Add 'full' parameter to alpm_pkg_load

In most cases, we want to fully scan a package when we load it, which serves
as a integrity verification check. However, there are times when it is only
desired to read the metadata and nothing else, so allow the caller of pkg_load
to choose the behavior they need.

This pays big dividends in speeding up pacman cache cleaning functionality.
Old (729 packages):
real    1m43.717s
user    1m20.785s
sys     0m2.993s
New (729 packages):
real    0m25.607s
user    0m19.389s
sys     0m0.543s

Signed-off-by: Dan McGee <dan@archlinux.org>

7 files changed, 35 insertions, 22 deletions

diff --git a/lib/libalpm/add.c b/lib/libalpm/add.c
index 442b3180..47a885ec 100644
--- a/lib/libalpm/add.c
+++ b/lib/libalpm/add.c
@@ -65,7 +65,7 @@ int _alpm_add_loadtarget(pmtrans_t *trans, pmdb_t *db, char *name)
 
 	_alpm_log(PM_LOG_DEBUG, "loading target '%s'\n", name);
 
-	if(alpm_pkg_load(name, &pkg) != 0) {
+	if(alpm_pkg_load(name, 1, &pkg) != 0) {
 		goto error;
 	}
 	pkgname = alpm_pkg_get_name(pkg);
@@ -97,7 +97,7 @@ int _alpm_add_loadtarget(pmtrans_t *trans, pmdb_t *db, char *name)
 				pmpkg_t *newpkg;
 				_alpm_log(PM_LOG_WARNING, _("replacing older version %s-%s by %s in target list\n"),
 				          pkg->name, pkg->version, pkgver);
-				if((newpkg = _alpm_pkg_load(name)) == NULL) {
+				if((newpkg = _alpm_pkg_load(name, 1)) == NULL) {
 					/* pm_errno is already set by pkg_load() */
 					goto error;
 				}
diff --git a/lib/libalpm/alpm.h b/lib/libalpm/alpm.h
index 53bbaaa0..f208398d 100644
--- a/lib/libalpm/alpm.h
+++ b/lib/libalpm/alpm.h
@@ -180,7 +180,7 @@ typedef enum _pmpkgreason_t {
 	PM_PKG_REASON_DEPEND = 1  /* installed as a dependency for another package */
 } pmpkgreason_t;
 
-int alpm_pkg_load(const char *filename, pmpkg_t **pkg);
+int alpm_pkg_load(const char *filename, unsigned short full, pmpkg_t **pkg);
 int alpm_pkg_free(pmpkg_t *pkg);
 int alpm_pkg_checkmd5sum(pmpkg_t *pkg);
 char *alpm_fetch_pkgurl(const char *url);
diff --git a/lib/libalpm/package.c b/lib/libalpm/package.c
index 30d77a74..96c264ac 100644
--- a/lib/libalpm/package.c
+++ b/lib/libalpm/package.c
@@ -57,10 +57,13 @@
 
 /** Create a package from a file.
  * @param filename location of the package tarball
+ * @param full whether to stop the load after metadata is read or continue
+ *             through the full archive
  * @param pkg address of the package pointer
  * @return 0 on success, -1 on error (pm_errno is set accordingly)
  */
-int SYMEXPORT alpm_pkg_load(const char *filename, pmpkg_t **pkg)
+int SYMEXPORT alpm_pkg_load(const char *filename, unsigned short full,
+		pmpkg_t **pkg)
 {
 	_alpm_log(PM_LOG_FUNCTION, "enter alpm_pkg_load\n");
 
@@ -69,7 +72,7 @@ int SYMEXPORT alpm_pkg_load(const char *filename, pmpkg_t **pkg)
 			RET_ERR(PM_ERR_WRONG_ARGS, -1));
 	ASSERT(pkg != NULL, RET_ERR(PM_ERR_WRONG_ARGS, -1));
 
-	*pkg = _alpm_pkg_load(filename);
+	*pkg = _alpm_pkg_load(filename, full);
 	if(*pkg == NULL) {
 		/* pm_errno is set by pkg_load */
 		return(-1);
@@ -870,13 +873,19 @@ static int parse_descfile(const char *descfile, pmpkg_t *info)
 	return(0);
 }
 
-pmpkg_t *_alpm_pkg_load(const char *pkgfile)
+
+/**
+ * Load a package and create the corresponding pmpkg_t struct.
+ * @param pkgfile path to the package file
+ * @param full whether to stop the load after metadata is read or continue
+ *             through the full archive
+ * @return An information filled pmpkg_t struct
+ */
+pmpkg_t *_alpm_pkg_load(const char *pkgfile, unsigned short full)
 {
-	char *expath;
 	int ret = ARCHIVE_OK;
 	int config = 0;
 	int filelist = 0;
-	int scriptcheck = 0;
 	struct archive *archive;
 	struct archive_entry *entry;
 	pmpkg_t *info = NULL;
@@ -920,9 +929,9 @@ pmpkg_t *_alpm_pkg_load(const char *pkgfile)
 	 * requires changes 'parse_descfile' as well
 	 * */
 
-	/* Read through the entire archive for metadata.  We will continue reading
-	 * even if all metadata is found, to verify the integrity of the archive in
-	 * full */
+	/* If full is false, only read through the archive until we find our needed
+	 * metadata. If it is true, read through the entire archive, which serves
+	 * as a verfication of integrity. */
 	while((ret = archive_read_next_header (archive, &entry)) == ARCHIVE_OK) {
 		const char *entry_name = archive_entry_pathname(entry);
 
@@ -951,7 +960,6 @@ pmpkg_t *_alpm_pkg_load(const char *pkgfile)
 			continue;
 		} else if(strcmp(entry_name,  ".INSTALL") == 0) {
 			info->scriptlet = 1;
-			scriptcheck = 1;
 		} else if(strcmp(entry_name, ".FILELIST") == 0) {
 			/* Build info->files from the filelist */
 			FILE *fp;
@@ -983,10 +991,9 @@ pmpkg_t *_alpm_pkg_load(const char *pkgfile)
 			filelist = 1;
 			continue;
 		} else if(*entry_name == '.') {
-					/* for now, ignore all files starting with '.' that haven't
-					 * already been handled (for future possibilities) */
+			/* for now, ignore all files starting with '.' that haven't
+			 * already been handled (for future possibilities) */
 		} else {
-			scriptcheck = 1;
 			/* Keep track of all files so we can generate a filelist later if missing */
 			all_files = alpm_list_add(all_files, strdup(entry_name));
 		}
@@ -996,9 +1003,14 @@ pmpkg_t *_alpm_pkg_load(const char *pkgfile)
 			pm_errno = PM_ERR_LIBARCHIVE_ERROR;
 			goto error;
 		}
-		expath = NULL;
+
+		/* if we are not doing a full read, see if we have all we need */
+		if(!full && config && filelist) {
+			break;
+		}
 	}
-	if(ret != ARCHIVE_EOF) { /* An error occured */
+
+	if(ret != ARCHIVE_EOF && ret != ARCHIVE_OK) { /* An error occured */
 		_alpm_log(PM_LOG_ERROR, _("error while reading package: %s\n"), archive_error_string(archive));
 		pm_errno = PM_ERR_LIBARCHIVE_ERROR;
 		goto error;
diff --git a/lib/libalpm/package.h b/lib/libalpm/package.h
index 30dd3961..47d384b6 100644
--- a/lib/libalpm/package.h
+++ b/lib/libalpm/package.h
@@ -94,7 +94,7 @@ pmpkg_t *_alpm_pkg_dup(pmpkg_t *pkg);
 void _alpm_pkg_free(pmpkg_t *pkg);
 int _alpm_pkg_cmp(const void *p1, const void *p2);
 int alpm_pkg_compare_versions(pmpkg_t *local_pkg, pmpkg_t *pkg);
-pmpkg_t *_alpm_pkg_load(const char *pkgfile);
+pmpkg_t *_alpm_pkg_load(const char *pkgfile, unsigned short full);
 pmpkg_t *_alpm_pkg_find(const char *needle, alpm_list_t *haystack);
 int _alpm_pkg_istoonew(pmpkg_t *pkg);
 void _alpm_pkg_update_requiredby(pmpkg_t *pkg);
diff --git a/src/pacman/query.c b/src/pacman/query.c
index 26a1bd8f..4e4002cf 100644
--- a/src/pacman/query.c
+++ b/src/pacman/query.c
@@ -369,7 +369,7 @@ int pacman_query(alpm_list_t *targets)
 		pmpkg_t *pkg = NULL;
 
 		if(config->op_q_isfile) {
-			alpm_pkg_load(strname, &pkg);
+			alpm_pkg_load(strname, 1, &pkg);
 		} else {
 			pkg = alpm_db_get_pkg(db_local, strname);
 		}
diff --git a/src/pacman/sync.c b/src/pacman/sync.c
index bb4926e2..11941eb3 100644
--- a/src/pacman/sync.c
+++ b/src/pacman/sync.c
@@ -82,8 +82,9 @@ static int sync_cleancache(int level)
 			snprintf(path, PATH_MAX, "%s/%s", cachedir, ent->d_name);
 
 			/* attempt to load the package, skip file on failures as we may have
-			 * files here that aren't valid packages */
-			if(alpm_pkg_load(path, &localpkg) != 0 || localpkg == NULL) {
+			 * files here that aren't valid packages. we also don't need a full
+			 * load of the package, just the metadata. */
+			if(alpm_pkg_load(path, 0, &localpkg) != 0 || localpkg == NULL) {
 				continue;
 			}
 			/* check if this package is in the local DB */
diff --git a/src/util/testpkg.c b/src/util/testpkg.c
index 7da18ca6..7d844e85 100644
--- a/src/util/testpkg.c
+++ b/src/util/testpkg.c
@@ -57,7 +57,7 @@ int main(int argc, char **argv)
     /* let us get log messages from libalpm */
 	alpm_option_set_logcb(output_cb);
 
-	if(alpm_pkg_load(argv[1], &pkg) == -1 || pkg == NULL) {
+	if(alpm_pkg_load(argv[1], 1, &pkg) == -1 || pkg == NULL) {
         retval = 1;
 	} else {
 		alpm_pkg_free(pkg);