tar/asm: testing the disassembler

adding an archive to pass through and check that it is precisely the
same archive on the handed through io.Reader.

tar/asm/assemble_test.go

@@ -1,48 +1,197 @@
 package asm
 
 import (
+	"bytes"
+	"compress/gzip"
+	"crypto/sha1"
+	"fmt"
+	"io"
+	"io/ioutil"
+	"os"
 	"testing"
 
 	"github.com/vbatts/tar-split/tar/storage"
 )
 
-var entries = storage.Entries{
-	storage.Entry{
+var entries = []struct {
+	Entry storage.Entry
+	Body  []byte
+}{
+	{
+		Entry: storage.Entry{
 			Type:    storage.SegmentType,
-		Payload:  []byte("how"),
-		Position: 0,
+			Payload: []byte("y'all"), // FIXME need real header here
 		},
-	storage.Entry{
-		Type:     storage.SegmentType,
-		Payload:  []byte("y'all"),
-		Position: 1,
 	},
-	storage.Entry{
+	{
+		Entry: storage.Entry{
 			Type:    storage.FileType,
 			Name:    "./hurr.txt",
-		Payload:  []byte("deadbeef"),
-		Size:     8,
-		Position: 2,
+			Payload: []byte{2, 116, 164, 177, 171, 236, 107, 78},
+			Size:    20,
 		},
-	storage.Entry{
+		Body: []byte("imma hurr til I derp"),
+	},
+	{
+		Entry: storage.Entry{
 			Type:    storage.SegmentType,
-		Payload:  []byte("doin"),
-		Position: 3,
+			Payload: []byte("doin"), // FIXME need real header here
 		},
-	storage.Entry{
+	},
+	{
+		Entry: storage.Entry{
 			Type:    storage.FileType,
 			Name:    "./ermahgerd.txt",
-		Payload:  []byte("cafebabe"),
-		Size:     8,
-		Position: 4,
+			Payload: []byte{126, 72, 89, 239, 230, 252, 160, 187},
+			Size:    26,
+		},
+
+		Body: []byte("café con leche, por favor"),
+	},
+	{
+		Entry: storage.Entry{
+			Type: storage.SegmentType,
+			Payload: []byte{00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00,
+				00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00,
+				00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00,
+				00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00,
+				00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00,
+				00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00,
+				00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00,
+				00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00,
+				00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00,
+				00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00,
+				00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00,
+				00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00,
+				00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00,
+				00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00,
+				00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00,
+				00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00,
+				00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00,
+				00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00,
+				00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00,
+				00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00,
+				00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00,
+				00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00,
+				00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00,
+				00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00,
+				00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00,
+				00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00,
+				00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00,
+				00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00,
+				00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00,
+				00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00,
+				00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00,
+				00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00,
+				00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00,
+				00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00,
+				00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00,
+				00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00,
+				00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00,
+				00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00,
+				00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00,
+				00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00,
+				00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00,
+				00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00,
+				00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00,
+				00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00,
+				00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00,
+				00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00,
+				00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00,
+				00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00,
+				00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00,
+				00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00,
+				00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00,
+				00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00,
+				00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00,
+				00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00,
+				00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00,
+				00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00,
+				00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00, 00,
+				00, 00},
+		},
 	},
 }
 
-func TestNewOutputTarStream(t *testing.T) {
-	// TODO disassembly
+func TestTarStream(t *testing.T) {
 	fgp := NewBufferFileGetPutter()
+
+	// first lets prep a GetPutter and Packer
+	for i := range entries {
+		if entries[i].Entry.Type == storage.FileType {
+			j, csum, err := fgp.Put(entries[i].Entry.Name, bytes.NewBuffer(entries[i].Body))
+			if err != nil {
+				t.Error(err)
+			}
+			if j != entries[i].Entry.Size {
+				t.Errorf("size %q: expected %d; got %d",
+					entries[i].Entry.Name,
+					entries[i].Entry.Size,
+					j)
+			}
+			if !bytes.Equal(csum, entries[i].Entry.Payload) {
+				t.Errorf("checksum %q: expected %v; got %v",
+					entries[i].Entry.Name,
+					entries[i].Entry.Payload,
+					csum)
+			}
+		}
+	}
+
+	// next we'll use these to produce a tar stream.
 	_ = NewOutputTarStream(fgp, nil)
+	// TODO finish this
 }
 
-func TestNewInputTarStream(t *testing.T) {
+func TestInputTarStream(t *testing.T) {
+	var (
+		expectedSum        = "1eb237ff69bca6e22789ecb05b45d35ca307adbd"
+		expectedSize int64 = 10240
+	)
+
+	fh, err := os.Open("./testdata/t.tar.gz")
+	if err != nil {
+		t.Fatal(err)
+	}
+	defer fh.Close()
+	gzRdr, err := gzip.NewReader(fh)
+	if err != nil {
+		t.Fatal(err)
+	}
+	defer gzRdr.Close()
+
+	// Setup where we'll store the metadata
+	w := bytes.NewBuffer([]byte{})
+	sp := storage.NewJsonPacker(w)
+	fgp := NewBufferFileGetPutter()
+
+	// check the tar on the front end too
+	h0 := sha1.New()
+	tRdr0 := io.TeeReader(gzRdr, h0)
+
+	// wrap the disassembly stream
+	tarStream, err := NewInputTarStream(tRdr0, sp, fgp)
+	if err != nil {
+		t.Fatal(err)
+	}
+
+	// get a sum of the stream after it has passed through to ensure it's the same.
+	h1 := sha1.New()
+	tRdr1 := io.TeeReader(tarStream, h1)
+
+	// read it all to the bit bucket
+	i, err := io.Copy(ioutil.Discard, tRdr1)
+	if err != nil {
+		t.Fatal(err)
+	}
+
+	if i != expectedSize {
+		t.Errorf("size of tar: expected %d; got %d", expectedSize, i)
+	}
+
+	if fmt.Sprintf("%x", h1.Sum(nil)) != expectedSum {
+		t.Logf("h0 was %x", h0.Sum(nil))
+		t.Errorf("checksum of tar: expected %q; got %x", expectedSum, h1.Sum(nil))
+	}
+
 }

tar/asm/disassemble.go

@@ -36,6 +36,7 @@ func NewInputTarStream(r io.Reader, p storage.Packer, fp FilePutter) (io.Reader,
 	pR, pW := io.Pipe()
 	outputRdr := io.TeeReader(r, pW)
 
+	// we need a putter that will generate the crc64 sums of file payloads
 	if fp == nil {
 		fp = NewDiscardFilePutter()
 	}
@@ -58,10 +59,8 @@ func NewInputTarStream(r io.Reader, p storage.Packer, fp FilePutter) (io.Reader,
 				})
 				if err != nil {
 					pW.CloseWithError(err)
-				} else {
-					pW.Close()
 				}
-				return
+				break // not return. We need the end of the reader.
 			}
 
 			if _, err := p.AddEntry(storage.Entry{
@@ -77,7 +76,7 @@ func NewInputTarStream(r io.Reader, p storage.Packer, fp FilePutter) (io.Reader,
 				fileRdr, fileWrtr := io.Pipe()
 				go func() {
 					var err error
-					csum, err = fp.Put(hdr.Name, fileRdr)
+					_, csum, err = fp.Put(hdr.Name, fileRdr)
 					if err != nil {
 						pW.CloseWithError(err)
 					}
@@ -86,27 +85,30 @@ func NewInputTarStream(r io.Reader, p storage.Packer, fp FilePutter) (io.Reader,
 					pW.CloseWithError(err)
 					return
 				}
+				fileWrtr.Close()
 			}
 			// File entries added, regardless of size
-			if _, err := p.AddEntry(storage.Entry{
+			_, err = p.AddEntry(storage.Entry{
 				Type:    storage.FileType,
 				Name:    hdr.Name,
 				Size:    hdr.Size,
 				Payload: csum,
-			}); err != nil {
+			})
+			if err != nil {
 				pW.CloseWithError(err)
 			}
 
-			if _, err := p.AddEntry(storage.Entry{
+			_, err = p.AddEntry(storage.Entry{
 				Type:    storage.SegmentType,
 				Payload: tr.RawBytes(),
-			}); err != nil {
+			})
+			if err != nil {
 				pW.CloseWithError(err)
 			}
 		}
 
 		// it is allowable, and not uncommon that there is further padding on the
-		// end of an archive, apart from the expected 1024 null bytes
+		// end of an archive, apart from the expected 1024 null bytes.
 		remainder, err := ioutil.ReadAll(outputRdr)
 		if err != nil && err != io.EOF {
 			pW.CloseWithError(err)

tar/asm/getter.go

@@ -16,8 +16,8 @@ type FileGetter interface {
 }
 
 type FilePutter interface {
-	// Put returns a stream for the provided file path
-	Put(string, io.Reader) ([]byte, error)
+	// Put returns the crc64 checksum for the provided file
+	Put(string, io.Reader) (int64, []byte, error)
 }
 
 type FileGetPutter interface {
@@ -51,15 +51,16 @@ func (bfgp bufferFileGetPutter) Get(name string) (io.ReadCloser, error) {
 	return &readCloserWrapper{b}, nil
 }
 
-func (bfgp *bufferFileGetPutter) Put(name string, r io.Reader) ([]byte, error) {
+func (bfgp *bufferFileGetPutter) Put(name string, r io.Reader) (int64, []byte, error) {
 	c := crc64.New(crcTable)
 	tRdr := io.TeeReader(r, c)
 	b := bytes.NewBuffer([]byte{})
-	if _, err := io.Copy(b, tRdr); err != nil {
-		return nil, err
+	i, err := io.Copy(b, tRdr)
+	if err != nil {
+		return 0, nil, err
 	}
 	bfgp.files[name] = b.Bytes()
-	return c.Sum(nil), nil
+	return i, c.Sum(nil), nil
 }
 
 type readCloserWrapper struct {
@@ -86,11 +87,11 @@ func NewDiscardFilePutter() FilePutter {
 type bitBucketFilePutter struct {
 }
 
-func (bbfp *bitBucketFilePutter) Put(name string, r io.Reader) ([]byte, error) {
+func (bbfp *bitBucketFilePutter) Put(name string, r io.Reader) (int64, []byte, error) {
 	c := crc64.New(crcTable)
 	tRdr := io.TeeReader(r, c)
-	_, err := io.Copy(ioutil.Discard, tRdr)
-	return c.Sum(nil), err
+	i, err := io.Copy(ioutil.Discard, tRdr)
+	return i, c.Sum(nil), err
 }
 
 var crcTable = crc64.MakeTable(crc64.ISO)

tar/asm/getter_test.go

@@ -14,7 +14,7 @@ func TestGetter(t *testing.T) {
 	}
 	for n, b := range files {
 		for body, sum := range b {
-			csum, err := fgp.Put(n, bytes.NewBufferString(body))
+			_, csum, err := fgp.Put(n, bytes.NewBufferString(body))
 			if err != nil {
 				t.Error(err)
 			}
@@ -50,7 +50,7 @@ func TestPutter(t *testing.T) {
 	}
 	for n, b := range files {
 		for body, sum := range b {
-			csum, err := fp.Put(n, bytes.NewBufferString(body))
+			_, csum, err := fp.Put(n, bytes.NewBufferString(body))
 			if err != nil {
 				t.Error(err)
 			}