Merge pull request #5098 from schomatis/feat/unixfs/fsnode-include-pb

unixfs: integrate `pb.Data` into `FSNode` to avoid duplicating fields

importer/helpers/dagbuilder.go

@@ -116,7 +116,7 @@ func (db *DagBuilderHelper) GetDagServ() ipld.DAGService {
 func (db *DagBuilderHelper) NewUnixfsNode() *UnixfsNode {
 	n := &UnixfsNode{
 		node: new(dag.ProtoNode),
-		ufmt: &ft.FSNode{Type: ft.TFile},
+		ufmt: ft.NewFSNode(ft.TFile),
 	}
 	n.SetPrefix(db.prefix)
 	return n
@@ -161,7 +161,7 @@ func (db *DagBuilderHelper) NewLeaf(data []byte) (*UnixfsNode, error) {
 func (db *DagBuilderHelper) newUnixfsBlock() *UnixfsNode {
 	n := &UnixfsNode{
 		node: new(dag.ProtoNode),
-		ufmt: &ft.FSNode{Type: ft.TRaw},
+		ufmt: ft.NewFSNode(ft.TRaw),
 	}
 	n.SetPrefix(db.prefix)
 	return n

importer/helpers/helpers.go

@@ -77,7 +77,7 @@ func (n *UnixfsNode) Set(other *UnixfsNode) {
 	n.raw = other.raw
 	n.rawnode = other.rawnode
 	if other.ufmt != nil {
-		n.ufmt.Data = other.ufmt.Data
+		n.ufmt.SetData(other.ufmt.GetData())
 	}
 }
 
@@ -127,7 +127,7 @@ func (n *UnixfsNode) RemoveChild(index int, dbh *DagBuilderHelper) {
 
 // SetData stores data in this node.
 func (n *UnixfsNode) SetData(data []byte) {
-	n.ufmt.Data = data
+	n.ufmt.SetData(data)
 }
 
 // FileSize returns the total file size of this tree (including children)

mfs/file.go

@@ -60,7 +60,7 @@ func (fi *File) Open(flags int, sync bool) (FileDescriptor, error) {
 			return nil, err
 		}
 
-		switch fsn.Type {
+		switch fsn.GetType() {
 		default:
 			return nil, fmt.Errorf("unsupported fsnode type for 'file'")
 		case ft.TSymlink:

mfs/mfs_test.go

@@ -852,7 +852,7 @@ func TestFlushing(t *testing.T) {
 		t.Fatal(err)
 	}
 
-	if fsnode.Type != ft.TDirectory {
+	if fsnode.GetType() != ft.TDirectory {
 		t.Fatal("root wasnt a directory")
 	}
 

unixfs/mod/dagmodifier.go

@@ -526,7 +526,7 @@ func dagTruncate(ctx context.Context, n ipld.Node, size uint64, ds ipld.DAGServi
 	var cur uint64
 	end := 0
 	var modified ipld.Node
-	ndata := new(ft.FSNode)
+	ndata := ft.NewFSNode(ft.TRaw)
 	for i, lnk := range nd.Links() {
 		child, err := lnk.GetNode(ctx, ds)
 		if err != nil {

unixfs/unixfs.go

@@ -139,67 +139,101 @@ func DataSize(data []byte) (uint64, error) {
 	}
 }
 
-// An FSNode represents a filesystem object.
+// An FSNode represents a filesystem object using the UnixFS specification.
+//
+// The `NewFSNode` constructor should be used instead of just calling `new(FSNode)`
+// to guarantee that the required (`Type` and `Filesize`) fields in the `format`
+// structure are initialized before marshaling (in `GetBytes()`).
 type FSNode struct {
-	Data []byte
 
-	// total data size for each child
+	// UnixFS format defined as a protocol buffers message.
-	blocksizes []uint64
+	format pb.Data
-
-	// running sum of blocksizes
-	subtotal uint64
-
-	// node type of this node
-	Type pb.Data_DataType
 }
 
 // FSNodeFromBytes unmarshal a protobuf message onto an FSNode.
 func FSNodeFromBytes(b []byte) (*FSNode, error) {
-	pbn := new(pb.Data)
+	n := new(FSNode)
-	err := proto.Unmarshal(b, pbn)
+	err := proto.Unmarshal(b, &n.format)
 	if err != nil {
 		return nil, err
 	}
 
-	n := new(FSNode)
-	n.Data = pbn.Data
-	n.blocksizes = pbn.Blocksizes
-	n.subtotal = pbn.GetFilesize() - uint64(len(n.Data))
-	n.Type = pbn.GetType()
 	return n, nil
 }
 
+// NewFSNode creates a new FSNode structure with the given `dataType`.
+//
+// It initializes the (required) `Type` field (that doesn't have a `Set()`
+// accessor so it must be specified at creation), otherwise the `Marshal()`
+// method in `GetBytes()` would fail (`required field "Type" not set`).
+//
+// It also initializes the `Filesize` pointer field to ensure its value
+// is never nil before marshaling, this is not a required field but it is
+// done to be backwards compatible with previous `go-ipfs` versions hash.
+// (If it wasn't initialized there could be cases where `Filesize` could
+// have been left at nil, when the `FSNode` was created but no data or
+// child nodes were set to adjust it, as is the case in `NewLeaf()`.)
+func NewFSNode(dataType pb.Data_DataType) *FSNode {
+	n := new(FSNode)
+	n.format.Type = &dataType
+
+	// Initialize by `Filesize` by updating it with a dummy (zero) value.
+	n.UpdateFilesize(0)
+
+	return n
+}
+
 // AddBlockSize adds the size of the next child block of this node
 func (n *FSNode) AddBlockSize(s uint64) {
-	n.subtotal += s
+	n.UpdateFilesize(int64(s))
-	n.blocksizes = append(n.blocksizes, s)
+	n.format.Blocksizes = append(n.format.Blocksizes, s)
 }
 
 // RemoveBlockSize removes the given child block's size.
 func (n *FSNode) RemoveBlockSize(i int) {
-	n.subtotal -= n.blocksizes[i]
+	n.UpdateFilesize(-int64(n.format.Blocksizes[i]))
-	n.blocksizes = append(n.blocksizes[:i], n.blocksizes[i+1:]...)
+	n.format.Blocksizes = append(n.format.Blocksizes[:i], n.format.Blocksizes[i+1:]...)
 }
 
 // GetBytes marshals this node as a protobuf message.
 func (n *FSNode) GetBytes() ([]byte, error) {
-	pbn := new(pb.Data)
+	return proto.Marshal(&n.format)
-	pbn.Type = &n.Type
-	pbn.Filesize = proto.Uint64(uint64(len(n.Data)) + n.subtotal)
-	pbn.Blocksizes = n.blocksizes
-	pbn.Data = n.Data
-	return proto.Marshal(pbn)
 }
 
 // FileSize returns the total size of this tree. That is, the size of
 // the data in this node plus the size of all its children.
 func (n *FSNode) FileSize() uint64 {
-	return uint64(len(n.Data)) + n.subtotal
+	return n.format.GetFilesize()
 }
 
 // NumChildren returns the number of child blocks of this node
 func (n *FSNode) NumChildren() int {
-	return len(n.blocksizes)
+	return len(n.format.Blocksizes)
+}
+
+// GetData retrieves the `Data` field from the internal `format`.
+func (n *FSNode) GetData() []byte {
+	return n.format.GetData()
+}
+
+// SetData sets the `Data` field from the internal `format`
+// updating its `Filesize`.
+func (n *FSNode) SetData(newData []byte) {
+	n.UpdateFilesize(int64(len(newData) - len(n.GetData())))
+	n.format.Data = newData
+}
+
+// UpdateFilesize updates the `Filesize` field from the internal `format`
+// by a signed difference (`filesizeDiff`).
+// TODO: Add assert to check for `Filesize` > 0?
+func (n *FSNode) UpdateFilesize(filesizeDiff int64) {
+	n.format.Filesize = proto.Uint64(uint64(
+		int64(n.format.GetFilesize()) + filesizeDiff))
+}
+
+// GetType retrieves the `Type` field from the internal `format`.
+func (n *FSNode) GetType() pb.Data_DataType {
+	return n.format.GetType()
 }
 
 // Metadata is used to store additional FSNode information.

unixfs/unixfs_test.go

@@ -10,14 +10,13 @@ import (
 )
 
 func TestFSNode(t *testing.T) {
-	fsn := new(FSNode)
+	fsn := NewFSNode(TFile)
-	fsn.Type = TFile
 	for i := 0; i < 16; i++ {
 		fsn.AddBlockSize(100)
 	}
 	fsn.RemoveBlockSize(15)
 
-	fsn.Data = make([]byte, 128)
+	fsn.SetData(make([]byte, 128))
 
 	b, err := fsn.GetBytes()
 	if err != nil {