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