data: replace Tree with TreeNodeIterator

The TreeNodeIterator decodes nodes while iterating over a tree blob.
This should reduce peak memory usage as now only the serialized tree
blob and a single node have to alive at the same time. Using the
iterator has implications for the error handling however. Now it is
necessary that all loops that iterate through a tree check for errors
before using the node returned by the iterator.

The other change is that it is no longer possible to iterate over a tree
multiple times. Instead it must be loaded a second time. This only
affects the tree rewriting code.
This commit is contained in:
Michael Eischer
2026-01-31 20:03:38 +01:00
parent 1e183509d4
commit 350f29d921
23 changed files with 394 additions and 233 deletions
+7 -3
View File
@@ -30,7 +30,7 @@ func New(format string, repo restic.Loader, w io.Writer) *Dumper {
}
}
func (d *Dumper) DumpTree(ctx context.Context, tree *data.Tree, rootPath string) error {
func (d *Dumper) DumpTree(ctx context.Context, tree data.TreeNodeIterator, rootPath string) error {
wg, ctx := errgroup.WithContext(ctx)
// ch is buffered to deal with variable download/write speeds.
@@ -52,10 +52,14 @@ func (d *Dumper) DumpTree(ctx context.Context, tree *data.Tree, rootPath string)
return wg.Wait()
}
func sendTrees(ctx context.Context, repo restic.BlobLoader, tree *data.Tree, rootPath string, ch chan *data.Node) error {
func sendTrees(ctx context.Context, repo restic.BlobLoader, nodes data.TreeNodeIterator, rootPath string, ch chan *data.Node) error {
defer close(ch)
for _, node := range tree.Nodes {
for item := range nodes {
if item.Error != nil {
return item.Error
}
node := item.Node
node.Path = path.Join(rootPath, node.Name)
if err := sendNodes(ctx, repo, node, ch); err != nil {
return err