mono/io-layer/wapi_glob.c

   1 /*      $OpenBSD: glob.c,v 1.26 2005/11/28 17:50:12 deraadt Exp $ */
   2 /*
   3  * Copyright (c) 1989, 1993
   4  *      The Regents of the University of California.  All rights reserved.
   5  *
   6  * This code is derived from software contributed to Berkeley by
   7  * Guido van Rossum.
   8  *
   9  * Redistribution and use in source and binary forms, with or without
  10  * modification, are permitted provided that the following conditions
  11  * are met:
  12  * 1. Redistributions of source code must retain the above copyright
  13  *    notice, this list of conditions and the following disclaimer.
  14  * 2. Redistributions in binary form must reproduce the above copyright
  15  *    notice, this list of conditions and the following disclaimer in the
  16  *    documentation and/or other materials provided with the distribution.
  17  * 3. 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 /*
  35  * _wapi_glob(3) -- a subset of the one defined in POSIX 1003.2.
  36  *
  37  * Optional extra services, controlled by flags not defined by POSIX:
  38  *
  39  * GLOB_MAGCHAR:
  40  *      Set in gl_flags if pattern contained a globbing character.
  41  */
  42 #include <sys/param.h>
  43 #include <sys/stat.h>
  44
  45 #include <glib.h>
  46 #include <ctype.h>
  47 #include <dirent.h>
  48 #include <errno.h>
  49 #include <pwd.h>
  50 #include <stdio.h>
  51 #include <stdlib.h>
  52 #include <string.h>
  53 #include <unistd.h>
  54
  55 #include "wapi_glob.h"
  56
  57 #define EOS             '\0'
  58 #define NOT             '!'
  59 #define QUESTION        '?'
  60 #define QUOTE           '\\'
  61 #define STAR            '*'
  62
  63 #ifndef DEBUG
  64
  65 #define M_QUOTE         0x8000
  66 #define M_PROTECT       0x4000
  67 #define M_MASK          0xffff
  68 #define M_ASCII         0x00ff
  69
  70 typedef u_short Char;
  71
  72 #else
  73
  74 #define M_QUOTE         0x80
  75 #define M_PROTECT       0x40
  76 #define M_MASK          0xff
  77 #define M_ASCII         0x7f
  78
  79 typedef char Char;
  80
  81 #endif
  82
  83
  84 #define CHAR(c)         ((gchar)((c)&M_ASCII))
  85 #define META(c)         ((gchar)((c)|M_QUOTE))
  86 #define M_ALL           META('*')
  87 #define M_ONE           META('?')
  88 #define ismeta(c)       (((c)&M_QUOTE) != 0)
  89
  90
  91 static int       g_Ctoc(const gchar *, char *, u_int);
  92 static int       glob0(GDir *dir, const gchar *, wapi_glob_t *, gboolean,
  93                        gboolean);
  94 static int       glob1(GDir *dir, gchar *, gchar *, wapi_glob_t *, size_t *,
  95                        gboolean, gboolean);
  96 static int       glob3(GDir *dir, gchar *, gchar *, wapi_glob_t *, size_t *,
  97                        gboolean, gboolean);
  98 static int       globextend(const gchar *, wapi_glob_t *, size_t *);
  99 static int       match(const gchar *, gchar *, gchar *, gboolean);
 100 #ifdef DEBUG
 101 static void      qprintf(const char *, Char *);
 102 #endif
 103
 104 int
 105 _wapi_glob(GDir *dir, const char *pattern, int flags, wapi_glob_t *pglob)
 106 {
 107         const u_char *patnext;
 108         int c;
 109         gchar *bufnext, *bufend, patbuf[MAXPATHLEN];
 110
 111         patnext = (u_char *) pattern;
 112         if (!(flags & WAPI_GLOB_APPEND)) {
 113                 pglob->gl_pathc = 0;
 114                 pglob->gl_pathv = NULL;
 115                 pglob->gl_offs = 0;
 116         }
 117         pglob->gl_flags = flags & ~WAPI_GLOB_MAGCHAR;
 118
 119         bufnext = patbuf;
 120         bufend = bufnext + MAXPATHLEN - 1;
 121
 122         /* Protect the quoted characters. */
 123         while (bufnext < bufend && (c = *patnext++) != EOS)
 124                 if (c == QUOTE) {
 125                         if ((c = *patnext++) == EOS) {
 126                                 c = QUOTE;
 127                                 --patnext;
 128                         }
 129                         *bufnext++ = c | M_PROTECT;
 130                 } else
 131                         *bufnext++ = c;
 132
 133         *bufnext = EOS;
 134
 135         return glob0(dir, patbuf, pglob, flags & WAPI_GLOB_IGNORECASE,
 136                      flags & WAPI_GLOB_UNIQUE);
 137 }
 138
 139 /*
 140  * The main glob() routine: compiles the pattern (optionally processing
 141  * quotes), calls glob1() to do the real pattern matching, and finally
 142  * sorts the list (unless unsorted operation is requested).  Returns 0
 143  * if things went well, nonzero if errors occurred.  It is not an error
 144  * to find no matches.
 145  */
 146 static int
 147 glob0(GDir *dir, const gchar *pattern, wapi_glob_t *pglob, gboolean ignorecase,
 148         gboolean unique)
 149 {
 150         const gchar *qpatnext;
 151         int c, err, oldpathc;
 152         gchar *bufnext, patbuf[MAXPATHLEN];
 153         size_t limit = 0;
 154
 155         qpatnext = pattern;
 156         oldpathc = pglob->gl_pathc;
 157         bufnext = patbuf;
 158
 159         /* We don't need to check for buffer overflow any more. */
 160         while ((c = *qpatnext++) != EOS) {
 161                 switch (c) {
 162                 case QUESTION:
 163                         pglob->gl_flags |= WAPI_GLOB_MAGCHAR;
 164                         *bufnext++ = M_ONE;
 165                         break;
 166                 case STAR:
 167                         pglob->gl_flags |= WAPI_GLOB_MAGCHAR;
 168                         /* collapse adjacent stars to one,
 169                          * to avoid exponential behavior
 170                          */
 171                         if (bufnext == patbuf || bufnext[-1] != M_ALL)
 172                                 *bufnext++ = M_ALL;
 173                         break;
 174                 default:
 175                         *bufnext++ = CHAR(c);
 176                         break;
 177                 }
 178         }
 179         *bufnext = EOS;
 180 #ifdef DEBUG
 181         qprintf("glob0:", patbuf);
 182 #endif
 183
 184         if ((err = glob1(dir, patbuf, patbuf+MAXPATHLEN-1, pglob, &limit,
 185                          ignorecase, unique)) != 0)
 186                 return(err);
 187
 188         if (pglob->gl_pathc == oldpathc) {
 189                 return(WAPI_GLOB_NOMATCH);
 190         }
 191
 192         return(0);
 193 }
 194
 195 static int
 196 glob1(GDir *dir, gchar *pattern, gchar *pattern_last, wapi_glob_t *pglob,
 197       size_t *limitp, gboolean ignorecase, gboolean unique)
 198 {
 199         /* A null pathname is invalid -- POSIX 1003.1 sect. 2.4. */
 200         if (*pattern == EOS)
 201                 return(0);
 202         return(glob3(dir, pattern, pattern_last, pglob, limitp, ignorecase,
 203                      unique));
 204 }
 205
 206 static gboolean contains (wapi_glob_t *pglob, const gchar *name)
 207 {
 208         int i;
 209         char **pp;
 210
 211         if (pglob->gl_pathv != NULL) {
 212                 pp = pglob->gl_pathv + pglob->gl_offs;
 213                 for (i = pglob->gl_pathc; i--; ++pp) {
 214                         if (*pp) {
 215                                 if (!strcmp (*pp, name)) {
 216                                         return(TRUE);
 217                                 }
 218                         }
 219                 }
 220         }
 221
 222         return(FALSE);
 223 }
 224
 225 static int
 226 glob3(GDir *dir, gchar *pattern, gchar *pattern_last, wapi_glob_t *pglob,
 227       size_t *limitp, gboolean ignorecase, gboolean unique)
 228 {
 229         const gchar *name;
 230
 231         /* Search directory for matching names. */
 232         while ((name = g_dir_read_name(dir))) {
 233                 if (!match(name, pattern, pattern + strlen (pattern),
 234                            ignorecase)) {
 235                         continue;
 236                 }
 237                 if (!unique ||
 238                     !contains (pglob, name)) {
 239                         globextend (name, pglob, limitp);
 240                 }
 241         }
 242
 243         return(0);
 244 }
 245
 246
 247 /*
 248  * Extend the gl_pathv member of a wapi_glob_t structure to accommodate a new item,
 249  * add the new item, and update gl_pathc.
 250  *
 251  * This assumes the BSD realloc, which only copies the block when its size
 252  * crosses a power-of-two boundary; for v7 realloc, this would cause quadratic
 253  * behavior.
 254  *
 255  * Return 0 if new item added, error code if memory couldn't be allocated.
 256  *
 257  * Invariant of the wapi_glob_t structure:
 258  *      Either gl_pathc is zero and gl_pathv is NULL; or gl_pathc > 0 and
 259  *      gl_pathv points to (gl_offs + gl_pathc + 1) items.
 260  */
 261 static int
 262 globextend(const gchar *path, wapi_glob_t *pglob, size_t *limitp)
 263 {
 264         char **pathv;
 265         int i;
 266         u_int newsize, len;
 267         char *copy;
 268         const gchar *p;
 269
 270         newsize = sizeof(*pathv) * (2 + pglob->gl_pathc + pglob->gl_offs);
 271         pathv = pglob->gl_pathv ? realloc((char *)pglob->gl_pathv, newsize) :
 272             malloc(newsize);
 273         if (pathv == NULL) {
 274                 if (pglob->gl_pathv) {
 275                         free(pglob->gl_pathv);
 276                         pglob->gl_pathv = NULL;
 277                 }
 278                 return(WAPI_GLOB_NOSPACE);
 279         }
 280
 281         if (pglob->gl_pathv == NULL && pglob->gl_offs > 0) {
 282                 /* first time around -- clear initial gl_offs items */
 283                 pathv += pglob->gl_offs;
 284                 for (i = pglob->gl_offs; --i >= 0; )
 285                         *--pathv = NULL;
 286         }
 287         pglob->gl_pathv = pathv;
 288
 289         for (p = path; *p++;)
 290                 ;
 291         len = (size_t)(p - path);
 292         *limitp += len;
 293         if ((copy = malloc(len)) != NULL) {
 294                 if (g_Ctoc(path, copy, len)) {
 295                         free(copy);
 296                         return(WAPI_GLOB_NOSPACE);
 297                 }
 298                 pathv[pglob->gl_offs + pglob->gl_pathc++] = copy;
 299         }
 300         pathv[pglob->gl_offs + pglob->gl_pathc] = NULL;
 301
 302 #if 0
 303         /* Broken on opensuse 11 */
 304         if ((pglob->gl_flags & WAPI_GLOB_LIMIT) &&
 305             newsize + *limitp >= ARG_MAX) {
 306                 errno = 0;
 307                 return(WAPI_GLOB_NOSPACE);
 308         }
 309 #endif
 310
 311         return(copy == NULL ? WAPI_GLOB_NOSPACE : 0);
 312 }
 313
 314
 315 /*
 316  * pattern matching function for filenames.  Each occurrence of the *
 317  * pattern causes a recursion level.
 318  */
 319 static int
 320 match(const gchar *name, gchar *pat, gchar *patend, gboolean ignorecase)
 321 {
 322         gchar c;
 323
 324         while (pat < patend) {
 325                 c = *pat++;
 326                 switch (c & M_MASK) {
 327                 case M_ALL:
 328                         if (pat == patend)
 329                                 return(1);
 330                         do {
 331                                 if (match(name, pat, patend, ignorecase))
 332                                         return(1);
 333                         } while (*name++ != EOS);
 334                         return(0);
 335                 case M_ONE:
 336                         if (*name++ == EOS)
 337                                 return(0);
 338                         break;
 339                 default:
 340                         if (ignorecase) {
 341                                 if (g_ascii_tolower (*name++) != g_ascii_tolower (c))
 342                                         return(0);
 343                         } else {
 344                                 if (*name++ != c)
 345                                         return(0);
 346                         }
 347
 348                         break;
 349                 }
 350         }
 351         return(*name == EOS);
 352 }
 353
 354 /* Free allocated data belonging to a wapi_glob_t structure. */
 355 void
 356 _wapi_globfree(wapi_glob_t *pglob)
 357 {
 358         int i;
 359         char **pp;
 360
 361         if (pglob->gl_pathv != NULL) {
 362                 pp = pglob->gl_pathv + pglob->gl_offs;
 363                 for (i = pglob->gl_pathc; i--; ++pp)
 364                         if (*pp)
 365                                 free(*pp);
 366                 free(pglob->gl_pathv);
 367                 pglob->gl_pathv = NULL;
 368         }
 369 }
 370
 371 static int
 372 g_Ctoc(const gchar *str, char *buf, u_int len)
 373 {
 374
 375         while (len--) {
 376                 if ((*buf++ = *str++) == EOS)
 377                         return (0);
 378         }
 379         return (1);
 380 }
 381
 382 #ifdef DEBUG
 383 static void
 384 qprintf(const char *str, Char *s)
 385 {
 386         Char *p;
 387
 388         (void)printf("%s:\n", str);
 389         for (p = s; *p; p++)
 390                 (void)printf("%c", CHAR(*p));
 391         (void)printf("\n");
 392         for (p = s; *p; p++)
 393                 (void)printf("%c", *p & M_PROTECT ? '"' : ' ');
 394         (void)printf("\n");
 395         for (p = s; *p; p++)
 396                 (void)printf("%c", ismeta(*p) ? '_' : ' ');
 397         (void)printf("\n");
 398 }
 399 #endif