Adds support for bzip2 file format

- Adds new Bzip2Archive type
- Adds support for bzip2 in generic Archive type

go.mod

@@ -7,6 +7,7 @@ require (
 	github.com/Masterminds/semver/v3 v3.1.1
 	github.com/cheggaaa/pb/v3 v3.0.8
 	github.com/docker/distribution v2.7.1+incompatible
+	github.com/dsnet/compress v0.0.1
 	github.com/gabriel-vasile/mimetype v1.3.0
 	github.com/google/go-containerregistry v0.5.1
 	github.com/onsi/gomega v1.13.0

go.sum

@@ -83,6 +83,9 @@ github.com/docker/docker-credential-helpers v0.6.3/go.mod h1:WRaJzqw3CTB9bk10avu
 github.com/docker/go-connections v0.4.0/go.mod h1:Gbd7IOopHjR8Iph03tsViu4nIes5XhDvyHbTtUxmeec=
 github.com/docker/go-units v0.4.0/go.mod h1:fgPhTUdO+D/Jk86RDLlptpiXQzgHJF7gydDDbaIK4Dk=
 github.com/docopt/docopt-go v0.0.0-20180111231733-ee0de3bc6815/go.mod h1:WwZ+bS3ebgob9U8Nd0kOddGdZWjyMGR8Wziv+TBNwSE=
+github.com/dsnet/compress v0.0.1 h1:PlZu0n3Tuv04TzpfPbrnI0HW/YwodEXDS+oPKahKF0Q=
+github.com/dsnet/compress v0.0.1/go.mod h1:Aw8dCMJ7RioblQeTqt88akK31OvO8Dhf5JflhBbQEHo=
+github.com/dsnet/golib v0.0.0-20171103203638-1ea166775780/go.mod h1:Lj+Z9rebOhdfkVLjJ8T6VcRQv3SXugXy999NBtR9aFY=
 github.com/emicklei/go-restful v0.0.0-20170410110728-ff4f55a20633/go.mod h1:otzb+WCGbkyDHkqmQmT5YD2WR4BBwUdeQoFo8l/7tVs=
 github.com/emicklei/go-restful v2.9.5+incompatible/go.mod h1:otzb+WCGbkyDHkqmQmT5YD2WR4BBwUdeQoFo8l/7tVs=
 github.com/envoyproxy/go-control-plane v0.9.0/go.mod h1:YTl/9mNaCwkRvm6d1a2C3ymFceY/DCBVvsKhRF0iEA4=
@@ -211,6 +214,8 @@ github.com/julienschmidt/httprouter v1.2.0/go.mod h1:SYymIcj16QtmaHHD7aYtjjsJG7V
 github.com/kisielk/errcheck v1.1.0/go.mod h1:EZBBE59ingxPouuu3KfxchcWSUPOHkagtvWXihfKN4Q=
 github.com/kisielk/errcheck v1.2.0/go.mod h1:/BMXB+zMLi60iA8Vv6Ksmxu/1UDYcXs4uQLJ+jE2L00=
 github.com/kisielk/gotool v1.0.0/go.mod h1:XhKaO+MFFWcvkIS/tQcRk01m1F5IRFswLeQ+oQHNcck=
+github.com/klauspost/compress v1.4.1/go.mod h1:RyIbtBH6LamlWaDj8nUwkbUhJ87Yi3uG0guNDohfE1A=
+github.com/klauspost/cpuid v1.2.0/go.mod h1:Pj4uuM528wm8OyEC2QMXAi2YiTZ96dNQPGgoMS4s3ek=
 github.com/konsorten/go-windows-terminal-sequences v1.0.1/go.mod h1:T0+1ngSBFLxvqU3pZ+m/2kptfBszLMUkC4ZK/EgS/cQ=
 github.com/konsorten/go-windows-terminal-sequences v1.0.3/go.mod h1:T0+1ngSBFLxvqU3pZ+m/2kptfBszLMUkC4ZK/EgS/cQ=
 github.com/kr/logfmt v0.0.0-20140226030751-b84e30acd515/go.mod h1:+0opPa2QZZtGFBFZlji/RkVcI2GknAs/DXo4wKdlNEc=
@@ -339,6 +344,7 @@ github.com/stretchr/testify v1.6.1/go.mod h1:6Fq8oRcR53rry900zMqJjRRixrwX3KX962/
 github.com/subosito/gotenv v1.2.0/go.mod h1:N0PQaV/YGNqwC0u51sEeR/aUtSLEXKX9iv69rRypqCw=
 github.com/tmc/grpc-websocket-proxy v0.0.0-20190109142713-0ad062ec5ee5/go.mod h1:ncp9v5uamzpCO7NfCPTXjqaC+bZgJeR0sMTm6dMHP7U=
 github.com/ugorji/go v1.1.4/go.mod h1:uQMGLiO92mf5W77hV/PUCpI3pbzQx3CRekS0kk+RGrc=
+github.com/ulikunitz/xz v0.5.6/go.mod h1:2bypXElzHzzJZwzH67Y6wb67pO62Rzfn7BSiF4ABRW8=
 github.com/ulikunitz/xz v0.5.10 h1:t92gobL9l3HE202wg3rlk19F6X+JOxl9BBrCCMYEYd8=
 github.com/ulikunitz/xz v0.5.10/go.mod h1:nbz6k7qbPmH4IRqmfOplQw/tblSgqTqBwxkY0oWt/14=
 github.com/xiang90/probing v0.0.0-20190116061207-43a291ad63a2/go.mod h1:UETIi67q53MR2AWcXfiuqkDkRtnGDLqkBTpCHuJHxtU=

vacation/example_test.go

@@ -10,6 +10,7 @@ import (
 	"os"
 	"path/filepath"
 
+	dsnetBzip2 "github.com/dsnet/compress/bzip2"
 	"github.com/paketo-buildpacks/packit/vacation"
 	"github.com/ulikunitz/xz"
 )
@@ -567,6 +568,147 @@ func ExampleTarXZArchive_StripComponents() {
 	// some-other-dir/some-file
 }
 
+func ExampleTarBzip2Archive() {
+	buffer := bytes.NewBuffer(nil)
+
+	// Using the dsnet library because the Go compression library does not
+	// have a writer. There is recent discussion on this issue
+	// https://github.com/golang/go/issues/4828 to add an encoder. The
+	// library should be removed once there is a native encoder
+	bz, err := dsnetBzip2.NewWriter(buffer, nil)
+	if err != nil {
+		log.Fatal(err)
+	}
+
+	tw := tar.NewWriter(bz)
+
+	files := []ArchiveFile{
+		{Name: "some-dir/"},
+		{Name: "some-dir/some-other-dir/"},
+		{Name: "some-dir/some-other-dir/some-file", Content: []byte("some-dir/some-other-dir/some-file")},
+		{Name: "first", Content: []byte("first")},
+		{Name: "second", Content: []byte("second")},
+		{Name: "third", Content: []byte("third")},
+	}
+
+	for _, file := range files {
+		err := tw.WriteHeader(&tar.Header{Name: file.Name, Mode: 0755, Size: int64(len(file.Content))})
+		if err != nil {
+			log.Fatal(err)
+		}
+
+		_, err = tw.Write(file.Content)
+		if err != nil {
+			log.Fatal(err)
+		}
+	}
+
+	tw.Close()
+	bz.Close()
+
+	destination, err := os.MkdirTemp("", "destination")
+	if err != nil {
+		log.Fatal(err)
+	}
+	defer os.RemoveAll(destination)
+
+	archive := vacation.NewTarBzip2Archive(bytes.NewReader(buffer.Bytes()))
+	if err := archive.Decompress(destination); err != nil {
+		log.Fatal(err)
+	}
+
+	err = filepath.Walk(destination, func(path string, info os.FileInfo, err error) error {
+		if !info.IsDir() {
+			rel, err := filepath.Rel(destination, path)
+			if err != nil {
+				log.Fatal(err)
+			}
+
+			fmt.Printf("%s\n", rel)
+			return nil
+		}
+		return nil
+	})
+	if err != nil {
+		log.Fatal(err)
+	}
+
+	// Output:
+	// first
+	// second
+	// some-dir/some-other-dir/some-file
+	// third
+}
+
+func ExampleTarBzip2Archive_StripComponents() {
+	buffer := bytes.NewBuffer(nil)
+
+	// Using the dsnet library because the Go compression library does not
+	// have a writer. There is recent discussion on this issue
+	// https://github.com/golang/go/issues/4828 to add an encoder. The
+	// library should be removed once there is a native encoder
+	bz, err := dsnetBzip2.NewWriter(buffer, nil)
+	if err != nil {
+		log.Fatal(err)
+	}
+
+	tw := tar.NewWriter(bz)
+
+	files := []ArchiveFile{
+		{Name: "some-dir/"},
+		{Name: "some-dir/some-other-dir/"},
+		{Name: "some-dir/some-other-dir/some-file", Content: []byte("some-dir/some-other-dir/some-file")},
+		{Name: "first", Content: []byte("first")},
+		{Name: "second", Content: []byte("second")},
+		{Name: "third", Content: []byte("third")},
+	}
+
+	for _, file := range files {
+		err := tw.WriteHeader(&tar.Header{Name: file.Name, Mode: 0755, Size: int64(len(file.Content))})
+		if err != nil {
+			log.Fatal(err)
+		}
+
+		_, err = tw.Write(file.Content)
+		if err != nil {
+			log.Fatal(err)
+		}
+	}
+
+	tw.Close()
+	bz.Close()
+
+	destination, err := os.MkdirTemp("", "destination")
+	if err != nil {
+		log.Fatal(err)
+	}
+	defer os.RemoveAll(destination)
+
+	archive := vacation.NewTarBzip2Archive(bytes.NewReader(buffer.Bytes())).StripComponents(1)
+	if err := archive.Decompress(destination); err != nil {
+		log.Fatal(err)
+	}
+
+	err = filepath.Walk(destination, func(path string, info os.FileInfo, err error) error {
+		if !info.IsDir() {
+			rel, err := filepath.Rel(destination, path)
+			if err != nil {
+				log.Fatal(err)
+			}
+
+			fmt.Printf("%s\n", rel)
+			return nil
+		}
+		return nil
+	})
+	if err != nil {
+		log.Fatal(err)
+	}
+
+	// Output:
+	// some-other-dir/some-file
+}
+
 func ExampleZipArchive() {
 	buffer := bytes.NewBuffer(nil)
 	zw := zip.NewWriter(buffer)

vacation/init_test.go

@@ -10,6 +10,7 @@ import (
 func TestVacation(t *testing.T) {
 	suite := spec.New("vacation", spec.Report(report.Terminal{}))
 	suite("VacationArchive", testVacationArchive)
+	suite("VacationTarBzip2", testVacationTarBzip2)
 	suite("VacationSymlinkSorting", testVacationSymlinkSorting)
 	suite("VacationTar", testVacationTar)
 	suite("VacationTarGzip", testVacationTarGzip)

vacation/vacation.go

@@ -8,6 +8,7 @@ import (
 	"archive/tar"
 	"archive/zip"
 	"bufio"
+	"compress/bzip2"
 	"compress/gzip"
 	"fmt"
 	"io"
@@ -45,6 +46,12 @@ type TarXZArchive struct {
 	components int
 }
 
+// A TarBzip2Archive decompresses bzip2 files from an input stream.
+type TarBzip2Archive struct {
+	reader     io.Reader
+	components int
+}
+
 // NewArchive returns a new Archive that reads from inputReader.
 func NewArchive(inputReader io.Reader) Archive {
 	return Archive{reader: inputReader}
@@ -65,6 +72,11 @@ func NewTarXZArchive(inputReader io.Reader) TarXZArchive {
 	return TarXZArchive{reader: inputReader}
 }
 
+// NewTarBzip2Archive returns a new Bzip2Archive that reads from inputReader.
+func NewTarBzip2Archive(inputReader io.Reader) TarBzip2Archive {
+	return TarBzip2Archive{reader: inputReader}
+}
+
 // Decompress reads from TarArchive and writes files into the
 // destination specified.
 func (ta TarArchive) Decompress(destination string) error {
@@ -245,6 +257,8 @@ func (a Archive) Decompress(destination string) error {
 		return NewTarGzipArchive(bufferedReader).StripComponents(a.components).Decompress(destination)
 	case "application/x-xz":
 		return NewTarXZArchive(bufferedReader).StripComponents(a.components).Decompress(destination)
+	case "application/x-bzip2":
+		return NewTarBzip2Archive(bufferedReader).StripComponents(a.components).Decompress(destination)
 	case "application/zip":
 		return NewZipArchive(bufferedReader).Decompress(destination)
 	case "text/plain; charset=utf-8":
@@ -278,6 +292,12 @@ func (txz TarXZArchive) Decompress(destination string) error {
 	return NewTarArchive(xzr).StripComponents(txz.components).Decompress(destination)
 }
 
+// Decompress reads from TarBzip2Archive and writes files into the destination
+// specified.
+func (tbz TarBzip2Archive) Decompress(destination string) error {
+	return NewTarArchive(bzip2.NewReader(tbz.reader)).StripComponents(tbz.components).Decompress(destination)
+}
+
 func writeTextFile(reader io.Reader, destination string) error {
 	file, err := os.Create(filepath.Join(destination, "artifact"))
 	if err != nil {
@@ -322,6 +342,13 @@ func (txz TarXZArchive) StripComponents(components int) TarXZArchive {
 	return txz
 }
 
+// StripComponents behaves like the --strip-components flag on tar command
+// removing the first n levels from the final decompression destination.
+func (tbz TarBzip2Archive) StripComponents(components int) TarBzip2Archive {
+	tbz.components = components
+	return tbz
+}
+
 // A ZipArchive decompresses zip files from an input stream.
 type ZipArchive struct {
 	reader io.Reader

vacation/vacation_archive_test.go

@@ -9,6 +9,7 @@ import (
 	"path/filepath"
 	"testing"
 
+	dsnetBzip2 "github.com/dsnet/compress/bzip2"
 	"github.com/paketo-buildpacks/packit/vacation"
 	"github.com/sclevine/spec"
 	"github.com/ulikunitz/xz"
@@ -208,6 +209,75 @@ func testVacationArchive(t *testing.T, context spec.G, it spec.S) {
 			})
 		})
 
+		context("when passed the reader of a bzip2 file", func() {
+			var (
+				archive vacation.Archive
+				tempDir string
+			)
+
+			it.Before(func() {
+				var err error
+				tempDir, err = os.MkdirTemp("", "vacation")
+				Expect(err).NotTo(HaveOccurred())
+
+				buffer := bytes.NewBuffer(nil)
+
+				// Using the dsnet library because the Go compression library does not
+				// have a writer. There is recent discussion on this issue
+				// https://github.com/golang/go/issues/4828 to add an encoder. The
+				// library should be removed once there is a native encoder
+				bz, err := dsnetBzip2.NewWriter(buffer, nil)
+				Expect(err).NotTo(HaveOccurred())
+
+				tw := tar.NewWriter(bz)
+
+				Expect(tw.WriteHeader(&tar.Header{Name: "some-dir", Mode: 0755, Typeflag: tar.TypeDir})).To(Succeed())
+				_, err = tw.Write(nil)
+				Expect(err).NotTo(HaveOccurred())
+
+				nestedFile := filepath.Join("some-dir", "some-nested-file")
+				Expect(tw.WriteHeader(&tar.Header{Name: nestedFile, Mode: 0755, Size: int64(len(nestedFile))})).To(Succeed())
+				_, err = tw.Write([]byte(nestedFile))
+				Expect(err).NotTo(HaveOccurred())
+
+				Expect(tw.WriteHeader(&tar.Header{Name: "some-file", Mode: 0755, Size: int64(len("some-file"))})).To(Succeed())
+				_, err = tw.Write([]byte("some-file"))
+				Expect(err).NotTo(HaveOccurred())
+
+				Expect(tw.Close()).To(Succeed())
+				Expect(bz.Close()).To(Succeed())
+
+				archive = vacation.NewArchive(buffer)
+			})
+
+			it.After(func() {
+				Expect(os.RemoveAll(tempDir)).To(Succeed())
+			})
+
+			it("unpackages the archive into the path", func() {
+				err := archive.Decompress(tempDir)
+				Expect(err).NotTo(HaveOccurred())
+
+				files, err := filepath.Glob(filepath.Join(tempDir, "*"))
+				Expect(err).NotTo(HaveOccurred())
+				Expect(files).To(ConsistOf([]string{
+					filepath.Join(tempDir, "some-dir"),
+					filepath.Join(tempDir, "some-file"),
+				}))
+			})
+
+			it("unpackages the archive into the path but also strips the first component", func() {
+				err := archive.StripComponents(1).Decompress(tempDir)
+				Expect(err).NotTo(HaveOccurred())
+
+				files, err := filepath.Glob(filepath.Join(tempDir, "*"))
+				Expect(err).NotTo(HaveOccurred())
+				Expect(files).To(ConsistOf([]string{
+					filepath.Join(tempDir, "some-nested-file"),
+				}))
+			})
+		})
+
 		context("when passed the reader of a zip file", func() {
 			var (
 				archive vacation.Archive