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