| 1 | /*
|
|---|
| 2 | * Copyright (c) 2008 Jakub Jermar
|
|---|
| 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 | *
|
|---|
| 9 | * - Redistributions of source code must retain the above copyright
|
|---|
| 10 | * notice, this list of conditions and the following disclaimer.
|
|---|
| 11 | * - Redistributions in binary form must reproduce the above copyright
|
|---|
| 12 | * notice, this list of conditions and the following disclaimer in the
|
|---|
| 13 | * documentation and/or other materials provided with the distribution.
|
|---|
| 14 | * - The name of the author may not be used to endorse or promote products
|
|---|
| 15 | * derived from this software without specific prior written permission.
|
|---|
| 16 | *
|
|---|
| 17 | * THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR
|
|---|
| 18 | * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
|
|---|
| 19 | * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
|
|---|
| 20 | * IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT,
|
|---|
| 21 | * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
|
|---|
| 22 | * NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
|
|---|
| 23 | * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
|
|---|
| 24 | * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
|
|---|
| 25 | * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
|
|---|
| 26 | * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
|
|---|
| 27 | */
|
|---|
| 28 |
|
|---|
| 29 | /** @addtogroup libc
|
|---|
| 30 | * @{
|
|---|
| 31 | */
|
|---|
| 32 |
|
|---|
| 33 | /**
|
|---|
| 34 | * @file
|
|---|
| 35 | * @brief
|
|---|
| 36 | */
|
|---|
| 37 |
|
|---|
| 38 | #include <stdlib.h>
|
|---|
| 39 | #include <stddef.h>
|
|---|
| 40 | #include <vfs/canonify.h>
|
|---|
| 41 |
|
|---|
| 42 | /** Token types used for tokenization of path. */
|
|---|
| 43 | typedef enum {
|
|---|
| 44 | TK_INVALID,
|
|---|
| 45 | TK_SLASH,
|
|---|
| 46 | TK_DOT,
|
|---|
| 47 | TK_DOTDOT,
|
|---|
| 48 | TK_COMP,
|
|---|
| 49 | TK_NUL
|
|---|
| 50 | } tokval_t;
|
|---|
| 51 |
|
|---|
| 52 | typedef struct {
|
|---|
| 53 | tokval_t kind;
|
|---|
| 54 | char *start;
|
|---|
| 55 | char *stop;
|
|---|
| 56 | } token_t;
|
|---|
| 57 |
|
|---|
| 58 | /** Fake up the TK_SLASH token. */
|
|---|
| 59 | static token_t slash_token(char *start)
|
|---|
| 60 | {
|
|---|
| 61 | token_t ret;
|
|---|
| 62 | ret.kind = TK_SLASH;
|
|---|
| 63 | ret.start = start;
|
|---|
| 64 | ret.stop = start;
|
|---|
| 65 | return ret;
|
|---|
| 66 | }
|
|---|
| 67 |
|
|---|
| 68 | /** Given a token, return the next token. */
|
|---|
| 69 | static token_t next_token(token_t *cur)
|
|---|
| 70 | {
|
|---|
| 71 | token_t ret;
|
|---|
| 72 |
|
|---|
| 73 | if (cur->stop[1] == '\0') {
|
|---|
| 74 | ret.kind = TK_NUL;
|
|---|
| 75 | ret.start = cur->stop + 1;
|
|---|
| 76 | ret.stop = ret.start;
|
|---|
| 77 | return ret;
|
|---|
| 78 | }
|
|---|
| 79 | if (cur->stop[1] == '/') {
|
|---|
| 80 | ret.kind = TK_SLASH;
|
|---|
| 81 | ret.start = cur->stop + 1;
|
|---|
| 82 | ret.stop = ret.start;
|
|---|
| 83 | return ret;
|
|---|
| 84 | }
|
|---|
| 85 | if (cur->stop[1] == '.' && (!cur->stop[2] || cur->stop[2] == '/')) {
|
|---|
| 86 | ret.kind = TK_DOT;
|
|---|
| 87 | ret.start = cur->stop + 1;
|
|---|
| 88 | ret.stop = ret.start;
|
|---|
| 89 | return ret;
|
|---|
| 90 | }
|
|---|
| 91 | if (cur->stop[1] == '.' && cur->stop[2] == '.' &&
|
|---|
| 92 | (!cur->stop[3] || cur->stop[3] == '/')) {
|
|---|
| 93 | ret.kind = TK_DOTDOT;
|
|---|
| 94 | ret.start = cur->stop + 1;
|
|---|
| 95 | ret.stop = cur->stop + 2;
|
|---|
| 96 | return ret;
|
|---|
| 97 | }
|
|---|
| 98 | unsigned i = 1;
|
|---|
| 99 | while (cur->stop[i] && cur->stop[i] != '/')
|
|---|
| 100 | i++;
|
|---|
| 101 |
|
|---|
| 102 | ret.kind = TK_COMP;
|
|---|
| 103 | ret.start = &cur->stop[1];
|
|---|
| 104 | ret.stop = &cur->stop[i - 1];
|
|---|
| 105 | return ret;
|
|---|
| 106 | }
|
|---|
| 107 |
|
|---|
| 108 | /** States used by canonify(). */
|
|---|
| 109 | typedef enum {
|
|---|
| 110 | S_INI,
|
|---|
| 111 | S_A,
|
|---|
| 112 | S_B,
|
|---|
| 113 | S_C,
|
|---|
| 114 | S_ACCEPT,
|
|---|
| 115 | S_RESTART,
|
|---|
| 116 | S_REJECT
|
|---|
| 117 | } state_t;
|
|---|
| 118 |
|
|---|
| 119 | typedef struct {
|
|---|
| 120 | state_t s;
|
|---|
| 121 | void (*f)(token_t *, token_t *, token_t *);
|
|---|
| 122 | } change_state_t;
|
|---|
| 123 |
|
|---|
| 124 | /*
|
|---|
| 125 | * Actions that can be performed when transitioning from one
|
|---|
| 126 | * state of canonify() to another.
|
|---|
| 127 | */
|
|---|
| 128 | static void set_first_slash(token_t *t, token_t *tfsl, token_t *tlcomp)
|
|---|
| 129 | {
|
|---|
| 130 | *tfsl = *t;
|
|---|
| 131 | *tlcomp = *t;
|
|---|
| 132 | }
|
|---|
| 133 | static void save_component(token_t *t, token_t *tfsl, token_t *tlcomp)
|
|---|
| 134 | {
|
|---|
| 135 | *tlcomp = *t;
|
|---|
| 136 | }
|
|---|
| 137 | static void terminate_slash(token_t *t, token_t *tfsl, token_t *tlcomp)
|
|---|
| 138 | {
|
|---|
| 139 | if (tfsl->stop[1]) /* avoid writing to a well-formatted path */
|
|---|
| 140 | tfsl->stop[1] = '\0';
|
|---|
| 141 | }
|
|---|
| 142 | static void remove_trailing_slash(token_t *t, token_t *tfsl, token_t *tlcomp)
|
|---|
| 143 | {
|
|---|
| 144 | t->start[-1] = '\0';
|
|---|
| 145 | }
|
|---|
| 146 | /** Eat the extra '/'.
|
|---|
| 147 | *
|
|---|
| 148 | * @param t The current TK_SLASH token.
|
|---|
| 149 | */
|
|---|
| 150 | static void shift_slash(token_t *t, token_t *tfsl, token_t *tlcomp)
|
|---|
| 151 | {
|
|---|
| 152 | char *p = t->start;
|
|---|
| 153 | char *q = t->stop + 1;
|
|---|
| 154 | while ((*p++ = *q++))
|
|---|
| 155 | ;
|
|---|
| 156 | }
|
|---|
| 157 | /** Eat the extra '.'.
|
|---|
| 158 | *
|
|---|
| 159 | * @param t The current TK_DOT token.
|
|---|
| 160 | */
|
|---|
| 161 | static void shift_dot(token_t *t, token_t *tfsl, token_t *tlcomp)
|
|---|
| 162 | {
|
|---|
| 163 | char *p = t->start;
|
|---|
| 164 | char *q = t->stop + 1;
|
|---|
| 165 | while ((*p++ = *q++))
|
|---|
| 166 | ;
|
|---|
| 167 | }
|
|---|
| 168 | /** Collapse the TK_COMP TK_SLASH TK_DOTDOT pattern.
|
|---|
| 169 | *
|
|---|
| 170 | * @param t The current TK_DOTDOT token.
|
|---|
| 171 | * @param tlcomp The last TK_COMP token.
|
|---|
| 172 | */
|
|---|
| 173 | static void shift_dotdot(token_t *t, token_t *tfsl, token_t *tlcomp)
|
|---|
| 174 | {
|
|---|
| 175 | char *p = tlcomp->start;
|
|---|
| 176 | char *q = t->stop + 1;
|
|---|
| 177 | while ((*p++ = *q++))
|
|---|
| 178 | ;
|
|---|
| 179 | }
|
|---|
| 180 |
|
|---|
| 181 | /** Transition function for canonify(). */
|
|---|
| 182 | static change_state_t trans[4][6] = {
|
|---|
| 183 | [S_INI] = {
|
|---|
| 184 | [TK_SLASH] = {
|
|---|
| 185 | .s = S_A,
|
|---|
| 186 | .f = set_first_slash,
|
|---|
| 187 | },
|
|---|
| 188 | [TK_DOT] = {
|
|---|
| 189 | .s = S_REJECT,
|
|---|
| 190 | .f = NULL,
|
|---|
| 191 | },
|
|---|
| 192 | [TK_DOTDOT] = {
|
|---|
| 193 | .s = S_REJECT,
|
|---|
| 194 | .f = NULL,
|
|---|
| 195 | },
|
|---|
| 196 | [TK_COMP] = {
|
|---|
| 197 | .s = S_REJECT,
|
|---|
| 198 | .f = NULL,
|
|---|
| 199 | },
|
|---|
| 200 | [TK_NUL] = {
|
|---|
| 201 | .s = S_REJECT,
|
|---|
| 202 | .f = NULL,
|
|---|
| 203 | },
|
|---|
| 204 | [TK_INVALID] = {
|
|---|
| 205 | .s = S_REJECT,
|
|---|
| 206 | .f = NULL,
|
|---|
| 207 | },
|
|---|
| 208 | },
|
|---|
| 209 | [S_A] = {
|
|---|
| 210 | [TK_SLASH] = {
|
|---|
| 211 | .s = S_A,
|
|---|
| 212 | .f = set_first_slash,
|
|---|
| 213 | },
|
|---|
| 214 | [TK_DOT] = {
|
|---|
| 215 | .s = S_A,
|
|---|
| 216 | .f = NULL,
|
|---|
| 217 | },
|
|---|
| 218 | [TK_DOTDOT] = {
|
|---|
| 219 | .s = S_A,
|
|---|
| 220 | .f = NULL,
|
|---|
| 221 | },
|
|---|
| 222 | [TK_COMP] = {
|
|---|
| 223 | .s = S_B,
|
|---|
| 224 | .f = save_component,
|
|---|
| 225 | },
|
|---|
| 226 | [TK_NUL] = {
|
|---|
| 227 | .s = S_ACCEPT,
|
|---|
| 228 | .f = terminate_slash,
|
|---|
| 229 | },
|
|---|
| 230 | [TK_INVALID] = {
|
|---|
| 231 | .s = S_REJECT,
|
|---|
| 232 | .f = NULL,
|
|---|
| 233 | },
|
|---|
| 234 | },
|
|---|
| 235 | [S_B] = {
|
|---|
| 236 | [TK_SLASH] = {
|
|---|
| 237 | .s = S_C,
|
|---|
| 238 | .f = NULL,
|
|---|
| 239 | },
|
|---|
| 240 | [TK_DOT] = {
|
|---|
| 241 | .s = S_REJECT,
|
|---|
| 242 | .f = NULL,
|
|---|
| 243 | },
|
|---|
| 244 | [TK_DOTDOT] = {
|
|---|
| 245 | .s = S_REJECT,
|
|---|
| 246 | .f = NULL,
|
|---|
| 247 | },
|
|---|
| 248 | [TK_COMP] = {
|
|---|
| 249 | .s = S_REJECT,
|
|---|
| 250 | .f = NULL,
|
|---|
| 251 | },
|
|---|
| 252 | [TK_NUL] = {
|
|---|
| 253 | .s = S_ACCEPT,
|
|---|
| 254 | .f = NULL,
|
|---|
| 255 | },
|
|---|
| 256 | [TK_INVALID] = {
|
|---|
| 257 | .s = S_REJECT,
|
|---|
| 258 | .f = NULL,
|
|---|
| 259 | },
|
|---|
| 260 | },
|
|---|
| 261 | [S_C] = {
|
|---|
| 262 | [TK_SLASH] = {
|
|---|
| 263 | .s = S_RESTART,
|
|---|
| 264 | .f = shift_slash,
|
|---|
| 265 | },
|
|---|
| 266 | [TK_DOT] = {
|
|---|
| 267 | .s = S_RESTART,
|
|---|
| 268 | .f = shift_dot,
|
|---|
| 269 | },
|
|---|
| 270 | [TK_DOTDOT] = {
|
|---|
| 271 | .s = S_RESTART,
|
|---|
| 272 | .f = shift_dotdot,
|
|---|
| 273 | },
|
|---|
| 274 | [TK_COMP] = {
|
|---|
| 275 | .s = S_B,
|
|---|
| 276 | .f = save_component,
|
|---|
| 277 | },
|
|---|
| 278 | [TK_NUL] = {
|
|---|
| 279 | .s = S_ACCEPT,
|
|---|
| 280 | .f = remove_trailing_slash,
|
|---|
| 281 | },
|
|---|
| 282 | [TK_INVALID] = {
|
|---|
| 283 | .s = S_REJECT,
|
|---|
| 284 | .f = NULL,
|
|---|
| 285 | },
|
|---|
| 286 | }
|
|---|
| 287 | };
|
|---|
| 288 |
|
|---|
| 289 | /** Canonify a file system path.
|
|---|
| 290 | *
|
|---|
| 291 | * A file system path is canonical, if the following holds:
|
|---|
| 292 | *
|
|---|
| 293 | * 1) the path is absolute
|
|---|
| 294 | * (i.e. a/b/c is not canonical)
|
|---|
| 295 | * 2) there is no trailing slash in the path if it has components
|
|---|
| 296 | * (i.e. /a/b/c/ is not canonical)
|
|---|
| 297 | * 3) there is no extra slash in the path
|
|---|
| 298 | * (i.e. /a//b/c is not canonical)
|
|---|
| 299 | * 4) there is no '.' component in the path
|
|---|
| 300 | * (i.e. /a/./b/c is not canonical)
|
|---|
| 301 | * 5) there is no '..' component in the path
|
|---|
| 302 | * (i.e. /a/b/../c is not canonical)
|
|---|
| 303 | *
|
|---|
| 304 | * This function makes a potentially non-canonical file system path canonical.
|
|---|
| 305 | * It works in-place and requires a NULL-terminated input string.
|
|---|
| 306 | *
|
|---|
| 307 | * @param path Path to be canonified.
|
|---|
| 308 | * @param lenp Pointer where the length of the final path will be
|
|---|
| 309 | * stored. Can be NULL.
|
|---|
| 310 | *
|
|---|
| 311 | * @return Canonified path or NULL on failure.
|
|---|
| 312 | */
|
|---|
| 313 | char *canonify(char *path, size_t *lenp)
|
|---|
| 314 | {
|
|---|
| 315 | state_t state;
|
|---|
| 316 | token_t t;
|
|---|
| 317 | token_t tfsl; /* first slash */
|
|---|
| 318 | token_t tlcomp; /* last component */
|
|---|
| 319 | if (*path != '/')
|
|---|
| 320 | return NULL;
|
|---|
| 321 | tfsl = slash_token(path);
|
|---|
| 322 | restart:
|
|---|
| 323 | state = S_INI;
|
|---|
| 324 | t = tfsl;
|
|---|
| 325 | tlcomp = tfsl;
|
|---|
| 326 | while (state != S_ACCEPT && state != S_RESTART && state != S_REJECT) {
|
|---|
| 327 | if (trans[state][t.kind].f)
|
|---|
| 328 | trans[state][t.kind].f(&t, &tfsl, &tlcomp);
|
|---|
| 329 | state = trans[state][t.kind].s;
|
|---|
| 330 | t = next_token(&t);
|
|---|
| 331 | }
|
|---|
| 332 |
|
|---|
| 333 | switch (state) {
|
|---|
| 334 | case S_RESTART:
|
|---|
| 335 | goto restart;
|
|---|
| 336 | case S_REJECT:
|
|---|
| 337 | return NULL;
|
|---|
| 338 | case S_ACCEPT:
|
|---|
| 339 | if (lenp)
|
|---|
| 340 | *lenp = (size_t)((tlcomp.stop - tfsl.start) + 1);
|
|---|
| 341 | return tfsl.start;
|
|---|
| 342 | default:
|
|---|
| 343 | abort();
|
|---|
| 344 | }
|
|---|
| 345 | }
|
|---|
| 346 |
|
|---|
| 347 | /**
|
|---|
| 348 | * @}
|
|---|
| 349 | */
|
|---|