; replace && replace[n]; n++)
136 				*p++ = replace[n];
137 			t += flen;
138 		} else
139 			*p++ = *t++; /* copy character and increment */
140 	}
141 
142 	/* Terminate the string */
143 	*p = '\0';
144 
145 	/* Free the temporary allocated memory */
146 	if (temp != source)
147 		free(temp);
148 
149 	/* Return the length of the completed string */
150 	return (strlen(source));
151 }
152 
153 /*
154  * Expands escape sequences in a buffer pointed to by `source'. This function
155  * steps through each character, and converts escape sequences such as "\n",
156  * "\r", "\t" and others into their respective meanings.
157  *
158  * You should not pass a string constant or literal to this function or the
159  * program will likely segmentation fault when it tries to modify the data.
160  *
161  * The string length will either shorten or stay the same depending on whether
162  * any escape sequences were converted but the amount of memory allocated does
163  * not change.
164  *
165  * Interpreted sequences are:
166  *
167  * 	\0NNN	character with octal value NNN (0 to 3 digits)
168  * 	\N	character with octal value N (0 thru 7)
169  * 	\a	alert (BEL)
170  * 	\b	backslash
171  * 	\f	form feed
172  * 	\n	new line
173  * 	\r	carriage return
174  * 	\t	horizontal tab
175  * 	\v	vertical tab
176  * 	\xNN	byte with hexadecimal value NN (1 to 2 digits)
177  *
178  * All other sequences are unescaped (ie. '\"' and '\#').
179  */
180 void strexpand(char *source)
181 {
182 	uint8_t c;
183 	char *chr;
184 	char *pos;
185 	char d[4];
186 
187 	/* Initialize position elements */
188 	pos = chr = source;
189 
190 	/* Loop until we hit the end of the string */
191 	while (*pos != '\0') {
192 		if (*chr != '\\') {
193 			*pos = *chr; /* copy character to current offset */
194 			pos++;
195 			chr++;
196 			continue;
197 		}
198 
199 		/* Replace the backslash with the correct character */
200 		switch (*++chr) {
201 		case 'a': *pos = '\a'; break; /* bell/alert (BEL) */
202 		case 'b': *pos = '\b'; break; /* backspace */
203 		case 'f': *pos = '\f'; break; /* form feed */
204 		case 'n': *pos = '\n'; break; /* new line */
205 		case 'r': *pos = '\r'; break; /* carriage return */
206 		case 't': *pos = '\t'; break; /* horizontal tab */
207 		case 'v': *pos = '\v'; break; /* vertical tab */
208 		case 'x': /* hex value (1 to 2 digits)(\xNN) */
209 			d[2] = '\0'; /* pre-terminate the string */
210 
211 			/* verify next two characters are hex */
212 			d[0] = isxdigit(*(chr+1)) ? *++chr : '\0';
213 			if (d[0] != '\0')
214 				d[1] = isxdigit(*(chr+1)) ? *++chr : '\0';
215 
216 			/* convert the characters to decimal */
217 			c = (uint8_t)strtoul(d, 0, 16);
218 
219 			/* assign the converted value */
220 			*pos = (c != 0 || d[0] == '0') ? c : *++chr;
221 			break;
222 		case '0': /* octal value (0 to 3 digits)(\0NNN) */
223 			d[3] = '\0'; /* pre-terminate the string */
224 
225 			/* verify next three characters are octal */
226 			d[0] = (isdigit(*(chr+1)) && *(chr+1) < '8') ?
227 			    *++chr : '\0';
228 			if (d[0] != '\0')
229 				d[1] = (isdigit(*(chr+1)) && *(chr+1) < '8') ?
230 				    *++chr : '\0';
231 			if (d[1] != '\0')
232 				d[2] = (isdigit(*(chr+1)) && *(chr+1) < '8') ?
233 				    *++chr : '\0';
234 
235 			/* convert the characters to decimal */
236 			c = (uint8_t)strtoul(d, 0, 8);
237 
238 			/* assign the converted value */
239 			*pos = c;
240 			break;
241 		default: /* single octal (\0..7) or unknown sequence */
242 			if (isdigit(*chr) && *chr < '8') {
243 				d[0] = *chr;
244 				d[1] = '\0';
245 				*pos = (uint8_t)strtoul(d, 0, 8);
246 			} else
247 				*pos = *chr;
248 		}
249 
250 		/* Increment to next offset, possible next escape sequence */
251 		pos++;
252 		chr++;
253 	}
254 }
255 
256 /*
257  * Expand only the escaped newlines in a buffer pointed to by `source'. This
258  * function steps through each character, and converts the "\n" sequence into
259  * a literal newline and the "\\n" sequence into "\n".
260  *
261  * You should not pass a string constant or literal to this function or the
262  * program will likely segmentation fault when it tries to modify the data.
263  *
264  * The string length will either shorten or stay the same depending on whether
265  * any escaped newlines were converted but the amount of memory allocated does
266  * not change.
267  */
268 void strexpandnl(char *source)
269 {
270 	uint8_t backslash = 0;
271 	char *cp1;
272 	char *cp2;
273 
274 	/* Replace '\n' with literal in dprompt */
275 	cp1 = cp2 = source;
276 	while (*cp2 != '\0') {
277 		*cp1 = *cp2;
278 		if (*cp2 == '\\')
279 			backslash++;
280 		else if (*cp2 != 'n')
281 			backslash = 0;
282 		else if (backslash > 0) {
283 			*(--cp1) = (backslash & 1) == 1 ? '\n' : 'n';
284 			backslash = 0;
285 		}
286 		cp1++;
287 		cp2++;
288 	}
289 	*cp1 = *cp2;
290 }
291 
292 /*
293  * Convert a string to lower case. You should not pass a string constant to
294  * this function. Only pass pointers to allocated memory with null terminated
295  * string data.
296  */
297 void
298 strtolower(char *source)
299 {
300 	char *p = source;
301 
302 	if (source == NULL)
303 		return;
304 
305 	while (*p != '\0') {
306 		*p = tolower(*p);
307 		p++; /* would have just used `*p++' but gcc 3.x warns */
308 	}
309 }
310