libarchive 2.0

  * libarchive_test program exercises many of the core features
  * Refactored old "read_extract" into new "archive_write_disk", which
    uses archive_write methods to put entries onto disk.  In particular,
    you can now use archive_write_disk to create objects on disk
    without having an archive available.
  * Pushed some security checks from bsdtar down into libarchive, where
    they can be better optimized.
  * Rearchitected the logic for creating objects on disk to reduce
    the number of system calls.  Several common cases now use a
    minimum number of system calls.
  * Virtualized some internal interfaces to provide a clearer separation
    of read and write handling and make it simpler to override key
    methods.
  * New "empty" format reader.
  * Corrected return types (this ABI breakage required the "2.0" version bump)
  * Many bug fixes.

1 files changed, 212 insertions, 0 deletions

diff --git a/lib/libarchive/archive_write_disk_set_standard_lookup.c b/lib/libarchive/archive_write_disk_set_standard_lookup.c
new file mode 100644
index 000000000000..273b9af83e99
--- /dev/null
+++ b/lib/libarchive/archive_write_disk_set_standard_lookup.c
@@ -0,0 +1,212 @@
+/*-
+ * Copyright (c) 2003-2007 Tim Kientzle
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ * 1. Redistributions of source code must retain the above copyright
+ *    notice, this list of conditions and the following disclaimer.
+ * 2. Redistributions in binary form must reproduce the above copyright
+ *    notice, this list of conditions and the following disclaimer in the
+ *    documentation and/or other materials provided with the distribution.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE AUTHOR(S) ``AS IS'' AND ANY EXPRESS OR
+ * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
+ * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
+ * IN NO EVENT SHALL THE AUTHOR(S) BE LIABLE FOR ANY DIRECT, INDIRECT,
+ * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
+ * NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
+ * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#include "archive_platform.h"
+__FBSDID("$FreeBSD$");
+
+#ifdef HAVE_SYS_TYPES_H
+#include <sys/types.h>
+#endif
+#ifdef HAVE_SYS_ACL_H
+#include <sys/acl.h>
+#endif
+#ifdef HAVE_ATTR_XATTR_H
+#include <attr/xattr.h>
+#endif
+#ifdef HAVE_SYS_IOCTL_H
+#include <sys/ioctl.h>
+#endif
+#ifdef HAVE_SYS_STAT_H
+#include <sys/stat.h>
+#endif
+#ifdef HAVE_SYS_TIME_H
+#include <sys/time.h>
+#endif
+
+#ifdef HAVE_ERRNO_H
+#include <errno.h>
+#endif
+#ifdef HAVE_GRP_H
+#include <grp.h>
+#endif
+#ifdef HAVE_PWD_H
+#include <pwd.h>
+#endif
+#ifdef HAVE_STDLIB_H
+#include <stdlib.h>
+#endif
+#ifdef HAVE_STRING_H
+#include <string.h>
+#endif
+
+#include "archive.h"
+#include "archive_private.h"
+#include "archive_read_private.h"
+#include "archive_write_disk_private.h"
+
+struct bucket {
+	char	*name;
+	int	 hash;
+	id_t	 id;
+};
+
+static unsigned int	hash(const char *);
+static gid_t	lookup_gid(void *, const char *uname, gid_t);
+static uid_t	lookup_uid(void *, const char *uname, uid_t);
+static void	cleanup(void *);
+
+/*
+ * Installs functions that use getpwnam()/getgrnam()---along with
+ * a simple cache to accelerate such lookups---into the archive_write_disk
+ * object.  This is in a separate file because getpwnam()/getgrnam()
+ * can pull in a LOT of library code (including NIS/LDAP functions, which
+ * pull in DNS resolveers, etc).  This can easily top 500kB, which makes
+ * it inappropriate for some space-constrained applications.
+ *
+ * Applications that are size-sensitive may want to just use the
+ * real default functions (defined in archive_write_disk.c) that just
+ * use the uid/gid without the lookup.  Or define your own custom functions
+ * if you prefer.
+ *
+ * TODO: Replace these hash tables with simpler move-to-front LRU
+ * lists with a bounded size (128 items?).  The hash is a bit faster,
+ * but has a bad pathology in which it thrashes a single bucket.  Even
+ * walking a list of 128 items is a lot faster than calling
+ * getpwnam()!
+ */
+int
+archive_write_disk_set_standard_lookup(struct archive *a)
+{
+	struct bucket *ucache = malloc(sizeof(struct bucket[127]));
+	struct bucket *gcache = malloc(sizeof(struct bucket[127]));
+	memset(ucache, 0, sizeof(struct bucket[127]));
+	memset(gcache, 0, sizeof(struct bucket[127]));
+	archive_write_disk_set_group_lookup(a, gcache, lookup_gid, cleanup);
+	archive_write_disk_set_user_lookup(a, ucache, lookup_uid, cleanup);
+	return (ARCHIVE_OK);
+}
+
+static gid_t
+lookup_gid(void *private_data, const char *gname, gid_t gid)
+{
+	int h;
+	struct bucket *b;
+	int cache_size;
+	struct bucket *gcache = (struct bucket *)private_data;
+
+	cache_size = 127;
+
+	/* If no gname, just use the gid provided. */
+	if (gname == NULL || *gname == '\0')
+		return (gid);
+
+	/* Try to find gname in the cache. */
+	h = hash(gname);
+	b = &gcache[h % cache_size ];
+	if (b->name != NULL && b->hash == h && strcmp(gname, b->name) == 0)
+		return ((gid_t)b->id);
+
+	/* Free the cache slot for a new entry. */
+	if (b->name != NULL)
+		free(b->name);
+	b->name = strdup(gname);
+	/* Note: If strdup fails, that's okay; we just won't cache. */
+	b->hash = h;
+#if HAVE_GRP_H
+	{
+		struct group	*grent = getgrnam(gname);
+		if (grent != NULL)
+			gid = grent->gr_gid;
+	}
+#elif _WIN32
+	/* TODO: do a gname->gid lookup for Windows. */
+#endif
+	b->id = gid;
+
+	return (gid);
+}
+
+static uid_t
+lookup_uid(void *private_data, const char *uname, uid_t uid)
+{
+	int h;
+	struct bucket *b;
+	int cache_size;
+	struct bucket *ucache = (struct bucket *)private_data;
+
+	cache_size = 127;
+
+	/* If no uname, just use the uid provided. */
+	if (uname == NULL || *uname == '\0')
+		return (uid);
+
+	/* Try to find uname in the cache. */
+	h = hash(uname);
+	b = &ucache[h % cache_size ];
+	if (b->name != NULL && b->hash == h && strcmp(uname, b->name) == 0)
+		return ((uid_t)b->id);
+
+	/* Free the cache slot for a new entry. */
+	if (b->name != NULL)
+		free(b->name);
+	b->name = strdup(uname);
+	/* Note: If strdup fails, that's okay; we just won't cache. */
+	b->hash = h;
+#if HAVE_PWD_H
+	{
+		struct passwd	*pwent = getpwnam(uname);
+		if (pwent != NULL)
+			uid = pwent->pw_uid;
+	}
+#elif _WIN32
+	/* TODO: do a uname->uid lookup for Windows. */
+#endif
+	b->id = uid;
+
+	return (uid);
+}
+
+static void
+cleanup(void *private)
+{
+	free(private);
+}
+
+
+static unsigned int
+hash(const char *p)
+{
+	/* A 32-bit version of Peter Weinberger's (PJW) hash algorithm,
+	   as used by ELF for hashing function names. */
+	unsigned g, h = 0;
+	while (*p != '\0') {
+		h = ( h << 4 ) + *p++;
+		if (( g = h & 0xF0000000 )) {
+			h ^= g >> 24;
+			h &= 0x0FFFFFFF;
+		}
+	}
+	return h;
+}