prune: support .payload-link files

Prune payload-links to files that are not present in the repository.

A .payload-link is stored only once, even if there are two objects
with different checksums but same payload.  In the case prune deletes
the file referred by the payload-link, the other file won't have any
.payload-reference left.  A better algorithm to deal with this case
would be to delete every .payload-link file and then recreate them all
for the objects that are left in the repository at the expense of
calculating the checksum for each object in repository.

Signed-off-by: Giuseppe Scrivano <[email protected]>

src/libostree/ostree-repo-prune.c

@@ -46,18 +46,43 @@ maybe_prune_loose_object (OtPruneData        *data,
                           GCancellable       *cancellable,
                           GError            **error)
 {
+  gboolean reachable = FALSE;
   g_autoptr(GVariant) key = NULL;
 
   key = ostree_object_name_serialize (checksum, objtype);
 
-  if (!g_hash_table_lookup_extended (data->reachable, key, NULL, NULL))
+  if (g_hash_table_lookup_extended (data->reachable, key, NULL, NULL))
+    reachable = TRUE;
+  else
     {
       g_debug ("Pruning unneeded object %s.%s", checksum,
                ostree_object_type_to_string (objtype));
       if (!(flags & OSTREE_REPO_PRUNE_FLAGS_NO_PRUNE))
         {
           guint64 storage_size = 0;
 
+          if (objtype == OSTREE_OBJECT_TYPE_PAYLOAD_LINK)
+            {
+              char loose_path_buf[_OSTREE_LOOSE_PATH_MAX];
+              char target_checksum[OSTREE_SHA256_STRING_LEN+1];
+
+              _ostree_loose_path (loose_path_buf, checksum, OSTREE_OBJECT_TYPE_PAYLOAD_LINK, data->repo->mode);
+              if (readlinkat (data->repo->objects_dir_fd, loose_path_buf, target_checksum, sizeof (target_checksum)) < 0)
+                return glnx_throw_errno (error);
+
+              g_autoptr(GVariant) target_key = ostree_object_name_serialize (target_checksum, OSTREE_OBJECT_TYPE_FILE);
+
+              if (g_hash_table_lookup_extended (data->reachable, target_key, NULL, NULL))
+                {
+                  if (!ostree_repo_query_object_storage_size (data->repo, OSTREE_OBJECT_TYPE_FILE, target_checksum,
+                                                              &storage_size, cancellable, error))
+                    return FALSE;
+
+                  reachable = storage_size >= data->repo->payload_link_threshold;
+                  if (reachable)
+                    goto exit;
+                }
+            }
           if (objtype == OSTREE_OBJECT_TYPE_COMMIT)
             {
               if (!ostree_repo_mark_commit_partial (data->repo, checksum, FALSE, error))
@@ -79,7 +104,9 @@ maybe_prune_loose_object (OtPruneData        *data,
       else
         data->n_unreachable_content++;
     }
-  else
+
+ exit:
+  if (reachable)
     {
       g_debug ("Keeping needed object %s.%s", checksum,
                ostree_object_type_to_string (objtype));
@@ -233,6 +260,7 @@ repo_prune_internal (OstreeRepo        *self,
   g_autoptr(GHashTable) reachable_owned = g_hash_table_ref (options->reachable);
   data.reachable = reachable_owned;
 
+
   GLNX_HASH_TABLE_FOREACH_KV (objects, GVariant*, serialized_key, GVariant*, objdata)
     {
       const char *checksum;
@@ -283,7 +311,7 @@ repo_prune_internal (OstreeRepo        *self,
  * of traversing all commits, only refs will be used.  Particularly
  * when combined with @depth, this is a convenient way to delete
  * history from the repository.
- * 
+ *
  * Use the %OSTREE_REPO_PRUNE_FLAGS_NO_PRUNE to just determine
  * statistics on objects that would be deleted, without actually
  * deleting them.