mono/metadata/w32file-unix-glob.c

   1 /*      $OpenBSD: glob.c,v 1.26 2005/11/28 17:50:12 deraadt Exp $ */
   2 /**
   3  * \file
   4  * Copyright (c) 1989, 1993
   5  *      The Regents of the University of California.  All rights reserved.
   6  *
   7  * This code is derived from software contributed to Berkeley by
   8  * Guido van Rossum.
   9  *
  10  * Redistribution and use in source and binary forms, with or without
  11  * modification, are permitted provided that the following conditions
  12  * are met:
  13  * 1. Redistributions of source code must retain the above copyright
  14  *    notice, this list of conditions and the following disclaimer.
  15  * 2. Redistributions in binary form must reproduce the above copyright
  16  *    notice, this list of conditions and the following disclaimer in the
  17  *    documentation and/or other materials provided with the distribution.
  18  * 3. Neither the name of the University nor the names of its contributors
  19  *    may be used to endorse or promote products derived from this software
  20  *    without specific prior written permission.
  21  *
  22  * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
  23  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
  24  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
  25  * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
  26  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
  27  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
  28  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
  29  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
  30  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
  31  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
  32  * SUCH DAMAGE.
  33  */
  34
  35 /*
  36  * mono_w32file_unix_glob(3) -- a subset of the one defined in POSIX 1003.2.
  37  *
  38  * Optional extra services, controlled by flags not defined by POSIX:
  39  *
  40  * GLOB_MAGCHAR:
  41  *      Set in gl_flags if pattern contained a globbing character.
  42  */
  43 #include <sys/types.h>
  44 #include <sys/stat.h>
  45
  46 #include <glib.h>
  47 #include <ctype.h>
  48 #include <errno.h>
  49 #include <mono/utils/mono-errno.h>
  50 #include <stdio.h>
  51 #include <stdlib.h>
  52 #include <string.h>
  53 #include <unistd.h>
  54
  55 #include "w32file-unix-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 unsigned 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
  92 g_Ctoc(const gchar *, char *, unsigned int);
  93
  94 static int
  95 glob0(GDir *dir, const gchar *, mono_w32file_unix_glob_t *, gboolean, gboolean);
  96 static int
  97 glob1(GDir *dir, gchar *, gchar *, mono_w32file_unix_glob_t *, size_t *, gboolean, gboolean);
  98
  99 static int
 100 glob3(GDir *dir, gchar *, gchar *, mono_w32file_unix_glob_t *, size_t *, gboolean, gboolean);
 101
 102 static int
 103 globextend(const gchar *, mono_w32file_unix_glob_t *, size_t *);
 104
 105 static int
 106 match(const gchar *, gchar *, gchar *, gboolean);
 107
 108 #ifdef DEBUG_ENABLED
 109 static void      qprintf(const char *, Char *);
 110 #endif
 111
 112 int
 113 mono_w32file_unix_glob(GDir *dir, const char *pattern, int flags, mono_w32file_unix_glob_t *pglob)
 114 {
 115         const unsigned char *patnext;
 116         int c;
 117         gchar *bufnext, *bufend, patbuf[PATH_MAX];
 118
 119         patnext = (unsigned char *) pattern;
 120         if (!(flags & W32FILE_UNIX_GLOB_APPEND)) {
 121                 pglob->gl_pathc = 0;
 122                 pglob->gl_pathv = NULL;
 123                 pglob->gl_offs = 0;
 124         }
 125         pglob->gl_flags = flags & ~W32FILE_UNIX_GLOB_MAGCHAR;
 126
 127         bufnext = patbuf;
 128         bufend = bufnext + PATH_MAX - 1;
 129
 130         /* Protect the quoted characters. */
 131         while (bufnext < bufend && (c = *patnext++) != EOS)
 132                 if (c == QUOTE) {
 133                         if ((c = *patnext++) == EOS) {
 134                                 c = QUOTE;
 135                                 --patnext;
 136                         }
 137                         *bufnext++ = c | M_PROTECT;
 138                 } else
 139                         *bufnext++ = c;
 140
 141         *bufnext = EOS;
 142
 143         return glob0(dir, patbuf, pglob, flags & W32FILE_UNIX_GLOB_IGNORECASE,
 144                      flags & W32FILE_UNIX_GLOB_UNIQUE);
 145 }
 146
 147 /*
 148  * The main glob() routine: compiles the pattern (optionally processing
 149  * quotes), calls glob1() to do the real pattern matching, and finally
 150  * sorts the list (unless unsorted operation is requested).  Returns 0
 151  * if things went well, nonzero if errors occurred.  It is not an error
 152  * to find no matches.
 153  */
 154 static int
 155 glob0(GDir *dir, const gchar *pattern, mono_w32file_unix_glob_t *pglob, gboolean ignorecase,
 156         gboolean unique)
 157 {
 158         const gchar *qpatnext;
 159         int c, err, oldpathc;
 160         gchar *bufnext, patbuf[PATH_MAX];
 161         size_t limit = 0;
 162
 163         qpatnext = pattern;
 164         oldpathc = pglob->gl_pathc;
 165         bufnext = patbuf;
 166
 167         /* We don't need to check for buffer overflow any more. */
 168         while ((c = *qpatnext++) != EOS) {
 169                 switch (c) {
 170                 case QUESTION:
 171                         pglob->gl_flags |= W32FILE_UNIX_GLOB_MAGCHAR;
 172                         *bufnext++ = M_ONE;
 173                         break;
 174                 case STAR:
 175                         pglob->gl_flags |= W32FILE_UNIX_GLOB_MAGCHAR;
 176                         /* collapse adjacent stars to one,
 177                          * to avoid exponential behavior
 178                          */
 179                         if (bufnext == patbuf || bufnext[-1] != M_ALL)
 180                                 *bufnext++ = M_ALL;
 181                         break;
 182                 default:
 183                         *bufnext++ = CHAR(c);
 184                         break;
 185                 }
 186         }
 187         *bufnext = EOS;
 188 #ifdef DEBUG_ENABLED
 189         qprintf("glob0:", patbuf);
 190 #endif
 191
 192         if ((err = glob1(dir, patbuf, patbuf+PATH_MAX-1, pglob, &limit,
 193                          ignorecase, unique)) != 0)
 194                 return(err);
 195
 196         if (pglob->gl_pathc == oldpathc) {
 197                 return(W32FILE_UNIX_GLOB_NOMATCH);
 198         }
 199
 200         return(0);
 201 }
 202
 203 static int
 204 glob1(GDir *dir, gchar *pattern, gchar *pattern_last, mono_w32file_unix_glob_t *pglob,
 205       size_t *limitp, gboolean ignorecase, gboolean unique)
 206 {
 207         /* A null pathname is invalid -- POSIX 1003.1 sect. 2.4. */
 208         if (*pattern == EOS)
 209                 return(0);
 210         return(glob3(dir, pattern, pattern_last, pglob, limitp, ignorecase,
 211                      unique));
 212 }
 213
 214 static gboolean contains (mono_w32file_unix_glob_t *pglob, const gchar *name)
 215 {
 216         int i;
 217         char **pp;
 218
 219         if (pglob->gl_pathv != NULL) {
 220                 pp = pglob->gl_pathv + pglob->gl_offs;
 221                 for (i = pglob->gl_pathc; i--; ++pp) {
 222                         if (*pp) {
 223                                 if (!strcmp (*pp, name)) {
 224                                         return(TRUE);
 225                                 }
 226                         }
 227                 }
 228         }
 229
 230         return(FALSE);
 231 }
 232
 233 static int
 234 glob3(GDir *dir, gchar *pattern, gchar *pattern_last, mono_w32file_unix_glob_t *pglob,
 235       size_t *limitp, gboolean ignorecase, gboolean unique)
 236 {
 237         const gchar *name;
 238
 239         /* Search directory for matching names. */
 240         while ((name = g_dir_read_name(dir))) {
 241                 if (!match(name, pattern, pattern + strlen (pattern),
 242                            ignorecase)) {
 243                         continue;
 244                 }
 245                 if (!unique ||
 246                     !contains (pglob, name)) {
 247                         globextend (name, pglob, limitp);
 248                 }
 249         }
 250
 251         return(0);
 252 }
 253
 254
 255 /*
 256  * Extend the gl_pathv member of a mono_w32file_unix_glob_t structure to accommodate a new item,
 257  * add the new item, and update gl_pathc.
 258  *
 259  * This assumes the BSD realloc, which only copies the block when its size
 260  * crosses a power-of-two boundary; for v7 realloc, this would cause quadratic
 261  * behavior.
 262  *
 263  * Return 0 if new item added, error code if memory couldn't be allocated.
 264  *
 265  * Invariant of the mono_w32file_unix_glob_t structure:
 266  *      Either gl_pathc is zero and gl_pathv is NULL; or gl_pathc > 0 and
 267  *      gl_pathv points to (gl_offs + gl_pathc + 1) items.
 268  */
 269 static int
 270 globextend(const gchar *path, mono_w32file_unix_glob_t *pglob, size_t *limitp)
 271 {
 272         char **pathv;
 273         int i;
 274         unsigned int newsize, len;
 275         char *copy;
 276         const gchar *p;
 277
 278         newsize = sizeof(*pathv) * (2 + pglob->gl_pathc + pglob->gl_offs);
 279         /* FIXME: Can just use realloc(). */
 280         pathv = pglob->gl_pathv ? (char**)g_realloc (pglob->gl_pathv, newsize) :
 281             (char**)g_malloc (newsize);
 282         if (pathv == NULL) {
 283                 g_free (pglob->gl_pathv);
 284                 pglob->gl_pathv = NULL;
 285                 return(W32FILE_UNIX_GLOB_NOSPACE);
 286         }
 287
 288         if (pglob->gl_pathv == NULL && pglob->gl_offs > 0) {
 289                 /* first time around -- clear initial gl_offs items */
 290                 pathv += pglob->gl_offs;
 291                 for (i = pglob->gl_offs; --i >= 0; )
 292                         *--pathv = NULL;
 293         }
 294         pglob->gl_pathv = pathv;
 295
 296         for (p = path; *p++;)
 297                 ;
 298         len = (size_t)(p - path);
 299         *limitp += len;
 300         if ((copy = (char *)malloc(len)) != NULL) {
 301                 if (g_Ctoc(path, copy, len)) {
 302                         g_free (copy);
 303                         return(W32FILE_UNIX_GLOB_NOSPACE);
 304                 }
 305                 pathv[pglob->gl_offs + pglob->gl_pathc++] = copy;
 306         }
 307         pathv[pglob->gl_offs + pglob->gl_pathc] = NULL;
 308
 309 #if 0
 310         /* Broken on opensuse 11 */
 311         if ((pglob->gl_flags & W32FILE_UNIX_GLOB_LIMIT) &&
 312             newsize + *limitp >= ARG_MAX) {
 313                 mono_set_errno (0);
 314                 return(W32FILE_UNIX_GLOB_NOSPACE);
 315         }
 316 #endif
 317
 318         return(copy == NULL ? W32FILE_UNIX_GLOB_NOSPACE : 0);
 319 }
 320
 321
 322 /*
 323  * pattern matching function for filenames.  Each occurrence of the *
 324  * pattern causes a recursion level.
 325  */
 326 static int
 327 match(const gchar *name, gchar *pat, gchar *patend, gboolean ignorecase)
 328 {
 329         gchar c;
 330
 331         while (pat < patend) {
 332                 c = *pat++;
 333                 switch (c & M_MASK) {
 334                 case M_ALL:
 335                         if (pat == patend)
 336                                 return(1);
 337                         do {
 338                                 if (match(name, pat, patend, ignorecase))
 339                                         return(1);
 340                         } while (*name++ != EOS);
 341                         return(0);
 342                 case M_ONE:
 343                         if (*name++ == EOS)
 344                                 return(0);
 345                         break;
 346                 default:
 347                         if (ignorecase) {
 348                                 if (g_ascii_tolower (*name++) != g_ascii_tolower (c))
 349                                         return(0);
 350                         } else {
 351                                 if (*name++ != c)
 352                                         return(0);
 353                         }
 354
 355                         break;
 356                 }
 357         }
 358         return(*name == EOS);
 359 }
 360
 361 /* Free allocated data belonging to a mono_w32file_unix_glob_t structure. */
 362 void
 363 mono_w32file_unix_globfree(mono_w32file_unix_glob_t *pglob)
 364 {
 365         int i;
 366         char **pp;
 367
 368         if (pglob->gl_pathv != NULL) {
 369                 pp = pglob->gl_pathv + pglob->gl_offs;
 370                 for (i = pglob->gl_pathc; i--; ++pp)
 371                         if (*pp)
 372                                 g_free (*pp);
 373                 g_free (pglob->gl_pathv);
 374                 pglob->gl_pathv = NULL;
 375         }
 376 }
 377
 378 static int
 379 g_Ctoc(const gchar *str, char *buf, unsigned int len)
 380 {
 381
 382         while (len--) {
 383                 if ((*buf++ = *str++) == EOS)
 384                         return (0);
 385         }
 386         return (1);
 387 }
 388
 389 #ifdef DEBUG_ENABLED
 390 static void
 391 qprintf(const char *str, Char *s)
 392 {
 393         Char *p;
 394
 395         (void)printf("%s:\n", str);
 396         for (p = s; *p; p++)
 397                 (void)printf("%c", CHAR(*p));
 398         (void)printf("\n");
 399         for (p = s; *p; p++)
 400                 (void)printf("%c", *p & M_PROTECT ? '"' : ' ');
 401         (void)printf("\n");
 402         for (p = s; *p; p++)
 403                 (void)printf("%c", ismeta(*p) ? '_' : ' ');
 404         (void)printf("\n");
 405 }
 406 #endif