| 1 | /* -*-c-*- |
| 2 | * |
| 3 | * $Id: str.c,v 1.6 2004/04/08 01:36:13 mdw Exp $ |
| 4 | * |
| 5 | * Functions for hacking with strings |
| 6 | * |
| 7 | * (c) 1999 Straylight/Edgeware |
| 8 | */ |
| 9 | |
| 10 | /*----- Licensing notice --------------------------------------------------* |
| 11 | * |
| 12 | * This file is part of the mLib utilities library. |
| 13 | * |
| 14 | * mLib is free software; you can redistribute it and/or modify |
| 15 | * it under the terms of the GNU Library General Public License as |
| 16 | * published by the Free Software Foundation; either version 2 of the |
| 17 | * License, or (at your option) any later version. |
| 18 | * |
| 19 | * mLib is distributed in the hope that it will be useful, |
| 20 | * but WITHOUT ANY WARRANTY; without even the implied warranty of |
| 21 | * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
| 22 | * GNU Library General Public License for more details. |
| 23 | * |
| 24 | * You should have received a copy of the GNU Library General Public |
| 25 | * License along with mLib; if not, write to the Free |
| 26 | * Software Foundation, Inc., 59 Temple Place - Suite 330, Boston, |
| 27 | * MA 02111-1307, USA. |
| 28 | */ |
| 29 | |
| 30 | /*----- Header files ------------------------------------------------------*/ |
| 31 | |
| 32 | #include <ctype.h> |
| 33 | #include <stdio.h> |
| 34 | #include <stdlib.h> |
| 35 | #include <string.h> |
| 36 | |
| 37 | #include "str.h" |
| 38 | |
| 39 | /*----- Main code ---------------------------------------------------------*/ |
| 40 | |
| 41 | /* --- @str_qword@ --- * |
| 42 | * |
| 43 | * Arguments: @char **pp@ = address of pointer into string |
| 44 | * @unsigned f@ = various flags |
| 45 | * |
| 46 | * Returns: Pointer to the next space-separated possibly-quoted word from |
| 47 | * the string, or null. |
| 48 | * |
| 49 | * Use: Fetches the next word from a string. If the flag |
| 50 | * @STRF_QUOTE@ is set, the `\' character acts as an escape, and |
| 51 | * single and double quotes protect whitespace. |
| 52 | */ |
| 53 | |
| 54 | char *str_qword(char **pp, unsigned f) |
| 55 | { |
| 56 | char *p = *pp, *q, *qq; |
| 57 | int st = 0, pst = 0; |
| 58 | |
| 59 | /* --- Preliminaries --- */ |
| 60 | |
| 61 | if (!p) |
| 62 | return (0); |
| 63 | while (isspace((unsigned char)*p)) |
| 64 | p++; |
| 65 | if (!*p) { |
| 66 | *pp = 0; |
| 67 | return (0); |
| 68 | } |
| 69 | |
| 70 | /* --- Main work --- */ |
| 71 | |
| 72 | for (q = qq = p; *q; q++) { |
| 73 | switch (st) { |
| 74 | case '\\': |
| 75 | *qq++ = *q; |
| 76 | st = pst; |
| 77 | break; |
| 78 | case '\'': |
| 79 | case '\"': |
| 80 | if (*q == st) |
| 81 | st = pst = 0; |
| 82 | else if (*q == '\\') |
| 83 | st = '\\'; |
| 84 | else |
| 85 | *qq++ = *q; |
| 86 | break; |
| 87 | default: |
| 88 | if (isspace((unsigned char)*q)) { |
| 89 | do q++; while (*q && isspace((unsigned char)*q)); |
| 90 | goto done; |
| 91 | } else if (!(f & STRF_QUOTE)) |
| 92 | goto stdchar; |
| 93 | switch (*q) { |
| 94 | case '\\': |
| 95 | st = '\\'; |
| 96 | break; |
| 97 | case '\'': |
| 98 | case '\"': |
| 99 | st = pst = *q; |
| 100 | break; |
| 101 | default: |
| 102 | stdchar: |
| 103 | *qq++ = *q; |
| 104 | break; |
| 105 | } |
| 106 | } |
| 107 | } |
| 108 | |
| 109 | /* --- Finished --- */ |
| 110 | |
| 111 | done: |
| 112 | *pp = *q ? q : 0; |
| 113 | *qq++ = 0; |
| 114 | return (p); |
| 115 | } |
| 116 | |
| 117 | /* --- @str_qsplit@ --- * |
| 118 | * |
| 119 | * Arguments: @char *p@ = pointer to string |
| 120 | * @char *v[]@ = pointer to array to fill in |
| 121 | * @size_t c@ = count of strings to fill in |
| 122 | * @char **rest@ = where to store the remainder of the string |
| 123 | * @unsigned f@ = flags for @str_qword@ |
| 124 | * |
| 125 | * Returns: Number of strings filled in. |
| 126 | * |
| 127 | * Use: Fills an array with pointers to the individual words of a |
| 128 | * string. The string is modified in place to contain zero |
| 129 | * bytes at the word boundaries, and the words have leading |
| 130 | * and trailing space stripped off. No more than @c@ words |
| 131 | * are read; the actual number is returned as the value of the |
| 132 | * function. Unused slots in the array are populated with |
| 133 | * null bytes. If there's any string left, the address of the |
| 134 | * remainder is stored in @rest@ (if it's non-null); otherwise |
| 135 | * @rest@ is set to a null pointer. |
| 136 | */ |
| 137 | |
| 138 | size_t str_qsplit(char *p, char *v[], size_t c, char **rest, unsigned f) |
| 139 | { |
| 140 | size_t n = 0; |
| 141 | char *q; |
| 142 | |
| 143 | while (c && (q = str_qword(&p, f)) != 0) { |
| 144 | *v++ = q; |
| 145 | c--; |
| 146 | n++; |
| 147 | } |
| 148 | while (c) { |
| 149 | *v++ = 0; |
| 150 | c--; |
| 151 | } |
| 152 | if (rest) |
| 153 | *rest = p; |
| 154 | return (n); |
| 155 | } |
| 156 | |
| 157 | /* --- @str_getword@ --- * |
| 158 | * |
| 159 | * Arguments: @char **pp@ = address of pointer into string |
| 160 | * |
| 161 | * Returns: Pointer to the next space-separated word from the string, |
| 162 | * or null. |
| 163 | * |
| 164 | * Use: Parses off space-separated words from a string. This is a |
| 165 | * compatibility veneer over @str_qword@. |
| 166 | */ |
| 167 | |
| 168 | char *str_getword(char **pp) { return (str_qword(pp, 0)); } |
| 169 | |
| 170 | /* --- @str_split@ --- * |
| 171 | * |
| 172 | * Arguments: @char *p@ = pointer to string |
| 173 | * @char *v[]@ = pointer to array to fill in |
| 174 | * @size_t c@ = count of strings to fill in |
| 175 | * @char **rest@ = where to store the remainder of the string |
| 176 | * |
| 177 | * Returns: Number of strings filled in. |
| 178 | * |
| 179 | * Use: Fills an array with pointers to the individual words of a |
| 180 | * string. This is a compatibility veneer over @str_qsplit@. |
| 181 | */ |
| 182 | |
| 183 | size_t str_split(char *p, char *v[], size_t c, char **rest) |
| 184 | { return (str_qsplit(p, v, c, rest, 0)); } |
| 185 | |
| 186 | /* --- @str_matchx@ --- * |
| 187 | * |
| 188 | * Arguments: @const char *p@ = pointer to pattern string |
| 189 | * @const char *s@ = string to compare with |
| 190 | * @unsigned f@ = various flags |
| 191 | * |
| 192 | * Returns: Nonzero if the pattern matches the string. |
| 193 | * |
| 194 | * Use: Does simple wildcard matching. This is quite nasty and more |
| 195 | * than a little slow. Supports metacharacters `*', `?' and |
| 196 | * '['. |
| 197 | */ |
| 198 | |
| 199 | int str_matchx(const char *p, const char *s, unsigned f) |
| 200 | { |
| 201 | for (;;) { |
| 202 | char pch = *p++, pche, sch; |
| 203 | int sense; |
| 204 | |
| 205 | if ((f & STRF_PREFIX) && !*s) |
| 206 | return (1); |
| 207 | switch (pch) { |
| 208 | case '?': |
| 209 | if (!*s) |
| 210 | return (0); |
| 211 | s++; |
| 212 | break; |
| 213 | case '*': |
| 214 | if (!*p || (f & STRF_PREFIX)) |
| 215 | return (1); |
| 216 | while (*s) { |
| 217 | if (str_match(p, s)) |
| 218 | return (1); |
| 219 | s++; |
| 220 | } |
| 221 | return (0); |
| 222 | case '[': |
| 223 | if (!*s) |
| 224 | return (0); |
| 225 | sch = *s++; |
| 226 | pch = *p++; |
| 227 | sense = 1; |
| 228 | if (pch == '^' || pch == '!') { |
| 229 | sense = !sense; |
| 230 | pch = *p++; |
| 231 | } |
| 232 | if (pch == ']') { |
| 233 | if (*p == '-' && p[1] && p[1] != ']') { |
| 234 | pche = p[1]; |
| 235 | p += 2; |
| 236 | if (pch <= sch && sch <= pche) |
| 237 | goto class_match; |
| 238 | } else if (pch == sch) |
| 239 | goto class_match; |
| 240 | pch = *p++; |
| 241 | } |
| 242 | for (;; pch = *p++) { |
| 243 | if (!pch || pch == ']') |
| 244 | goto class_nomatch; |
| 245 | if (*p == '-' && p[1] && p[1] != ']') { |
| 246 | pche = p[1]; |
| 247 | p += 2; |
| 248 | if (pch <= sch && sch <= pche) |
| 249 | goto class_match; |
| 250 | } else if (pch == sch) |
| 251 | goto class_match; |
| 252 | } |
| 253 | class_match: |
| 254 | if (!sense) |
| 255 | return (0); |
| 256 | for (;;) { |
| 257 | pch = *p++; |
| 258 | if (!pch) |
| 259 | return (0); |
| 260 | if (pch == ']') |
| 261 | break; |
| 262 | if (*p == '-' && p[1] && p[1] != ']') |
| 263 | p += 2; |
| 264 | } |
| 265 | break; |
| 266 | class_nomatch: |
| 267 | if (sense) |
| 268 | return (0); |
| 269 | break; |
| 270 | case '\\': |
| 271 | pch = *p++; |
| 272 | default: |
| 273 | if (pch != *s) |
| 274 | return (0); |
| 275 | if (!pch) |
| 276 | return (1); |
| 277 | s++; |
| 278 | break; |
| 279 | } |
| 280 | } |
| 281 | } |
| 282 | |
| 283 | /* --- @str_match@ --- * |
| 284 | * |
| 285 | * Arguments: @const char *p@ = pointer to pattern string |
| 286 | * @const char *s@ = string to compare with |
| 287 | * |
| 288 | * Returns: Nonzero if the pattern matches the string. |
| 289 | * |
| 290 | * Use: Does simple wildcard matching. Equivalent to @str_matchx@ |
| 291 | * with zero flags word. |
| 292 | */ |
| 293 | |
| 294 | int str_match(const char *p, const char *s) |
| 295 | { return (str_matchx(p, s, 0)); } |
| 296 | |
| 297 | /* --- @str_sanitize@ --- * |
| 298 | * |
| 299 | * Arguments: @char *d@ = destination buffer |
| 300 | * @const char *p@ = pointer to source string |
| 301 | * @size_t sz@ = size of destination buffer |
| 302 | * |
| 303 | * Returns: --- |
| 304 | * |
| 305 | * Use: Writes a string into a buffer, being careful not to overflow |
| 306 | * the buffer, to null terminate the result, and to prevent |
| 307 | * nasty nonprintable characters ending up in the buffer. |
| 308 | */ |
| 309 | |
| 310 | void str_sanitize(char *d, const char *p, size_t sz) |
| 311 | { |
| 312 | if (!sz) |
| 313 | return; |
| 314 | sz--; |
| 315 | while (*p && sz) { |
| 316 | int ch = *p++; |
| 317 | if (!isgraph((unsigned char)ch)) |
| 318 | ch = '_'; |
| 319 | *d++ = ch; |
| 320 | sz--; |
| 321 | } |
| 322 | *d++ = 0; |
| 323 | } |
| 324 | |
| 325 | /*----- That's all, folks -------------------------------------------------*/ |