contrib/libarchive-2.0/libarchive/archive_write_disk_set_standard_lookup.c

   1 /*-
   2  * Copyright (c) 2003-2007 Tim Kientzle
   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  * 1. Redistributions of source code must retain the above copyright
   9  *    notice, this list of conditions and the following disclaimer.
  10  * 2. Redistributions in binary form must reproduce the above copyright
  11  *    notice, this list of conditions and the following disclaimer in the
  12  *    documentation and/or other materials provided with the distribution.
  13  *
  14  * THIS SOFTWARE IS PROVIDED BY THE AUTHOR(S) ``AS IS'' AND ANY EXPRESS OR
  15  * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
  16  * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
  17  * IN NO EVENT SHALL THE AUTHOR(S) BE LIABLE FOR ANY DIRECT, INDIRECT,
  18  * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
  19  * NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
  20  * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
  21  * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
  22  * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
  23  * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
  24  */
  25
  26 #include "archive_platform.h"
  27 __FBSDID("$FreeBSD: src/lib/libarchive/archive_write_disk_set_standard_lookup.c,v 1.1 2007/03/03 07:37:36 kientzle Exp $");
  28
  29 #ifdef HAVE_SYS_TYPES_H
  30 #include <sys/types.h>
  31 #endif
  32 #ifdef HAVE_SYS_ACL_H
  33 #include <sys/acl.h>
  34 #endif
  35 #ifdef HAVE_ATTR_XATTR_H
  36 #include <attr/xattr.h>
  37 #endif
  38 #ifdef HAVE_SYS_IOCTL_H
  39 #include <sys/ioctl.h>
  40 #endif
  41 #ifdef HAVE_SYS_STAT_H
  42 #include <sys/stat.h>
  43 #endif
  44 #ifdef HAVE_SYS_TIME_H
  45 #include <sys/time.h>
  46 #endif
  47
  48 #ifdef HAVE_ERRNO_H
  49 #include <errno.h>
  50 #endif
  51 #ifdef HAVE_GRP_H
  52 #include <grp.h>
  53 #endif
  54 #ifdef HAVE_PWD_H
  55 #include <pwd.h>
  56 #endif
  57 #ifdef HAVE_STDLIB_H
  58 #include <stdlib.h>
  59 #endif
  60 #ifdef HAVE_STRING_H
  61 #include <string.h>
  62 #endif
  63
  64 #include "archive.h"
  65 #include "archive_private.h"
  66 #include "archive_read_private.h"
  67 #include "archive_write_disk_private.h"
  68
  69 struct bucket {
  70         char    *name;
  71         int      hash;
  72         id_t     id;
  73 };
  74
  75 static unsigned int     hash(const char *);
  76 static gid_t    lookup_gid(void *, const char *uname, gid_t);
  77 static uid_t    lookup_uid(void *, const char *uname, uid_t);
  78 static void     cleanup(void *);
  79
  80 /*
  81  * Installs functions that use getpwnam()/getgrnam()---along with
  82  * a simple cache to accelerate such lookups---into the archive_write_disk
  83  * object.  This is in a separate file because getpwnam()/getgrnam()
  84  * can pull in a LOT of library code (including NIS/LDAP functions, which
  85  * pull in DNS resolveers, etc).  This can easily top 500kB, which makes
  86  * it inappropriate for some space-constrained applications.
  87  *
  88  * Applications that are size-sensitive may want to just use the
  89  * real default functions (defined in archive_write_disk.c) that just
  90  * use the uid/gid without the lookup.  Or define your own custom functions
  91  * if you prefer.
  92  *
  93  * TODO: Replace these hash tables with simpler move-to-front LRU
  94  * lists with a bounded size (128 items?).  The hash is a bit faster,
  95  * but has a bad pathology in which it thrashes a single bucket.  Even
  96  * walking a list of 128 items is a lot faster than calling
  97  * getpwnam()!
  98  */
  99 int
 100 archive_write_disk_set_standard_lookup(struct archive *a)
 101 {
 102         struct bucket *ucache = malloc(sizeof(struct bucket[127]));
 103         struct bucket *gcache = malloc(sizeof(struct bucket[127]));
 104         memset(ucache, 0, sizeof(struct bucket[127]));
 105         memset(gcache, 0, sizeof(struct bucket[127]));
 106         archive_write_disk_set_group_lookup(a, gcache, lookup_gid, cleanup);
 107         archive_write_disk_set_user_lookup(a, ucache, lookup_uid, cleanup);
 108         return (ARCHIVE_OK);
 109 }
 110
 111 static gid_t
 112 lookup_gid(void *private_data, const char *gname, gid_t gid)
 113 {
 114         int h;
 115         struct bucket *b;
 116         int cache_size;
 117         struct bucket *gcache = (struct bucket *)private_data;
 118
 119         cache_size = 127;
 120
 121         /* If no gname, just use the gid provided. */
 122         if (gname == NULL || *gname == '\0')
 123                 return (gid);
 124
 125         /* Try to find gname in the cache. */
 126         h = hash(gname);
 127         b = &gcache[h % cache_size ];
 128         if (b->name != NULL && b->hash == h && strcmp(gname, b->name) == 0)
 129                 return ((gid_t)b->id);
 130
 131         /* Free the cache slot for a new entry. */
 132         if (b->name != NULL)
 133                 free(b->name);
 134         b->name = strdup(gname);
 135         /* Note: If strdup fails, that's okay; we just won't cache. */
 136         b->hash = h;
 137 #if HAVE_GRP_H
 138         {
 139                 struct group    *grent = getgrnam(gname);
 140                 if (grent != NULL)
 141                         gid = grent->gr_gid;
 142         }
 143 #elif _WIN32
 144         /* TODO: do a gname->gid lookup for Windows. */
 145 #endif
 146         b->id = gid;
 147
 148         return (gid);
 149 }
 150
 151 static uid_t
 152 lookup_uid(void *private_data, const char *uname, uid_t uid)
 153 {
 154         int h;
 155         struct bucket *b;
 156         int cache_size;
 157         struct bucket *ucache = (struct bucket *)private_data;
 158
 159         cache_size = 127;
 160
 161         /* If no uname, just use the uid provided. */
 162         if (uname == NULL || *uname == '\0')
 163                 return (uid);
 164
 165         /* Try to find uname in the cache. */
 166         h = hash(uname);
 167         b = &ucache[h % cache_size ];
 168         if (b->name != NULL && b->hash == h && strcmp(uname, b->name) == 0)
 169                 return ((uid_t)b->id);
 170
 171         /* Free the cache slot for a new entry. */
 172         if (b->name != NULL)
 173                 free(b->name);
 174         b->name = strdup(uname);
 175         /* Note: If strdup fails, that's okay; we just won't cache. */
 176         b->hash = h;
 177 #if HAVE_PWD_H
 178         {
 179                 struct passwd   *pwent = getpwnam(uname);
 180                 if (pwent != NULL)
 181                         uid = pwent->pw_uid;
 182         }
 183 #elif _WIN32
 184         /* TODO: do a uname->uid lookup for Windows. */
 185 #endif
 186         b->id = uid;
 187
 188         return (uid);
 189 }
 190
 191 static void
 192 cleanup(void *private)
 193 {
 194         free(private);
 195 }
 196
 197
 198 static unsigned int
 199 hash(const char *p)
 200 {
 201         /* A 32-bit version of Peter Weinberger's (PJW) hash algorithm,
 202            as used by ELF for hashing function names. */
 203         unsigned g, h = 0;
 204         while (*p != '\0') {
 205                 h = ( h << 4 ) + *p++;
 206                 if (( g = h & 0xF0000000 )) {
 207                         h ^= g >> 24;
 208                         h &= 0x0FFFFFFF;
 209                 }
 210         }
 211         return h;
 212 }