summaryrefslogtreecommitdiff
path: root/src/archive
diff options
context:
space:
mode:
authorRuss Cox <rsc@golang.org>2014-09-08 00:08:51 -0400
committerRuss Cox <rsc@golang.org>2014-09-08 00:08:51 -0400
commit8528da672cc093d4dd06732819abc1f7b6b5a46e (patch)
tree334be80d4a4c85b77db6f6fdb67cbf0528cba5f5 /src/archive
parent73bcb69f272cbf34ddcc9daa56427a8683b5a95d (diff)
downloadgo-8528da672cc093d4dd06732819abc1f7b6b5a46e.tar.gz
build: move package sources from src/pkg to src
Preparation was in CL 134570043. This CL contains only the effect of 'hg mv src/pkg/* src'. For more about the move, see golang.org/s/go14nopkg.
Diffstat (limited to 'src/archive')
-rw-r--r--src/archive/tar/common.go305
-rw-r--r--src/archive/tar/example_test.go79
-rw-r--r--src/archive/tar/reader.go820
-rw-r--r--src/archive/tar/reader_test.go743
-rw-r--r--src/archive/tar/stat_atim.go20
-rw-r--r--src/archive/tar/stat_atimespec.go20
-rw-r--r--src/archive/tar/stat_unix.go32
-rw-r--r--src/archive/tar/tar_test.go284
-rw-r--r--src/archive/tar/testdata/gnu.tarbin0 -> 3072 bytes
-rw-r--r--src/archive/tar/testdata/nil-uid.tarbin0 -> 1024 bytes
-rw-r--r--src/archive/tar/testdata/pax.tarbin0 -> 10240 bytes
-rw-r--r--src/archive/tar/testdata/small.txt1
-rw-r--r--src/archive/tar/testdata/small2.txt1
-rw-r--r--src/archive/tar/testdata/sparse-formats.tarbin0 -> 17920 bytes
-rw-r--r--src/archive/tar/testdata/star.tarbin0 -> 3072 bytes
-rw-r--r--src/archive/tar/testdata/ustar.tarbin0 -> 2048 bytes
-rw-r--r--src/archive/tar/testdata/v7.tarbin0 -> 3584 bytes
-rw-r--r--src/archive/tar/testdata/writer-big-long.tarbin0 -> 4096 bytes
-rw-r--r--src/archive/tar/testdata/writer-big.tarbin0 -> 4096 bytes
-rw-r--r--src/archive/tar/testdata/writer.tarbin0 -> 3584 bytes
-rw-r--r--src/archive/tar/testdata/xattrs.tarbin0 -> 5120 bytes
-rw-r--r--src/archive/tar/writer.go396
-rw-r--r--src/archive/tar/writer_test.go491
-rw-r--r--src/archive/zip/example_test.go75
-rw-r--r--src/archive/zip/reader.go453
-rw-r--r--src/archive/zip/reader_test.go533
-rw-r--r--src/archive/zip/register.go110
-rw-r--r--src/archive/zip/struct.go313
-rw-r--r--src/archive/zip/testdata/crc32-not-streamed.zipbin0 -> 314 bytes
-rw-r--r--src/archive/zip/testdata/dd.zipbin0 -> 154 bytes
-rw-r--r--src/archive/zip/testdata/go-no-datadesc-sig.zipbin0 -> 330 bytes
-rw-r--r--src/archive/zip/testdata/go-with-datadesc-sig.zipbin0 -> 242 bytes
-rw-r--r--src/archive/zip/testdata/gophercolor16x16.pngbin0 -> 785 bytes
-rw-r--r--src/archive/zip/testdata/readme.notzipbin0 -> 1905 bytes
-rw-r--r--src/archive/zip/testdata/readme.zipbin0 -> 1885 bytes
-rw-r--r--src/archive/zip/testdata/symlink.zipbin0 -> 173 bytes
-rw-r--r--src/archive/zip/testdata/test-trailing-junk.zipbin0 -> 1184 bytes
-rw-r--r--src/archive/zip/testdata/test.zipbin0 -> 1170 bytes
-rw-r--r--src/archive/zip/testdata/unix.zipbin0 -> 620 bytes
-rw-r--r--src/archive/zip/testdata/winxp.zipbin0 -> 412 bytes
-rw-r--r--src/archive/zip/testdata/zip64-2.zipbin0 -> 266 bytes
-rw-r--r--src/archive/zip/testdata/zip64.zipbin0 -> 242 bytes
-rw-r--r--src/archive/zip/writer.go357
-rw-r--r--src/archive/zip/writer_test.go164
-rw-r--r--src/archive/zip/zip_test.go395
45 files changed, 5592 insertions, 0 deletions
diff --git a/src/archive/tar/common.go b/src/archive/tar/common.go
new file mode 100644
index 000000000..e363aa793
--- /dev/null
+++ b/src/archive/tar/common.go
@@ -0,0 +1,305 @@
+// Copyright 2009 The Go Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+// Package tar implements access to tar archives.
+// It aims to cover most of the variations, including those produced
+// by GNU and BSD tars.
+//
+// References:
+// http://www.freebsd.org/cgi/man.cgi?query=tar&sektion=5
+// http://www.gnu.org/software/tar/manual/html_node/Standard.html
+// http://pubs.opengroup.org/onlinepubs/9699919799/utilities/pax.html
+package tar
+
+import (
+ "bytes"
+ "errors"
+ "fmt"
+ "os"
+ "path"
+ "time"
+)
+
+const (
+ blockSize = 512
+
+ // Types
+ TypeReg = '0' // regular file
+ TypeRegA = '\x00' // regular file
+ TypeLink = '1' // hard link
+ TypeSymlink = '2' // symbolic link
+ TypeChar = '3' // character device node
+ TypeBlock = '4' // block device node
+ TypeDir = '5' // directory
+ TypeFifo = '6' // fifo node
+ TypeCont = '7' // reserved
+ TypeXHeader = 'x' // extended header
+ TypeXGlobalHeader = 'g' // global extended header
+ TypeGNULongName = 'L' // Next file has a long name
+ TypeGNULongLink = 'K' // Next file symlinks to a file w/ a long name
+ TypeGNUSparse = 'S' // sparse file
+)
+
+// A Header represents a single header in a tar archive.
+// Some fields may not be populated.
+type Header struct {
+ Name string // name of header file entry
+ Mode int64 // permission and mode bits
+ Uid int // user id of owner
+ Gid int // group id of owner
+ Size int64 // length in bytes
+ ModTime time.Time // modified time
+ Typeflag byte // type of header entry
+ Linkname string // target name of link
+ Uname string // user name of owner
+ Gname string // group name of owner
+ Devmajor int64 // major number of character or block device
+ Devminor int64 // minor number of character or block device
+ AccessTime time.Time // access time
+ ChangeTime time.Time // status change time
+ Xattrs map[string]string
+}
+
+// File name constants from the tar spec.
+const (
+ fileNameSize = 100 // Maximum number of bytes in a standard tar name.
+ fileNamePrefixSize = 155 // Maximum number of ustar extension bytes.
+)
+
+// FileInfo returns an os.FileInfo for the Header.
+func (h *Header) FileInfo() os.FileInfo {
+ return headerFileInfo{h}
+}
+
+// headerFileInfo implements os.FileInfo.
+type headerFileInfo struct {
+ h *Header
+}
+
+func (fi headerFileInfo) Size() int64 { return fi.h.Size }
+func (fi headerFileInfo) IsDir() bool { return fi.Mode().IsDir() }
+func (fi headerFileInfo) ModTime() time.Time { return fi.h.ModTime }
+func (fi headerFileInfo) Sys() interface{} { return fi.h }
+
+// Name returns the base name of the file.
+func (fi headerFileInfo) Name() string {
+ if fi.IsDir() {
+ return path.Base(path.Clean(fi.h.Name))
+ }
+ return path.Base(fi.h.Name)
+}
+
+// Mode returns the permission and mode bits for the headerFileInfo.
+func (fi headerFileInfo) Mode() (mode os.FileMode) {
+ // Set file permission bits.
+ mode = os.FileMode(fi.h.Mode).Perm()
+
+ // Set setuid, setgid and sticky bits.
+ if fi.h.Mode&c_ISUID != 0 {
+ // setuid
+ mode |= os.ModeSetuid
+ }
+ if fi.h.Mode&c_ISGID != 0 {
+ // setgid
+ mode |= os.ModeSetgid
+ }
+ if fi.h.Mode&c_ISVTX != 0 {
+ // sticky
+ mode |= os.ModeSticky
+ }
+
+ // Set file mode bits.
+ // clear perm, setuid, setgid and sticky bits.
+ m := os.FileMode(fi.h.Mode) &^ 07777
+ if m == c_ISDIR {
+ // directory
+ mode |= os.ModeDir
+ }
+ if m == c_ISFIFO {
+ // named pipe (FIFO)
+ mode |= os.ModeNamedPipe
+ }
+ if m == c_ISLNK {
+ // symbolic link
+ mode |= os.ModeSymlink
+ }
+ if m == c_ISBLK {
+ // device file
+ mode |= os.ModeDevice
+ }
+ if m == c_ISCHR {
+ // Unix character device
+ mode |= os.ModeDevice
+ mode |= os.ModeCharDevice
+ }
+ if m == c_ISSOCK {
+ // Unix domain socket
+ mode |= os.ModeSocket
+ }
+
+ switch fi.h.Typeflag {
+ case TypeLink, TypeSymlink:
+ // hard link, symbolic link
+ mode |= os.ModeSymlink
+ case TypeChar:
+ // character device node
+ mode |= os.ModeDevice
+ mode |= os.ModeCharDevice
+ case TypeBlock:
+ // block device node
+ mode |= os.ModeDevice
+ case TypeDir:
+ // directory
+ mode |= os.ModeDir
+ case TypeFifo:
+ // fifo node
+ mode |= os.ModeNamedPipe
+ }
+
+ return mode
+}
+
+// sysStat, if non-nil, populates h from system-dependent fields of fi.
+var sysStat func(fi os.FileInfo, h *Header) error
+
+// Mode constants from the tar spec.
+const (
+ c_ISUID = 04000 // Set uid
+ c_ISGID = 02000 // Set gid
+ c_ISVTX = 01000 // Save text (sticky bit)
+ c_ISDIR = 040000 // Directory
+ c_ISFIFO = 010000 // FIFO
+ c_ISREG = 0100000 // Regular file
+ c_ISLNK = 0120000 // Symbolic link
+ c_ISBLK = 060000 // Block special file
+ c_ISCHR = 020000 // Character special file
+ c_ISSOCK = 0140000 // Socket
+)
+
+// Keywords for the PAX Extended Header
+const (
+ paxAtime = "atime"
+ paxCharset = "charset"
+ paxComment = "comment"
+ paxCtime = "ctime" // please note that ctime is not a valid pax header.
+ paxGid = "gid"
+ paxGname = "gname"
+ paxLinkpath = "linkpath"
+ paxMtime = "mtime"
+ paxPath = "path"
+ paxSize = "size"
+ paxUid = "uid"
+ paxUname = "uname"
+ paxXattr = "SCHILY.xattr."
+ paxNone = ""
+)
+
+// FileInfoHeader creates a partially-populated Header from fi.
+// If fi describes a symlink, FileInfoHeader records link as the link target.
+// If fi describes a directory, a slash is appended to the name.
+// Because os.FileInfo's Name method returns only the base name of
+// the file it describes, it may be necessary to modify the Name field
+// of the returned header to provide the full path name of the file.
+func FileInfoHeader(fi os.FileInfo, link string) (*Header, error) {
+ if fi == nil {
+ return nil, errors.New("tar: FileInfo is nil")
+ }
+ fm := fi.Mode()
+ h := &Header{
+ Name: fi.Name(),
+ ModTime: fi.ModTime(),
+ Mode: int64(fm.Perm()), // or'd with c_IS* constants later
+ }
+ switch {
+ case fm.IsRegular():
+ h.Mode |= c_ISREG
+ h.Typeflag = TypeReg
+ h.Size = fi.Size()
+ case fi.IsDir():
+ h.Typeflag = TypeDir
+ h.Mode |= c_ISDIR
+ h.Name += "/"
+ case fm&os.ModeSymlink != 0:
+ h.Typeflag = TypeSymlink
+ h.Mode |= c_ISLNK
+ h.Linkname = link
+ case fm&os.ModeDevice != 0:
+ if fm&os.ModeCharDevice != 0 {
+ h.Mode |= c_ISCHR
+ h.Typeflag = TypeChar
+ } else {
+ h.Mode |= c_ISBLK
+ h.Typeflag = TypeBlock
+ }
+ case fm&os.ModeNamedPipe != 0:
+ h.Typeflag = TypeFifo
+ h.Mode |= c_ISFIFO
+ case fm&os.ModeSocket != 0:
+ h.Mode |= c_ISSOCK
+ default:
+ return nil, fmt.Errorf("archive/tar: unknown file mode %v", fm)
+ }
+ if fm&os.ModeSetuid != 0 {
+ h.Mode |= c_ISUID
+ }
+ if fm&os.ModeSetgid != 0 {
+ h.Mode |= c_ISGID
+ }
+ if fm&os.ModeSticky != 0 {
+ h.Mode |= c_ISVTX
+ }
+ if sysStat != nil {
+ return h, sysStat(fi, h)
+ }
+ return h, nil
+}
+
+var zeroBlock = make([]byte, blockSize)
+
+// POSIX specifies a sum of the unsigned byte values, but the Sun tar uses signed byte values.
+// We compute and return both.
+func checksum(header []byte) (unsigned int64, signed int64) {
+ for i := 0; i < len(header); i++ {
+ if i == 148 {
+ // The chksum field (header[148:156]) is special: it should be treated as space bytes.
+ unsigned += ' ' * 8
+ signed += ' ' * 8
+ i += 7
+ continue
+ }
+ unsigned += int64(header[i])
+ signed += int64(int8(header[i]))
+ }
+ return
+}
+
+type slicer []byte
+
+func (sp *slicer) next(n int) (b []byte) {
+ s := *sp
+ b, *sp = s[0:n], s[n:]
+ return
+}
+
+func isASCII(s string) bool {
+ for _, c := range s {
+ if c >= 0x80 {
+ return false
+ }
+ }
+ return true
+}
+
+func toASCII(s string) string {
+ if isASCII(s) {
+ return s
+ }
+ var buf bytes.Buffer
+ for _, c := range s {
+ if c < 0x80 {
+ buf.WriteByte(byte(c))
+ }
+ }
+ return buf.String()
+}
diff --git a/src/archive/tar/example_test.go b/src/archive/tar/example_test.go
new file mode 100644
index 000000000..351eaa0e6
--- /dev/null
+++ b/src/archive/tar/example_test.go
@@ -0,0 +1,79 @@
+// Copyright 2013 The Go Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+package tar_test
+
+import (
+ "archive/tar"
+ "bytes"
+ "fmt"
+ "io"
+ "log"
+ "os"
+)
+
+func Example() {
+ // Create a buffer to write our archive to.
+ buf := new(bytes.Buffer)
+
+ // Create a new tar archive.
+ tw := tar.NewWriter(buf)
+
+ // Add some files to the archive.
+ var files = []struct {
+ Name, Body string
+ }{
+ {"readme.txt", "This archive contains some text files."},
+ {"gopher.txt", "Gopher names:\nGeorge\nGeoffrey\nGonzo"},
+ {"todo.txt", "Get animal handling licence."},
+ }
+ for _, file := range files {
+ hdr := &tar.Header{
+ Name: file.Name,
+ Size: int64(len(file.Body)),
+ }
+ if err := tw.WriteHeader(hdr); err != nil {
+ log.Fatalln(err)
+ }
+ if _, err := tw.Write([]byte(file.Body)); err != nil {
+ log.Fatalln(err)
+ }
+ }
+ // Make sure to check the error on Close.
+ if err := tw.Close(); err != nil {
+ log.Fatalln(err)
+ }
+
+ // Open the tar archive for reading.
+ r := bytes.NewReader(buf.Bytes())
+ tr := tar.NewReader(r)
+
+ // Iterate through the files in the archive.
+ for {
+ hdr, err := tr.Next()
+ if err == io.EOF {
+ // end of tar archive
+ break
+ }
+ if err != nil {
+ log.Fatalln(err)
+ }
+ fmt.Printf("Contents of %s:\n", hdr.Name)
+ if _, err := io.Copy(os.Stdout, tr); err != nil {
+ log.Fatalln(err)
+ }
+ fmt.Println()
+ }
+
+ // Output:
+ // Contents of readme.txt:
+ // This archive contains some text files.
+ // Contents of gopher.txt:
+ // Gopher names:
+ // George
+ // Geoffrey
+ // Gonzo
+ // Contents of todo.txt:
+ // Get animal handling licence.
+}
diff --git a/src/archive/tar/reader.go b/src/archive/tar/reader.go
new file mode 100644
index 000000000..a27559d0f
--- /dev/null
+++ b/src/archive/tar/reader.go
@@ -0,0 +1,820 @@
+// Copyright 2009 The Go Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+package tar
+
+// TODO(dsymonds):
+// - pax extensions
+
+import (
+ "bytes"
+ "errors"
+ "io"
+ "io/ioutil"
+ "os"
+ "strconv"
+ "strings"
+ "time"
+)
+
+var (
+ ErrHeader = errors.New("archive/tar: invalid tar header")
+)
+
+const maxNanoSecondIntSize = 9
+
+// A Reader provides sequential access to the contents of a tar archive.
+// A tar archive consists of a sequence of files.
+// The Next method advances to the next file in the archive (including the first),
+// and then it can be treated as an io.Reader to access the file's data.
+type Reader struct {
+ r io.Reader
+ err error
+ pad int64 // amount of padding (ignored) after current file entry
+ curr numBytesReader // reader for current file entry
+ hdrBuff [blockSize]byte // buffer to use in readHeader
+}
+
+// A numBytesReader is an io.Reader with a numBytes method, returning the number
+// of bytes remaining in the underlying encoded data.
+type numBytesReader interface {
+ io.Reader
+ numBytes() int64
+}
+
+// A regFileReader is a numBytesReader for reading file data from a tar archive.
+type regFileReader struct {
+ r io.Reader // underlying reader
+ nb int64 // number of unread bytes for current file entry
+}
+
+// A sparseFileReader is a numBytesReader for reading sparse file data from a tar archive.
+type sparseFileReader struct {
+ rfr *regFileReader // reads the sparse-encoded file data
+ sp []sparseEntry // the sparse map for the file
+ pos int64 // keeps track of file position
+ tot int64 // total size of the file
+}
+
+// Keywords for GNU sparse files in a PAX extended header
+const (
+ paxGNUSparseNumBlocks = "GNU.sparse.numblocks"
+ paxGNUSparseOffset = "GNU.sparse.offset"
+ paxGNUSparseNumBytes = "GNU.sparse.numbytes"
+ paxGNUSparseMap = "GNU.sparse.map"
+ paxGNUSparseName = "GNU.sparse.name"
+ paxGNUSparseMajor = "GNU.sparse.major"
+ paxGNUSparseMinor = "GNU.sparse.minor"
+ paxGNUSparseSize = "GNU.sparse.size"
+ paxGNUSparseRealSize = "GNU.sparse.realsize"
+)
+
+// Keywords for old GNU sparse headers
+const (
+ oldGNUSparseMainHeaderOffset = 386
+ oldGNUSparseMainHeaderIsExtendedOffset = 482
+ oldGNUSparseMainHeaderNumEntries = 4
+ oldGNUSparseExtendedHeaderIsExtendedOffset = 504
+ oldGNUSparseExtendedHeaderNumEntries = 21
+ oldGNUSparseOffsetSize = 12
+ oldGNUSparseNumBytesSize = 12
+)
+
+// NewReader creates a new Reader reading from r.
+func NewReader(r io.Reader) *Reader { return &Reader{r: r} }
+
+// Next advances to the next entry in the tar archive.
+func (tr *Reader) Next() (*Header, error) {
+ var hdr *Header
+ if tr.err == nil {
+ tr.skipUnread()
+ }
+ if tr.err != nil {
+ return hdr, tr.err
+ }
+ hdr = tr.readHeader()
+ if hdr == nil {
+ return hdr, tr.err
+ }
+ // Check for PAX/GNU header.
+ switch hdr.Typeflag {
+ case TypeXHeader:
+ // PAX extended header
+ headers, err := parsePAX(tr)
+ if err != nil {
+ return nil, err
+ }
+ // We actually read the whole file,
+ // but this skips alignment padding
+ tr.skipUnread()
+ hdr = tr.readHeader()
+ mergePAX(hdr, headers)
+
+ // Check for a PAX format sparse file
+ sp, err := tr.checkForGNUSparsePAXHeaders(hdr, headers)
+ if err != nil {
+ tr.err = err
+ return nil, err
+ }
+ if sp != nil {
+ // Current file is a PAX format GNU sparse file.
+ // Set the current file reader to a sparse file reader.
+ tr.curr = &sparseFileReader{rfr: tr.curr.(*regFileReader), sp: sp, tot: hdr.Size}
+ }
+ return hdr, nil
+ case TypeGNULongName:
+ // We have a GNU long name header. Its contents are the real file name.
+ realname, err := ioutil.ReadAll(tr)
+ if err != nil {
+ return nil, err
+ }
+ hdr, err := tr.Next()
+ hdr.Name = cString(realname)
+ return hdr, err
+ case TypeGNULongLink:
+ // We have a GNU long link header.
+ realname, err := ioutil.ReadAll(tr)
+ if err != nil {
+ return nil, err
+ }
+ hdr, err := tr.Next()
+ hdr.Linkname = cString(realname)
+ return hdr, err
+ }
+ return hdr, tr.err
+}
+
+// checkForGNUSparsePAXHeaders checks the PAX headers for GNU sparse headers. If they are found, then
+// this function reads the sparse map and returns it. Unknown sparse formats are ignored, causing the file to
+// be treated as a regular file.
+func (tr *Reader) checkForGNUSparsePAXHeaders(hdr *Header, headers map[string]string) ([]sparseEntry, error) {
+ var sparseFormat string
+
+ // Check for sparse format indicators
+ major, majorOk := headers[paxGNUSparseMajor]
+ minor, minorOk := headers[paxGNUSparseMinor]
+ sparseName, sparseNameOk := headers[paxGNUSparseName]
+ _, sparseMapOk := headers[paxGNUSparseMap]
+ sparseSize, sparseSizeOk := headers[paxGNUSparseSize]
+ sparseRealSize, sparseRealSizeOk := headers[paxGNUSparseRealSize]
+
+ // Identify which, if any, sparse format applies from which PAX headers are set
+ if majorOk && minorOk {
+ sparseFormat = major + "." + minor
+ } else if sparseNameOk && sparseMapOk {
+ sparseFormat = "0.1"
+ } else if sparseSizeOk {
+ sparseFormat = "0.0"
+ } else {
+ // Not a PAX format GNU sparse file.
+ return nil, nil
+ }
+
+ // Check for unknown sparse format
+ if sparseFormat != "0.0" && sparseFormat != "0.1" && sparseFormat != "1.0" {
+ return nil, nil
+ }
+
+ // Update hdr from GNU sparse PAX headers
+ if sparseNameOk {
+ hdr.Name = sparseName
+ }
+ if sparseSizeOk {
+ realSize, err := strconv.ParseInt(sparseSize, 10, 0)
+ if err != nil {
+ return nil, ErrHeader
+ }
+ hdr.Size = realSize
+ } else if sparseRealSizeOk {
+ realSize, err := strconv.ParseInt(sparseRealSize, 10, 0)
+ if err != nil {
+ return nil, ErrHeader
+ }
+ hdr.Size = realSize
+ }
+
+ // Set up the sparse map, according to the particular sparse format in use
+ var sp []sparseEntry
+ var err error
+ switch sparseFormat {
+ case "0.0", "0.1":
+ sp, err = readGNUSparseMap0x1(headers)
+ case "1.0":
+ sp, err = readGNUSparseMap1x0(tr.curr)
+ }
+ return sp, err
+}
+
+// mergePAX merges well known headers according to PAX standard.
+// In general headers with the same name as those found
+// in the header struct overwrite those found in the header
+// struct with higher precision or longer values. Esp. useful
+// for name and linkname fields.
+func mergePAX(hdr *Header, headers map[string]string) error {
+ for k, v := range headers {
+ switch k {
+ case paxPath:
+ hdr.Name = v
+ case paxLinkpath:
+ hdr.Linkname = v
+ case paxGname:
+ hdr.Gname = v
+ case paxUname:
+ hdr.Uname = v
+ case paxUid:
+ uid, err := strconv.ParseInt(v, 10, 0)
+ if err != nil {
+ return err
+ }
+ hdr.Uid = int(uid)
+ case paxGid:
+ gid, err := strconv.ParseInt(v, 10, 0)
+ if err != nil {
+ return err
+ }
+ hdr.Gid = int(gid)
+ case paxAtime:
+ t, err := parsePAXTime(v)
+ if err != nil {
+ return err
+ }
+ hdr.AccessTime = t
+ case paxMtime:
+ t, err := parsePAXTime(v)
+ if err != nil {
+ return err
+ }
+ hdr.ModTime = t
+ case paxCtime:
+ t, err := parsePAXTime(v)
+ if err != nil {
+ return err
+ }
+ hdr.ChangeTime = t
+ case paxSize:
+ size, err := strconv.ParseInt(v, 10, 0)
+ if err != nil {
+ return err
+ }
+ hdr.Size = int64(size)
+ default:
+ if strings.HasPrefix(k, paxXattr) {
+ if hdr.Xattrs == nil {
+ hdr.Xattrs = make(map[string]string)
+ }
+ hdr.Xattrs[k[len(paxXattr):]] = v
+ }
+ }
+ }
+ return nil
+}
+
+// parsePAXTime takes a string of the form %d.%d as described in
+// the PAX specification.
+func parsePAXTime(t string) (time.Time, error) {
+ buf := []byte(t)
+ pos := bytes.IndexByte(buf, '.')
+ var seconds, nanoseconds int64
+ var err error
+ if pos == -1 {
+ seconds, err = strconv.ParseInt(t, 10, 0)
+ if err != nil {
+ return time.Time{}, err
+ }
+ } else {
+ seconds, err = strconv.ParseInt(string(buf[:pos]), 10, 0)
+ if err != nil {
+ return time.Time{}, err
+ }
+ nano_buf := string(buf[pos+1:])
+ // Pad as needed before converting to a decimal.
+ // For example .030 -> .030000000 -> 30000000 nanoseconds
+ if len(nano_buf) < maxNanoSecondIntSize {
+ // Right pad
+ nano_buf += strings.Repeat("0", maxNanoSecondIntSize-len(nano_buf))
+ } else if len(nano_buf) > maxNanoSecondIntSize {
+ // Right truncate
+ nano_buf = nano_buf[:maxNanoSecondIntSize]
+ }
+ nanoseconds, err = strconv.ParseInt(string(nano_buf), 10, 0)
+ if err != nil {
+ return time.Time{}, err
+ }
+ }
+ ts := time.Unix(seconds, nanoseconds)
+ return ts, nil
+}
+
+// parsePAX parses PAX headers.
+// If an extended header (type 'x') is invalid, ErrHeader is returned
+func parsePAX(r io.Reader) (map[string]string, error) {
+ buf, err := ioutil.ReadAll(r)
+ if err != nil {
+ return nil, err
+ }
+
+ // For GNU PAX sparse format 0.0 support.
+ // This function transforms the sparse format 0.0 headers into sparse format 0.1 headers.
+ var sparseMap bytes.Buffer
+
+ headers := make(map[string]string)
+ // Each record is constructed as
+ // "%d %s=%s\n", length, keyword, value
+ for len(buf) > 0 {
+ // or the header was empty to start with.
+ var sp int
+ // The size field ends at the first space.
+ sp = bytes.IndexByte(buf, ' ')
+ if sp == -1 {
+ return nil, ErrHeader
+ }
+ // Parse the first token as a decimal integer.
+ n, err := strconv.ParseInt(string(buf[:sp]), 10, 0)
+ if err != nil {
+ return nil, ErrHeader
+ }
+ // Extract everything between the decimal and the n -1 on the
+ // beginning to eat the ' ', -1 on the end to skip the newline.
+ var record []byte
+ record, buf = buf[sp+1:n-1], buf[n:]
+ // The first equals is guaranteed to mark the end of the key.
+ // Everything else is value.
+ eq := bytes.IndexByte(record, '=')
+ if eq == -1 {
+ return nil, ErrHeader
+ }
+ key, value := record[:eq], record[eq+1:]
+
+ keyStr := string(key)
+ if keyStr == paxGNUSparseOffset || keyStr == paxGNUSparseNumBytes {
+ // GNU sparse format 0.0 special key. Write to sparseMap instead of using the headers map.
+ sparseMap.Write(value)
+ sparseMap.Write([]byte{','})
+ } else {
+ // Normal key. Set the value in the headers map.
+ headers[keyStr] = string(value)
+ }
+ }
+ if sparseMap.Len() != 0 {
+ // Add sparse info to headers, chopping off the extra comma
+ sparseMap.Truncate(sparseMap.Len() - 1)
+ headers[paxGNUSparseMap] = sparseMap.String()
+ }
+ return headers, nil
+}
+
+// cString parses bytes as a NUL-terminated C-style string.
+// If a NUL byte is not found then the whole slice is returned as a string.
+func cString(b []byte) string {
+ n := 0
+ for n < len(b) && b[n] != 0 {
+ n++
+ }
+ return string(b[0:n])
+}
+
+func (tr *Reader) octal(b []byte) int64 {
+ // Check for binary format first.
+ if len(b) > 0 && b[0]&0x80 != 0 {
+ var x int64
+ for i, c := range b {
+ if i == 0 {
+ c &= 0x7f // ignore signal bit in first byte
+ }
+ x = x<<8 | int64(c)
+ }
+ return x
+ }
+
+ // Because unused fields are filled with NULs, we need
+ // to skip leading NULs. Fields may also be padded with
+ // spaces or NULs.
+ // So we remove leading and trailing NULs and spaces to
+ // be sure.
+ b = bytes.Trim(b, " \x00")
+
+ if len(b) == 0 {
+ return 0
+ }
+ x, err := strconv.ParseUint(cString(b), 8, 64)
+ if err != nil {
+ tr.err = err
+ }
+ return int64(x)
+}
+
+// skipUnread skips any unread bytes in the existing file entry, as well as any alignment padding.
+func (tr *Reader) skipUnread() {
+ nr := tr.numBytes() + tr.pad // number of bytes to skip
+ tr.curr, tr.pad = nil, 0
+ if sr, ok := tr.r.(io.Seeker); ok {
+ if _, err := sr.Seek(nr, os.SEEK_CUR); err == nil {
+ return
+ }
+ }
+ _, tr.err = io.CopyN(ioutil.Discard, tr.r, nr)
+}
+
+func (tr *Reader) verifyChecksum(header []byte) bool {
+ if tr.err != nil {
+ return false
+ }
+
+ given := tr.octal(header[148:156])
+ unsigned, signed := checksum(header)
+ return given == unsigned || given == signed
+}
+
+func (tr *Reader) readHeader() *Header {
+ header := tr.hdrBuff[:]
+ copy(header, zeroBlock)
+
+ if _, tr.err = io.ReadFull(tr.r, header); tr.err != nil {
+ return nil
+ }
+
+ // Two blocks of zero bytes marks the end of the archive.
+ if bytes.Equal(header, zeroBlock[0:blockSize]) {
+ if _, tr.err = io.ReadFull(tr.r, header); tr.err != nil {
+ return nil
+ }
+ if bytes.Equal(header, zeroBlock[0:blockSize]) {
+ tr.err = io.EOF
+ } else {
+ tr.err = ErrHeader // zero block and then non-zero block
+ }
+ return nil
+ }
+
+ if !tr.verifyChecksum(header) {
+ tr.err = ErrHeader
+ return nil
+ }
+
+ // Unpack
+ hdr := new(Header)
+ s := slicer(header)
+
+ hdr.Name = cString(s.next(100))
+ hdr.Mode = tr.octal(s.next(8))
+ hdr.Uid = int(tr.octal(s.next(8)))
+ hdr.Gid = int(tr.octal(s.next(8)))
+ hdr.Size = tr.octal(s.next(12))
+ hdr.ModTime = time.Unix(tr.octal(s.next(12)), 0)
+ s.next(8) // chksum
+ hdr.Typeflag = s.next(1)[0]
+ hdr.Linkname = cString(s.next(100))
+
+ // The remainder of the header depends on the value of magic.
+ // The original (v7) version of tar had no explicit magic field,
+ // so its magic bytes, like the rest of the block, are NULs.
+ magic := string(s.next(8)) // contains version field as well.
+ var format string
+ switch {
+ case magic[:6] == "ustar\x00": // POSIX tar (1003.1-1988)
+ if string(header[508:512]) == "tar\x00" {
+ format = "star"
+ } else {
+ format = "posix"
+ }
+ case magic == "ustar \x00": // old GNU tar
+ format = "gnu"
+ }
+
+ switch format {
+ case "posix", "gnu", "star":
+ hdr.Uname = cString(s.next(32))
+ hdr.Gname = cString(s.next(32))
+ devmajor := s.next(8)
+ devminor := s.next(8)
+ if hdr.Typeflag == TypeChar || hdr.Typeflag == TypeBlock {
+ hdr.Devmajor = tr.octal(devmajor)
+ hdr.Devminor = tr.octal(devminor)
+ }
+ var prefix string
+ switch format {
+ case "posix", "gnu":
+ prefix = cString(s.next(155))
+ case "star":
+ prefix = cString(s.next(131))
+ hdr.AccessTime = time.Unix(tr.octal(s.next(12)), 0)
+ hdr.ChangeTime = time.Unix(tr.octal(s.next(12)), 0)
+ }
+ if len(prefix) > 0 {
+ hdr.Name = prefix + "/" + hdr.Name
+ }
+ }
+
+ if tr.err != nil {
+ tr.err = ErrHeader
+ return nil
+ }
+
+ // Maximum value of hdr.Size is 64 GB (12 octal digits),
+ // so there's no risk of int64 overflowing.
+ nb := int64(hdr.Size)
+ tr.pad = -nb & (blockSize - 1) // blockSize is a power of two
+
+ // Set the current file reader.
+ tr.curr = &regFileReader{r: tr.r, nb: nb}
+
+ // Check for old GNU sparse format entry.
+ if hdr.Typeflag == TypeGNUSparse {
+ // Get the real size of the file.
+ hdr.Size = tr.octal(header[483:495])
+
+ // Read the sparse map.
+ sp := tr.readOldGNUSparseMap(header)
+ if tr.err != nil {
+ return nil
+ }
+ // Current file is a GNU sparse file. Update the current file reader.
+ tr.curr = &sparseFileReader{rfr: tr.curr.(*regFileReader), sp: sp, tot: hdr.Size}
+ }
+
+ return hdr
+}
+
+// A sparseEntry holds a single entry in a sparse file's sparse map.
+// A sparse entry indicates the offset and size in a sparse file of a
+// block of data.
+type sparseEntry struct {
+ offset int64
+ numBytes int64
+}
+
+// readOldGNUSparseMap reads the sparse map as stored in the old GNU sparse format.
+// The sparse map is stored in the tar header if it's small enough. If it's larger than four entries,
+// then one or more extension headers are used to store the rest of the sparse map.
+func (tr *Reader) readOldGNUSparseMap(header []byte) []sparseEntry {
+ isExtended := header[oldGNUSparseMainHeaderIsExtendedOffset] != 0
+ spCap := oldGNUSparseMainHeaderNumEntries
+ if isExtended {
+ spCap += oldGNUSparseExtendedHeaderNumEntries
+ }
+ sp := make([]sparseEntry, 0, spCap)
+ s := slicer(header[oldGNUSparseMainHeaderOffset:])
+
+ // Read the four entries from the main tar header
+ for i := 0; i < oldGNUSparseMainHeaderNumEntries; i++ {
+ offset := tr.octal(s.next(oldGNUSparseOffsetSize))
+ numBytes := tr.octal(s.next(oldGNUSparseNumBytesSize))
+ if tr.err != nil {
+ tr.err = ErrHeader
+ return nil
+ }
+ if offset == 0 && numBytes == 0 {
+ break
+ }
+ sp = append(sp, sparseEntry{offset: offset, numBytes: numBytes})
+ }
+
+ for isExtended {
+ // There are more entries. Read an extension header and parse its entries.
+ sparseHeader := make([]byte, blockSize)
+ if _, tr.err = io.ReadFull(tr.r, sparseHeader); tr.err != nil {
+ return nil
+ }
+ isExtended = sparseHeader[oldGNUSparseExtendedHeaderIsExtendedOffset] != 0
+ s = slicer(sparseHeader)
+ for i := 0; i < oldGNUSparseExtendedHeaderNumEntries; i++ {
+ offset := tr.octal(s.next(oldGNUSparseOffsetSize))
+ numBytes := tr.octal(s.next(oldGNUSparseNumBytesSize))
+ if tr.err != nil {
+ tr.err = ErrHeader
+ return nil
+ }
+ if offset == 0 && numBytes == 0 {
+ break
+ }
+ sp = append(sp, sparseEntry{offset: offset, numBytes: numBytes})
+ }
+ }
+ return sp
+}
+
+// readGNUSparseMap1x0 reads the sparse map as stored in GNU's PAX sparse format version 1.0.
+// The sparse map is stored just before the file data and padded out to the nearest block boundary.
+func readGNUSparseMap1x0(r io.Reader) ([]sparseEntry, error) {
+ buf := make([]byte, 2*blockSize)
+ sparseHeader := buf[:blockSize]
+
+ // readDecimal is a helper function to read a decimal integer from the sparse map
+ // while making sure to read from the file in blocks of size blockSize
+ readDecimal := func() (int64, error) {
+ // Look for newline
+ nl := bytes.IndexByte(sparseHeader, '\n')
+ if nl == -1 {
+ if len(sparseHeader) >= blockSize {
+ // This is an error
+ return 0, ErrHeader
+ }
+ oldLen := len(sparseHeader)
+ newLen := oldLen + blockSize
+ if cap(sparseHeader) < newLen {
+ // There's more header, but we need to make room for the next block
+ copy(buf, sparseHeader)
+ sparseHeader = buf[:newLen]
+ } else {
+ // There's more header, and we can just reslice
+ sparseHeader = sparseHeader[:newLen]
+ }
+
+ // Now that sparseHeader is large enough, read next block
+ if _, err := io.ReadFull(r, sparseHeader[oldLen:newLen]); err != nil {
+ return 0, err
+ }
+
+ // Look for a newline in the new data
+ nl = bytes.IndexByte(sparseHeader[oldLen:newLen], '\n')
+ if nl == -1 {
+ // This is an error
+ return 0, ErrHeader
+ }
+ nl += oldLen // We want the position from the beginning
+ }
+ // Now that we've found a newline, read a number
+ n, err := strconv.ParseInt(string(sparseHeader[:nl]), 10, 0)
+ if err != nil {
+ return 0, ErrHeader
+ }
+
+ // Update sparseHeader to consume this number
+ sparseHeader = sparseHeader[nl+1:]
+ return n, nil
+ }
+
+ // Read the first block
+ if _, err := io.ReadFull(r, sparseHeader); err != nil {
+ return nil, err
+ }
+
+ // The first line contains the number of entries
+ numEntries, err := readDecimal()
+ if err != nil {
+ return nil, err
+ }
+
+ // Read all the entries
+ sp := make([]sparseEntry, 0, numEntries)
+ for i := int64(0); i < numEntries; i++ {
+ // Read the offset
+ offset, err := readDecimal()
+ if err != nil {
+ return nil, err
+ }
+ // Read numBytes
+ numBytes, err := readDecimal()
+ if err != nil {
+ return nil, err
+ }
+
+ sp = append(sp, sparseEntry{offset: offset, numBytes: numBytes})
+ }
+
+ return sp, nil
+}
+
+// readGNUSparseMap0x1 reads the sparse map as stored in GNU's PAX sparse format version 0.1.
+// The sparse map is stored in the PAX headers.
+func readGNUSparseMap0x1(headers map[string]string) ([]sparseEntry, error) {
+ // Get number of entries
+ numEntriesStr, ok := headers[paxGNUSparseNumBlocks]
+ if !ok {
+ return nil, ErrHeader
+ }
+ numEntries, err := strconv.ParseInt(numEntriesStr, 10, 0)
+ if err != nil {
+ return nil, ErrHeader
+ }
+
+ sparseMap := strings.Split(headers[paxGNUSparseMap], ",")
+
+ // There should be two numbers in sparseMap for each entry
+ if int64(len(sparseMap)) != 2*numEntries {
+ return nil, ErrHeader
+ }
+
+ // Loop through the entries in the sparse map
+ sp := make([]sparseEntry, 0, numEntries)
+ for i := int64(0); i < numEntries; i++ {
+ offset, err := strconv.ParseInt(sparseMap[2*i], 10, 0)
+ if err != nil {
+ return nil, ErrHeader
+ }
+ numBytes, err := strconv.ParseInt(sparseMap[2*i+1], 10, 0)
+ if err != nil {
+ return nil, ErrHeader
+ }
+ sp = append(sp, sparseEntry{offset: offset, numBytes: numBytes})
+ }
+
+ return sp, nil
+}
+
+// numBytes returns the number of bytes left to read in the current file's entry
+// in the tar archive, or 0 if there is no current file.
+func (tr *Reader) numBytes() int64 {
+ if tr.curr == nil {
+ // No current file, so no bytes
+ return 0
+ }
+ return tr.curr.numBytes()
+}
+
+// Read reads from the current entry in the tar archive.
+// It returns 0, io.EOF when it reaches the end of that entry,
+// until Next is called to advance to the next entry.
+func (tr *Reader) Read(b []byte) (n int, err error) {
+ if tr.curr == nil {
+ return 0, io.EOF
+ }
+ n, err = tr.curr.Read(b)
+ if err != nil && err != io.EOF {
+ tr.err = err
+ }
+ return
+}
+
+func (rfr *regFileReader) Read(b []byte) (n int, err error) {
+ if rfr.nb == 0 {
+ // file consumed
+ return 0, io.EOF
+ }
+ if int64(len(b)) > rfr.nb {
+ b = b[0:rfr.nb]
+ }
+ n, err = rfr.r.Read(b)
+ rfr.nb -= int64(n)
+
+ if err == io.EOF && rfr.nb > 0 {
+ err = io.ErrUnexpectedEOF
+ }
+ return
+}
+
+// numBytes returns the number of bytes left to read in the file's data in the tar archive.
+func (rfr *regFileReader) numBytes() int64 {
+ return rfr.nb
+}
+
+// readHole reads a sparse file hole ending at offset toOffset
+func (sfr *sparseFileReader) readHole(b []byte, toOffset int64) int {
+ n64 := toOffset - sfr.pos
+ if n64 > int64(len(b)) {
+ n64 = int64(len(b))
+ }
+ n := int(n64)
+ for i := 0; i < n; i++ {
+ b[i] = 0
+ }
+ sfr.pos += n64
+ return n
+}
+
+// Read reads the sparse file data in expanded form.
+func (sfr *sparseFileReader) Read(b []byte) (n int, err error) {
+ if len(sfr.sp) == 0 {
+ // No more data fragments to read from.
+ if sfr.pos < sfr.tot {
+ // We're in the last hole
+ n = sfr.readHole(b, sfr.tot)
+ return
+ }
+ // Otherwise, we're at the end of the file
+ return 0, io.EOF
+ }
+ if sfr.pos < sfr.sp[0].offset {
+ // We're in a hole
+ n = sfr.readHole(b, sfr.sp[0].offset)
+ return
+ }
+
+ // We're not in a hole, so we'll read from the next data fragment
+ posInFragment := sfr.pos - sfr.sp[0].offset
+ bytesLeft := sfr.sp[0].numBytes - posInFragment
+ if int64(len(b)) > bytesLeft {
+ b = b[0:bytesLeft]
+ }
+
+ n, err = sfr.rfr.Read(b)
+ sfr.pos += int64(n)
+
+ if int64(n) == bytesLeft {
+ // We're done with this fragment
+ sfr.sp = sfr.sp[1:]
+ }
+
+ if err == io.EOF && sfr.pos < sfr.tot {
+ // We reached the end of the last fragment's data, but there's a final hole
+ err = nil
+ }
+ return
+}
+
+// numBytes returns the number of bytes left to read in the sparse file's
+// sparse-encoded data in the tar archive.
+func (sfr *sparseFileReader) numBytes() int64 {
+ return sfr.rfr.nb
+}
diff --git a/src/archive/tar/reader_test.go b/src/archive/tar/reader_test.go
new file mode 100644
index 000000000..9601ffe45
--- /dev/null
+++ b/src/archive/tar/reader_test.go
@@ -0,0 +1,743 @@
+// Copyright 2009 The Go Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+package tar
+
+import (
+ "bytes"
+ "crypto/md5"
+ "fmt"
+ "io"
+ "io/ioutil"
+ "os"
+ "reflect"
+ "strings"
+ "testing"
+ "time"
+)
+
+type untarTest struct {
+ file string
+ headers []*Header
+ cksums []string
+}
+
+var gnuTarTest = &untarTest{
+ file: "testdata/gnu.tar",
+ headers: []*Header{
+ {
+ Name: "small.txt",
+ Mode: 0640,
+ Uid: 73025,
+ Gid: 5000,
+ Size: 5,
+ ModTime: time.Unix(1244428340, 0),
+ Typeflag: '0',
+ Uname: "dsymonds",
+ Gname: "eng",
+ },
+ {
+ Name: "small2.txt",
+ Mode: 0640,
+ Uid: 73025,
+ Gid: 5000,
+ Size: 11,
+ ModTime: time.Unix(1244436044, 0),
+ Typeflag: '0',
+ Uname: "dsymonds",
+ Gname: "eng",
+ },
+ },
+ cksums: []string{
+ "e38b27eaccb4391bdec553a7f3ae6b2f",
+ "c65bd2e50a56a2138bf1716f2fd56fe9",
+ },
+}
+
+var sparseTarTest = &untarTest{
+ file: "testdata/sparse-formats.tar",
+ headers: []*Header{
+ {
+ Name: "sparse-gnu",
+ Mode: 420,
+ Uid: 1000,
+ Gid: 1000,
+ Size: 200,
+ ModTime: time.Unix(1392395740, 0),
+ Typeflag: 0x53,
+ Linkname: "",
+ Uname: "david",
+ Gname: "david",
+ Devmajor: 0,
+ Devminor: 0,
+ },
+ {
+ Name: "sparse-posix-0.0",
+ Mode: 420,
+ Uid: 1000,
+ Gid: 1000,
+ Size: 200,
+ ModTime: time.Unix(1392342187, 0),
+ Typeflag: 0x30,
+ Linkname: "",
+ Uname: "david",
+ Gname: "david",
+ Devmajor: 0,
+ Devminor: 0,
+ },
+ {
+ Name: "sparse-posix-0.1",
+ Mode: 420,
+ Uid: 1000,
+ Gid: 1000,
+ Size: 200,
+ ModTime: time.Unix(1392340456, 0),
+ Typeflag: 0x30,
+ Linkname: "",
+ Uname: "david",
+ Gname: "david",
+ Devmajor: 0,
+ Devminor: 0,
+ },
+ {
+ Name: "sparse-posix-1.0",
+ Mode: 420,
+ Uid: 1000,
+ Gid: 1000,
+ Size: 200,
+ ModTime: time.Unix(1392337404, 0),
+ Typeflag: 0x30,
+ Linkname: "",
+ Uname: "david",
+ Gname: "david",
+ Devmajor: 0,
+ Devminor: 0,
+ },
+ {
+ Name: "end",
+ Mode: 420,
+ Uid: 1000,
+ Gid: 1000,
+ Size: 4,
+ ModTime: time.Unix(1392398319, 0),
+ Typeflag: 0x30,
+ Linkname: "",
+ Uname: "david",
+ Gname: "david",
+ Devmajor: 0,
+ Devminor: 0,
+ },
+ },
+ cksums: []string{
+ "6f53234398c2449fe67c1812d993012f",
+ "6f53234398c2449fe67c1812d993012f",
+ "6f53234398c2449fe67c1812d993012f",
+ "6f53234398c2449fe67c1812d993012f",
+ "b0061974914468de549a2af8ced10316",
+ },
+}
+
+var untarTests = []*untarTest{
+ gnuTarTest,
+ sparseTarTest,
+ {
+ file: "testdata/star.tar",
+ headers: []*Header{
+ {
+ Name: "small.txt",
+ Mode: 0640,
+ Uid: 73025,
+ Gid: 5000,
+ Size: 5,
+ ModTime: time.Unix(1244592783, 0),
+ Typeflag: '0',
+ Uname: "dsymonds",
+ Gname: "eng",
+ AccessTime: time.Unix(1244592783, 0),
+ ChangeTime: time.Unix(1244592783, 0),
+ },
+ {
+ Name: "small2.txt",
+ Mode: 0640,
+ Uid: 73025,
+ Gid: 5000,
+ Size: 11,
+ ModTime: time.Unix(1244592783, 0),
+ Typeflag: '0',
+ Uname: "dsymonds",
+ Gname: "eng",
+ AccessTime: time.Unix(1244592783, 0),
+ ChangeTime: time.Unix(1244592783, 0),
+ },
+ },
+ },
+ {
+ file: "testdata/v7.tar",
+ headers: []*Header{
+ {
+ Name: "small.txt",
+ Mode: 0444,
+ Uid: 73025,
+ Gid: 5000,
+ Size: 5,
+ ModTime: time.Unix(1244593104, 0),
+ Typeflag: '\x00',
+ },
+ {
+ Name: "small2.txt",
+ Mode: 0444,
+ Uid: 73025,
+ Gid: 5000,
+ Size: 11,
+ ModTime: time.Unix(1244593104, 0),
+ Typeflag: '\x00',
+ },
+ },
+ },
+ {
+ file: "testdata/pax.tar",
+ headers: []*Header{
+ {
+ Name: "a/123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100",
+ Mode: 0664,
+ Uid: 1000,
+ Gid: 1000,
+ Uname: "shane",
+ Gname: "shane",
+ Size: 7,
+ ModTime: time.Unix(1350244992, 23960108),
+ ChangeTime: time.Unix(1350244992, 23960108),
+ AccessTime: time.Unix(1350244992, 23960108),
+ Typeflag: TypeReg,
+ },
+ {
+ Name: "a/b",
+ Mode: 0777,
+ Uid: 1000,
+ Gid: 1000,
+ Uname: "shane",
+ Gname: "shane",
+ Size: 0,
+ ModTime: time.Unix(1350266320, 910238425),
+ ChangeTime: time.Unix(1350266320, 910238425),
+ AccessTime: time.Unix(1350266320, 910238425),
+ Typeflag: TypeSymlink,
+ Linkname: "123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100",
+ },
+ },
+ },
+ {
+ file: "testdata/nil-uid.tar", // golang.org/issue/5290
+ headers: []*Header{
+ {
+ Name: "P1050238.JPG.log",
+ Mode: 0664,
+ Uid: 0,
+ Gid: 0,
+ Size: 14,
+ ModTime: time.Unix(1365454838, 0),
+ Typeflag: TypeReg,
+ Linkname: "",
+ Uname: "eyefi",
+ Gname: "eyefi",
+ Devmajor: 0,
+ Devminor: 0,
+ },
+ },
+ },
+ {
+ file: "testdata/xattrs.tar",
+ headers: []*Header{
+ {
+ Name: "small.txt",
+ Mode: 0644,
+ Uid: 1000,
+ Gid: 10,
+ Size: 5,
+ ModTime: time.Unix(1386065770, 448252320),
+ Typeflag: '0',
+ Uname: "alex",
+ Gname: "wheel",
+ AccessTime: time.Unix(1389782991, 419875220),
+ ChangeTime: time.Unix(1389782956, 794414986),
+ Xattrs: map[string]string{
+ "user.key": "value",
+ "user.key2": "value2",
+ // Interestingly, selinux encodes the terminating null inside the xattr
+ "security.selinux": "unconfined_u:object_r:default_t:s0\x00",
+ },
+ },
+ {
+ Name: "small2.txt",
+ Mode: 0644,
+ Uid: 1000,
+ Gid: 10,
+ Size: 11,
+ ModTime: time.Unix(1386065770, 449252304),
+ Typeflag: '0',
+ Uname: "alex",
+ Gname: "wheel",
+ AccessTime: time.Unix(1389782991, 419875220),
+ ChangeTime: time.Unix(1386065770, 449252304),
+ Xattrs: map[string]string{
+ "security.selinux": "unconfined_u:object_r:default_t:s0\x00",
+ },
+ },
+ },
+ },
+}
+
+func TestReader(t *testing.T) {
+testLoop:
+ for i, test := range untarTests {
+ f, err := os.Open(test.file)
+ if err != nil {
+ t.Errorf("test %d: Unexpected error: %v", i, err)
+ continue
+ }
+ defer f.Close()
+ tr := NewReader(f)
+ for j, header := range test.headers {
+ hdr, err := tr.Next()
+ if err != nil || hdr == nil {
+ t.Errorf("test %d, entry %d: Didn't get entry: %v", i, j, err)
+ f.Close()
+ continue testLoop
+ }
+ if !reflect.DeepEqual(*hdr, *header) {
+ t.Errorf("test %d, entry %d: Incorrect header:\nhave %+v\nwant %+v",
+ i, j, *hdr, *header)
+ }
+ }
+ hdr, err := tr.Next()
+ if err == io.EOF {
+ continue testLoop
+ }
+ if hdr != nil || err != nil {
+ t.Errorf("test %d: Unexpected entry or error: hdr=%v err=%v", i, hdr, err)
+ }
+ }
+}
+
+func TestPartialRead(t *testing.T) {
+ f, err := os.Open("testdata/gnu.tar")
+ if err != nil {
+ t.Fatalf("Unexpected error: %v", err)
+ }
+ defer f.Close()
+
+ tr := NewReader(f)
+
+ // Read the first four bytes; Next() should skip the last byte.
+ hdr, err := tr.Next()
+ if err != nil || hdr == nil {
+ t.Fatalf("Didn't get first file: %v", err)
+ }
+ buf := make([]byte, 4)
+ if _, err := io.ReadFull(tr, buf); err != nil {
+ t.Fatalf("Unexpected error: %v", err)
+ }
+ if expected := []byte("Kilt"); !bytes.Equal(buf, expected) {
+ t.Errorf("Contents = %v, want %v", buf, expected)
+ }
+
+ // Second file
+ hdr, err = tr.Next()
+ if err != nil || hdr == nil {
+ t.Fatalf("Didn't get second file: %v", err)
+ }
+ buf = make([]byte, 6)
+ if _, err := io.ReadFull(tr, buf); err != nil {
+ t.Fatalf("Unexpected error: %v", err)
+ }
+ if expected := []byte("Google"); !bytes.Equal(buf, expected) {
+ t.Errorf("Contents = %v, want %v", buf, expected)
+ }
+}
+
+func TestIncrementalRead(t *testing.T) {
+ test := gnuTarTest
+ f, err := os.Open(test.file)
+ if err != nil {
+ t.Fatalf("Unexpected error: %v", err)
+ }
+ defer f.Close()
+
+ tr := NewReader(f)
+
+ headers := test.headers
+ cksums := test.cksums
+ nread := 0
+
+ // loop over all files
+ for ; ; nread++ {
+ hdr, err := tr.Next()
+ if hdr == nil || err == io.EOF {
+ break
+ }
+
+ // check the header
+ if !reflect.DeepEqual(*hdr, *headers[nread]) {
+ t.Errorf("Incorrect header:\nhave %+v\nwant %+v",
+ *hdr, headers[nread])
+ }
+
+ // read file contents in little chunks EOF,
+ // checksumming all the way
+ h := md5.New()
+ rdbuf := make([]uint8, 8)
+ for {
+ nr, err := tr.Read(rdbuf)
+ if err == io.EOF {
+ break
+ }
+ if err != nil {
+ t.Errorf("Read: unexpected error %v\n", err)
+ break
+ }
+ h.Write(rdbuf[0:nr])
+ }
+ // verify checksum
+ have := fmt.Sprintf("%x", h.Sum(nil))
+ want := cksums[nread]
+ if want != have {
+ t.Errorf("Bad checksum on file %s:\nhave %+v\nwant %+v", hdr.Name, have, want)
+ }
+ }
+ if nread != len(headers) {
+ t.Errorf("Didn't process all files\nexpected: %d\nprocessed %d\n", len(headers), nread)
+ }
+}
+
+func TestNonSeekable(t *testing.T) {
+ test := gnuTarTest
+ f, err := os.Open(test.file)
+ if err != nil {
+ t.Fatalf("Unexpected error: %v", err)
+ }
+ defer f.Close()
+
+ type readerOnly struct {
+ io.Reader
+ }
+ tr := NewReader(readerOnly{f})
+ nread := 0
+
+ for ; ; nread++ {
+ _, err := tr.Next()
+ if err == io.EOF {
+ break
+ }
+ if err != nil {
+ t.Fatalf("Unexpected error: %v", err)
+ }
+ }
+
+ if nread != len(test.headers) {
+ t.Errorf("Didn't process all files\nexpected: %d\nprocessed %d\n", len(test.headers), nread)
+ }
+}
+
+func TestParsePAXHeader(t *testing.T) {
+ paxTests := [][3]string{
+ {"a", "a=name", "10 a=name\n"}, // Test case involving multiple acceptable lengths
+ {"a", "a=name", "9 a=name\n"}, // Test case involving multiple acceptable length
+ {"mtime", "mtime=1350244992.023960108", "30 mtime=1350244992.023960108\n"}}
+ for _, test := range paxTests {
+ key, expected, raw := test[0], test[1], test[2]
+ reader := bytes.NewReader([]byte(raw))
+ headers, err := parsePAX(reader)
+ if err != nil {
+ t.Errorf("Couldn't parse correctly formatted headers: %v", err)
+ continue
+ }
+ if strings.EqualFold(headers[key], expected) {
+ t.Errorf("mtime header incorrectly parsed: got %s, wanted %s", headers[key], expected)
+ continue
+ }
+ trailer := make([]byte, 100)
+ n, err := reader.Read(trailer)
+ if err != io.EOF || n != 0 {
+ t.Error("Buffer wasn't consumed")
+ }
+ }
+ badHeader := bytes.NewReader([]byte("3 somelongkey="))
+ if _, err := parsePAX(badHeader); err != ErrHeader {
+ t.Fatal("Unexpected success when parsing bad header")
+ }
+}
+
+func TestParsePAXTime(t *testing.T) {
+ // Some valid PAX time values
+ timestamps := map[string]time.Time{
+ "1350244992.023960108": time.Unix(1350244992, 23960108), // The common case
+ "1350244992.02396010": time.Unix(1350244992, 23960100), // Lower precision value
+ "1350244992.0239601089": time.Unix(1350244992, 23960108), // Higher precision value
+ "1350244992": time.Unix(1350244992, 0), // Low precision value
+ }
+ for input, expected := range timestamps {
+ ts, err := parsePAXTime(input)
+ if err != nil {
+ t.Fatal(err)
+ }
+ if !ts.Equal(expected) {
+ t.Fatalf("Time parsing failure %s %s", ts, expected)
+ }
+ }
+}
+
+func TestMergePAX(t *testing.T) {
+ hdr := new(Header)
+ // Test a string, integer, and time based value.
+ headers := map[string]string{
+ "path": "a/b/c",
+ "uid": "1000",
+ "mtime": "1350244992.023960108",
+ }
+ err := mergePAX(hdr, headers)
+ if err != nil {
+ t.Fatal(err)
+ }
+ want := &Header{
+ Name: "a/b/c",
+ Uid: 1000,
+ ModTime: time.Unix(1350244992, 23960108),
+ }
+ if !reflect.DeepEqual(hdr, want) {
+ t.Errorf("incorrect merge: got %+v, want %+v", hdr, want)
+ }
+}
+
+func TestSparseEndToEnd(t *testing.T) {
+ test := sparseTarTest
+ f, err := os.Open(test.file)
+ if err != nil {
+ t.Fatalf("Unexpected error: %v", err)
+ }
+ defer f.Close()
+
+ tr := NewReader(f)
+
+ headers := test.headers
+ cksums := test.cksums
+ nread := 0
+
+ // loop over all files
+ for ; ; nread++ {
+ hdr, err := tr.Next()
+ if hdr == nil || err == io.EOF {
+ break
+ }
+
+ // check the header
+ if !reflect.DeepEqual(*hdr, *headers[nread]) {
+ t.Errorf("Incorrect header:\nhave %+v\nwant %+v",
+ *hdr, headers[nread])
+ }
+
+ // read and checksum the file data
+ h := md5.New()
+ _, err = io.Copy(h, tr)
+ if err != nil {
+ t.Fatalf("Unexpected error: %v", err)
+ }
+
+ // verify checksum
+ have := fmt.Sprintf("%x", h.Sum(nil))
+ want := cksums[nread]
+ if want != have {
+ t.Errorf("Bad checksum on file %s:\nhave %+v\nwant %+v", hdr.Name, have, want)
+ }
+ }
+ if nread != len(headers) {
+ t.Errorf("Didn't process all files\nexpected: %d\nprocessed %d\n", len(headers), nread)
+ }
+}
+
+type sparseFileReadTest struct {
+ sparseData []byte
+ sparseMap []sparseEntry
+ realSize int64
+ expected []byte
+}
+
+var sparseFileReadTests = []sparseFileReadTest{
+ {
+ sparseData: []byte("abcde"),
+ sparseMap: []sparseEntry{
+ {offset: 0, numBytes: 2},
+ {offset: 5, numBytes: 3},
+ },
+ realSize: 8,
+ expected: []byte("ab\x00\x00\x00cde"),
+ },
+ {
+ sparseData: []byte("abcde"),
+ sparseMap: []sparseEntry{
+ {offset: 0, numBytes: 2},
+ {offset: 5, numBytes: 3},
+ },
+ realSize: 10,
+ expected: []byte("ab\x00\x00\x00cde\x00\x00"),
+ },
+ {
+ sparseData: []byte("abcde"),
+ sparseMap: []sparseEntry{
+ {offset: 1, numBytes: 3},
+ {offset: 6, numBytes: 2},
+ },
+ realSize: 8,
+ expected: []byte("\x00abc\x00\x00de"),
+ },
+ {
+ sparseData: []byte("abcde"),
+ sparseMap: []sparseEntry{
+ {offset: 1, numBytes: 3},
+ {offset: 6, numBytes: 2},
+ },
+ realSize: 10,
+ expected: []byte("\x00abc\x00\x00de\x00\x00"),
+ },
+ {
+ sparseData: []byte(""),
+ sparseMap: nil,
+ realSize: 2,
+ expected: []byte("\x00\x00"),
+ },
+}
+
+func TestSparseFileReader(t *testing.T) {
+ for i, test := range sparseFileReadTests {
+ r := bytes.NewReader(test.sparseData)
+ nb := int64(r.Len())
+ sfr := &sparseFileReader{
+ rfr: &regFileReader{r: r, nb: nb},
+ sp: test.sparseMap,
+ pos: 0,
+ tot: test.realSize,
+ }
+ if sfr.numBytes() != nb {
+ t.Errorf("test %d: Before reading, sfr.numBytes() = %d, want %d", i, sfr.numBytes(), nb)
+ }
+ buf, err := ioutil.ReadAll(sfr)
+ if err != nil {
+ t.Errorf("test %d: Unexpected error: %v", i, err)
+ }
+ if e := test.expected; !bytes.Equal(buf, e) {
+ t.Errorf("test %d: Contents = %v, want %v", i, buf, e)
+ }
+ if sfr.numBytes() != 0 {
+ t.Errorf("test %d: After draining the reader, numBytes() was nonzero", i)
+ }
+ }
+}
+
+func TestSparseIncrementalRead(t *testing.T) {
+ sparseMap := []sparseEntry{{10, 2}}
+ sparseData := []byte("Go")
+ expected := "\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00Go\x00\x00\x00\x00\x00\x00\x00\x00"
+
+ r := bytes.NewReader(sparseData)
+ nb := int64(r.Len())
+ sfr := &sparseFileReader{
+ rfr: &regFileReader{r: r, nb: nb},
+ sp: sparseMap,
+ pos: 0,
+ tot: int64(len(expected)),
+ }
+
+ // We'll read the data 6 bytes at a time, with a hole of size 10 at
+ // the beginning and one of size 8 at the end.
+ var outputBuf bytes.Buffer
+ buf := make([]byte, 6)
+ for {
+ n, err := sfr.Read(buf)
+ if err == io.EOF {
+ break
+ }
+ if err != nil {
+ t.Errorf("Read: unexpected error %v\n", err)
+ }
+ if n > 0 {
+ _, err := outputBuf.Write(buf[:n])
+ if err != nil {
+ t.Errorf("Write: unexpected error %v\n", err)
+ }
+ }
+ }
+ got := outputBuf.String()
+ if got != expected {
+ t.Errorf("Contents = %v, want %v", got, expected)
+ }
+}
+
+func TestReadGNUSparseMap0x1(t *testing.T) {
+ headers := map[string]string{
+ paxGNUSparseNumBlocks: "4",
+ paxGNUSparseMap: "0,5,10,5,20,5,30,5",
+ }
+ expected := []sparseEntry{
+ {offset: 0, numBytes: 5},
+ {offset: 10, numBytes: 5},
+ {offset: 20, numBytes: 5},
+ {offset: 30, numBytes: 5},
+ }
+
+ sp, err := readGNUSparseMap0x1(headers)
+ if err != nil {
+ t.Errorf("Unexpected error: %v", err)
+ }
+ if !reflect.DeepEqual(sp, expected) {
+ t.Errorf("Incorrect sparse map: got %v, wanted %v", sp, expected)
+ }
+}
+
+func TestReadGNUSparseMap1x0(t *testing.T) {
+ // This test uses lots of holes so the sparse header takes up more than two blocks
+ numEntries := 100
+ expected := make([]sparseEntry, 0, numEntries)
+ sparseMap := new(bytes.Buffer)
+
+ fmt.Fprintf(sparseMap, "%d\n", numEntries)
+ for i := 0; i < numEntries; i++ {
+ offset := int64(2048 * i)
+ numBytes := int64(1024)
+ expected = append(expected, sparseEntry{offset: offset, numBytes: numBytes})
+ fmt.Fprintf(sparseMap, "%d\n%d\n", offset, numBytes)
+ }
+
+ // Make the header the smallest multiple of blockSize that fits the sparseMap
+ headerBlocks := (sparseMap.Len() + blockSize - 1) / blockSize
+ bufLen := blockSize * headerBlocks
+ buf := make([]byte, bufLen)
+ copy(buf, sparseMap.Bytes())
+
+ // Get an reader to read the sparse map
+ r := bytes.NewReader(buf)
+
+ // Read the sparse map
+ sp, err := readGNUSparseMap1x0(r)
+ if err != nil {
+ t.Errorf("Unexpected error: %v", err)
+ }
+ if !reflect.DeepEqual(sp, expected) {
+ t.Errorf("Incorrect sparse map: got %v, wanted %v", sp, expected)
+ }
+}
+
+func TestUninitializedRead(t *testing.T) {
+ test := gnuTarTest
+ f, err := os.Open(test.file)
+ if err != nil {
+ t.Fatalf("Unexpected error: %v", err)
+ }
+ defer f.Close()
+
+ tr := NewReader(f)
+ _, err = tr.Read([]byte{})
+ if err == nil || err != io.EOF {
+ t.Errorf("Unexpected error: %v, wanted %v", err, io.EOF)
+ }
+
+}
diff --git a/src/archive/tar/stat_atim.go b/src/archive/tar/stat_atim.go
new file mode 100644
index 000000000..cf9cc79c5
--- /dev/null
+++ b/src/archive/tar/stat_atim.go
@@ -0,0 +1,20 @@
+// Copyright 2012 The Go Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+// +build linux dragonfly openbsd solaris
+
+package tar
+
+import (
+ "syscall"
+ "time"
+)
+
+func statAtime(st *syscall.Stat_t) time.Time {
+ return time.Unix(st.Atim.Unix())
+}
+
+func statCtime(st *syscall.Stat_t) time.Time {
+ return time.Unix(st.Ctim.Unix())
+}
diff --git a/src/archive/tar/stat_atimespec.go b/src/archive/tar/stat_atimespec.go
new file mode 100644
index 000000000..6f17dbe30
--- /dev/null
+++ b/src/archive/tar/stat_atimespec.go
@@ -0,0 +1,20 @@
+// Copyright 2012 The Go Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+// +build darwin freebsd netbsd
+
+package tar
+
+import (
+ "syscall"
+ "time"
+)
+
+func statAtime(st *syscall.Stat_t) time.Time {
+ return time.Unix(st.Atimespec.Unix())
+}
+
+func statCtime(st *syscall.Stat_t) time.Time {
+ return time.Unix(st.Ctimespec.Unix())
+}
diff --git a/src/archive/tar/stat_unix.go b/src/archive/tar/stat_unix.go
new file mode 100644
index 000000000..cb843db4c
--- /dev/null
+++ b/src/archive/tar/stat_unix.go
@@ -0,0 +1,32 @@
+// Copyright 2012 The Go Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+// +build linux darwin dragonfly freebsd openbsd netbsd solaris
+
+package tar
+
+import (
+ "os"
+ "syscall"
+)
+
+func init() {
+ sysStat = statUnix
+}
+
+func statUnix(fi os.FileInfo, h *Header) error {
+ sys, ok := fi.Sys().(*syscall.Stat_t)
+ if !ok {
+ return nil
+ }
+ h.Uid = int(sys.Uid)
+ h.Gid = int(sys.Gid)
+ // TODO(bradfitz): populate username & group. os/user
+ // doesn't cache LookupId lookups, and lacks group
+ // lookup functions.
+ h.AccessTime = statAtime(sys)
+ h.ChangeTime = statCtime(sys)
+ // TODO(bradfitz): major/minor device numbers?
+ return nil
+}
diff --git a/src/archive/tar/tar_test.go b/src/archive/tar/tar_test.go
new file mode 100644
index 000000000..ed333f3ea
--- /dev/null
+++ b/src/archive/tar/tar_test.go
@@ -0,0 +1,284 @@
+// Copyright 2012 The Go Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+package tar
+
+import (
+ "bytes"
+ "io/ioutil"
+ "os"
+ "path"
+ "reflect"
+ "strings"
+ "testing"
+ "time"
+)
+
+func TestFileInfoHeader(t *testing.T) {
+ fi, err := os.Stat("testdata/small.txt")
+ if err != nil {
+ t.Fatal(err)
+ }
+ h, err := FileInfoHeader(fi, "")
+ if err != nil {
+ t.Fatalf("FileInfoHeader: %v", err)
+ }
+ if g, e := h.Name, "small.txt"; g != e {
+ t.Errorf("Name = %q; want %q", g, e)
+ }
+ if g, e := h.Mode, int64(fi.Mode().Perm())|c_ISREG; g != e {
+ t.Errorf("Mode = %#o; want %#o", g, e)
+ }
+ if g, e := h.Size, int64(5); g != e {
+ t.Errorf("Size = %v; want %v", g, e)
+ }
+ if g, e := h.ModTime, fi.ModTime(); !g.Equal(e) {
+ t.Errorf("ModTime = %v; want %v", g, e)
+ }
+ // FileInfoHeader should error when passing nil FileInfo
+ if _, err := FileInfoHeader(nil, ""); err == nil {
+ t.Fatalf("Expected error when passing nil to FileInfoHeader")
+ }
+}
+
+func TestFileInfoHeaderDir(t *testing.T) {
+ fi, err := os.Stat("testdata")
+ if err != nil {
+ t.Fatal(err)
+ }
+ h, err := FileInfoHeader(fi, "")
+ if err != nil {
+ t.Fatalf("FileInfoHeader: %v", err)
+ }
+ if g, e := h.Name, "testdata/"; g != e {
+ t.Errorf("Name = %q; want %q", g, e)
+ }
+ // Ignoring c_ISGID for golang.org/issue/4867
+ if g, e := h.Mode&^c_ISGID, int64(fi.Mode().Perm())|c_ISDIR; g != e {
+ t.Errorf("Mode = %#o; want %#o", g, e)
+ }
+ if g, e := h.Size, int64(0); g != e {
+ t.Errorf("Size = %v; want %v", g, e)
+ }
+ if g, e := h.ModTime, fi.ModTime(); !g.Equal(e) {
+ t.Errorf("ModTime = %v; want %v", g, e)
+ }
+}
+
+func TestFileInfoHeaderSymlink(t *testing.T) {
+ h, err := FileInfoHeader(symlink{}, "some-target")
+ if err != nil {
+ t.Fatal(err)
+ }
+ if g, e := h.Name, "some-symlink"; g != e {
+ t.Errorf("Name = %q; want %q", g, e)
+ }
+ if g, e := h.Linkname, "some-target"; g != e {
+ t.Errorf("Linkname = %q; want %q", g, e)
+ }
+}
+
+type symlink struct{}
+
+func (symlink) Name() string { return "some-symlink" }
+func (symlink) Size() int64 { return 0 }
+func (symlink) Mode() os.FileMode { return os.ModeSymlink }
+func (symlink) ModTime() time.Time { return time.Time{} }
+func (symlink) IsDir() bool { return false }
+func (symlink) Sys() interface{} { return nil }
+
+func TestRoundTrip(t *testing.T) {
+ data := []byte("some file contents")
+
+ var b bytes.Buffer
+ tw := NewWriter(&b)
+ hdr := &Header{
+ Name: "file.txt",
+ Uid: 1 << 21, // too big for 8 octal digits
+ Size: int64(len(data)),
+ ModTime: time.Now(),
+ }
+ // tar only supports second precision.
+ hdr.ModTime = hdr.ModTime.Add(-time.Duration(hdr.ModTime.Nanosecond()) * time.Nanosecond)
+ if err := tw.WriteHeader(hdr); err != nil {
+ t.Fatalf("tw.WriteHeader: %v", err)
+ }
+ if _, err := tw.Write(data); err != nil {
+ t.Fatalf("tw.Write: %v", err)
+ }
+ if err := tw.Close(); err != nil {
+ t.Fatalf("tw.Close: %v", err)
+ }
+
+ // Read it back.
+ tr := NewReader(&b)
+ rHdr, err := tr.Next()
+ if err != nil {
+ t.Fatalf("tr.Next: %v", err)
+ }
+ if !reflect.DeepEqual(rHdr, hdr) {
+ t.Errorf("Header mismatch.\n got %+v\nwant %+v", rHdr, hdr)
+ }
+ rData, err := ioutil.ReadAll(tr)
+ if err != nil {
+ t.Fatalf("Read: %v", err)
+ }
+ if !bytes.Equal(rData, data) {
+ t.Errorf("Data mismatch.\n got %q\nwant %q", rData, data)
+ }
+}
+
+type headerRoundTripTest struct {
+ h *Header
+ fm os.FileMode
+}
+
+func TestHeaderRoundTrip(t *testing.T) {
+ golden := []headerRoundTripTest{
+ // regular file.
+ {
+ h: &Header{
+ Name: "test.txt",
+ Mode: 0644 | c_ISREG,
+ Size: 12,
+ ModTime: time.Unix(1360600916, 0),
+ Typeflag: TypeReg,
+ },
+ fm: 0644,
+ },
+ // hard link.
+ {
+ h: &Header{
+ Name: "hard.txt",
+ Mode: 0644 | c_ISLNK,
+ Size: 0,
+ ModTime: time.Unix(1360600916, 0),
+ Typeflag: TypeLink,
+ },
+ fm: 0644 | os.ModeSymlink,
+ },
+ // symbolic link.
+ {
+ h: &Header{
+ Name: "link.txt",
+ Mode: 0777 | c_ISLNK,
+ Size: 0,
+ ModTime: time.Unix(1360600852, 0),
+ Typeflag: TypeSymlink,
+ },
+ fm: 0777 | os.ModeSymlink,
+ },
+ // character device node.
+ {
+ h: &Header{
+ Name: "dev/null",
+ Mode: 0666 | c_ISCHR,
+ Size: 0,
+ ModTime: time.Unix(1360578951, 0),
+ Typeflag: TypeChar,
+ },
+ fm: 0666 | os.ModeDevice | os.ModeCharDevice,
+ },
+ // block device node.
+ {
+ h: &Header{
+ Name: "dev/sda",
+ Mode: 0660 | c_ISBLK,
+ Size: 0,
+ ModTime: time.Unix(1360578954, 0),
+ Typeflag: TypeBlock,
+ },
+ fm: 0660 | os.ModeDevice,
+ },
+ // directory.
+ {
+ h: &Header{
+ Name: "dir/",
+ Mode: 0755 | c_ISDIR,
+ Size: 0,
+ ModTime: time.Unix(1360601116, 0),
+ Typeflag: TypeDir,
+ },
+ fm: 0755 | os.ModeDir,
+ },
+ // fifo node.
+ {
+ h: &Header{
+ Name: "dev/initctl",
+ Mode: 0600 | c_ISFIFO,
+ Size: 0,
+ ModTime: time.Unix(1360578949, 0),
+ Typeflag: TypeFifo,
+ },
+ fm: 0600 | os.ModeNamedPipe,
+ },
+ // setuid.
+ {
+ h: &Header{
+ Name: "bin/su",
+ Mode: 0755 | c_ISREG | c_ISUID,
+ Size: 23232,
+ ModTime: time.Unix(1355405093, 0),
+ Typeflag: TypeReg,
+ },
+ fm: 0755 | os.ModeSetuid,
+ },
+ // setguid.
+ {
+ h: &Header{
+ Name: "group.txt",
+ Mode: 0750 | c_ISREG | c_ISGID,
+ Size: 0,
+ ModTime: time.Unix(1360602346, 0),
+ Typeflag: TypeReg,
+ },
+ fm: 0750 | os.ModeSetgid,
+ },
+ // sticky.
+ {
+ h: &Header{
+ Name: "sticky.txt",
+ Mode: 0600 | c_ISREG | c_ISVTX,
+ Size: 7,
+ ModTime: time.Unix(1360602540, 0),
+ Typeflag: TypeReg,
+ },
+ fm: 0600 | os.ModeSticky,
+ },
+ }
+
+ for i, g := range golden {
+ fi := g.h.FileInfo()
+ h2, err := FileInfoHeader(fi, "")
+ if err != nil {
+ t.Error(err)
+ continue
+ }
+ if strings.Contains(fi.Name(), "/") {
+ t.Errorf("FileInfo of %q contains slash: %q", g.h.Name, fi.Name())
+ }
+ name := path.Base(g.h.Name)
+ if fi.IsDir() {
+ name += "/"
+ }
+ if got, want := h2.Name, name; got != want {
+ t.Errorf("i=%d: Name: got %v, want %v", i, got, want)
+ }
+ if got, want := h2.Size, g.h.Size; got != want {
+ t.Errorf("i=%d: Size: got %v, want %v", i, got, want)
+ }
+ if got, want := h2.Mode, g.h.Mode; got != want {
+ t.Errorf("i=%d: Mode: got %o, want %o", i, got, want)
+ }
+ if got, want := fi.Mode(), g.fm; got != want {
+ t.Errorf("i=%d: fi.Mode: got %o, want %o", i, got, want)
+ }
+ if got, want := h2.ModTime, g.h.ModTime; got != want {
+ t.Errorf("i=%d: ModTime: got %v, want %v", i, got, want)
+ }
+ if sysh, ok := fi.Sys().(*Header); !ok || sysh != g.h {
+ t.Errorf("i=%d: Sys didn't return original *Header", i)
+ }
+ }
+}
diff --git a/src/archive/tar/testdata/gnu.tar b/src/archive/tar/testdata/gnu.tar
new file mode 100644
index 000000000..fc899dc8d
--- /dev/null
+++ b/src/archive/tar/testdata/gnu.tar
Binary files differ
diff --git a/src/archive/tar/testdata/nil-uid.tar b/src/archive/tar/testdata/nil-uid.tar
new file mode 100644
index 000000000..cc9cfaa33
--- /dev/null
+++ b/src/archive/tar/testdata/nil-uid.tar
Binary files differ
diff --git a/src/archive/tar/testdata/pax.tar b/src/archive/tar/testdata/pax.tar
new file mode 100644
index 000000000..9bc24b658
--- /dev/null
+++ b/src/archive/tar/testdata/pax.tar
Binary files differ
diff --git a/src/archive/tar/testdata/small.txt b/src/archive/tar/testdata/small.txt
new file mode 100644
index 000000000..b249bfc51
--- /dev/null
+++ b/src/archive/tar/testdata/small.txt
@@ -0,0 +1 @@
+Kilts \ No newline at end of file
diff --git a/src/archive/tar/testdata/small2.txt b/src/archive/tar/testdata/small2.txt
new file mode 100644
index 000000000..394ee3ecd
--- /dev/null
+++ b/src/archive/tar/testdata/small2.txt
@@ -0,0 +1 @@
+Google.com
diff --git a/src/archive/tar/testdata/sparse-formats.tar b/src/archive/tar/testdata/sparse-formats.tar
new file mode 100644
index 000000000..8bd4e74d5
--- /dev/null
+++ b/src/archive/tar/testdata/sparse-formats.tar
Binary files differ
diff --git a/src/archive/tar/testdata/star.tar b/src/archive/tar/testdata/star.tar
new file mode 100644
index 000000000..59e2d4e60
--- /dev/null
+++ b/src/archive/tar/testdata/star.tar
Binary files differ
diff --git a/src/archive/tar/testdata/ustar.tar b/src/archive/tar/testdata/ustar.tar
new file mode 100644
index 000000000..29679d9a3
--- /dev/null
+++ b/src/archive/tar/testdata/ustar.tar
Binary files differ
diff --git a/src/archive/tar/testdata/v7.tar b/src/archive/tar/testdata/v7.tar
new file mode 100644
index 000000000..eb65fc941
--- /dev/null
+++ b/src/archive/tar/testdata/v7.tar
Binary files differ
diff --git a/src/archive/tar/testdata/writer-big-long.tar b/src/archive/tar/testdata/writer-big-long.tar
new file mode 100644
index 000000000..5960ee824
--- /dev/null
+++ b/src/archive/tar/testdata/writer-big-long.tar
Binary files differ
diff --git a/src/archive/tar/testdata/writer-big.tar b/src/archive/tar/testdata/writer-big.tar
new file mode 100644
index 000000000..753e883ce
--- /dev/null
+++ b/src/archive/tar/testdata/writer-big.tar
Binary files differ
diff --git a/src/archive/tar/testdata/writer.tar b/src/archive/tar/testdata/writer.tar
new file mode 100644
index 000000000..e6d816ad0
--- /dev/null
+++ b/src/archive/tar/testdata/writer.tar
Binary files differ
diff --git a/src/archive/tar/testdata/xattrs.tar b/src/archive/tar/testdata/xattrs.tar
new file mode 100644
index 000000000..9701950ed
--- /dev/null
+++ b/src/archive/tar/testdata/xattrs.tar
Binary files differ
diff --git a/src/archive/tar/writer.go b/src/archive/tar/writer.go
new file mode 100644
index 000000000..dafb2cabf
--- /dev/null
+++ b/src/archive/tar/writer.go
@@ -0,0 +1,396 @@
+// Copyright 2009 The Go Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+package tar
+
+// TODO(dsymonds):
+// - catch more errors (no first header, etc.)
+
+import (
+ "bytes"
+ "errors"
+ "fmt"
+ "io"
+ "os"
+ "path"
+ "strconv"
+ "strings"
+ "time"
+)
+
+var (
+ ErrWriteTooLong = errors.New("archive/tar: write too long")
+ ErrFieldTooLong = errors.New("archive/tar: header field too long")
+ ErrWriteAfterClose = errors.New("archive/tar: write after close")
+ errNameTooLong = errors.New("archive/tar: name too long")
+ errInvalidHeader = errors.New("archive/tar: header field too long or contains invalid values")
+)
+
+// A Writer provides sequential writing of a tar archive in POSIX.1 format.
+// A tar archive consists of a sequence of files.
+// Call WriteHeader to begin a new file, and then call Write to supply that file's data,
+// writing at most hdr.Size bytes in total.
+type Writer struct {
+ w io.Writer
+ err error
+ nb int64 // number of unwritten bytes for current file entry
+ pad int64 // amount of padding to write after current file entry
+ closed bool
+ usedBinary bool // whether the binary numeric field extension was used
+ preferPax bool // use pax header instead of binary numeric header
+ hdrBuff [blockSize]byte // buffer to use in writeHeader when writing a regular header
+ paxHdrBuff [blockSize]byte // buffer to use in writeHeader when writing a pax header
+}
+
+// NewWriter creates a new Writer writing to w.
+func NewWriter(w io.Writer) *Writer { return &Writer{w: w} }
+
+// Flush finishes writing the current file (optional).
+func (tw *Writer) Flush() error {
+ if tw.nb > 0 {
+ tw.err = fmt.Errorf("archive/tar: missed writing %d bytes", tw.nb)
+ return tw.err
+ }
+
+ n := tw.nb + tw.pad
+ for n > 0 && tw.err == nil {
+ nr := n
+ if nr > blockSize {
+ nr = blockSize
+ }
+ var nw int
+ nw, tw.err = tw.w.Write(zeroBlock[0:nr])
+ n -= int64(nw)
+ }
+ tw.nb = 0
+ tw.pad = 0
+ return tw.err
+}
+
+// Write s into b, terminating it with a NUL if there is room.
+// If the value is too long for the field and allowPax is true add a paxheader record instead
+func (tw *Writer) cString(b []byte, s string, allowPax bool, paxKeyword string, paxHeaders map[string]string) {
+ needsPaxHeader := allowPax && len(s) > len(b) || !isASCII(s)
+ if needsPaxHeader {
+ paxHeaders[paxKeyword] = s
+ return
+ }
+ if len(s) > len(b) {
+ if tw.err == nil {
+ tw.err = ErrFieldTooLong
+ }
+ return
+ }
+ ascii := toASCII(s)
+ copy(b, ascii)
+ if len(ascii) < len(b) {
+ b[len(ascii)] = 0
+ }
+}
+
+// Encode x as an octal ASCII string and write it into b with leading zeros.
+func (tw *Writer) octal(b []byte, x int64) {
+ s := strconv.FormatInt(x, 8)
+ // leading zeros, but leave room for a NUL.
+ for len(s)+1 < len(b) {
+ s = "0" + s
+ }
+ tw.cString(b, s, false, paxNone, nil)
+}
+
+// Write x into b, either as octal or as binary (GNUtar/star extension).
+// If the value is too long for the field and writingPax is enabled both for the field and the add a paxheader record instead
+func (tw *Writer) numeric(b []byte, x int64, allowPax bool, paxKeyword string, paxHeaders map[string]string) {
+ // Try octal first.
+ s := strconv.FormatInt(x, 8)
+ if len(s) < len(b) {
+ tw.octal(b, x)
+ return
+ }
+
+ // If it is too long for octal, and pax is preferred, use a pax header
+ if allowPax && tw.preferPax {
+ tw.octal(b, 0)
+ s := strconv.FormatInt(x, 10)
+ paxHeaders[paxKeyword] = s
+ return
+ }
+
+ // Too big: use binary (big-endian).
+ tw.usedBinary = true
+ for i := len(b) - 1; x > 0 && i >= 0; i-- {
+ b[i] = byte(x)
+ x >>= 8
+ }
+ b[0] |= 0x80 // highest bit indicates binary format
+}
+
+var (
+ minTime = time.Unix(0, 0)
+ // There is room for 11 octal digits (33 bits) of mtime.
+ maxTime = minTime.Add((1<<33 - 1) * time.Second)
+)
+
+// WriteHeader writes hdr and prepares to accept the file's contents.
+// WriteHeader calls Flush if it is not the first header.
+// Calling after a Close will return ErrWriteAfterClose.
+func (tw *Writer) WriteHeader(hdr *Header) error {
+ return tw.writeHeader(hdr, true)
+}
+
+// WriteHeader writes hdr and prepares to accept the file's contents.
+// WriteHeader calls Flush if it is not the first header.
+// Calling after a Close will return ErrWriteAfterClose.
+// As this method is called internally by writePax header to allow it to
+// suppress writing the pax header.
+func (tw *Writer) writeHeader(hdr *Header, allowPax bool) error {
+ if tw.closed {
+ return ErrWriteAfterClose
+ }
+ if tw.err == nil {
+ tw.Flush()
+ }
+ if tw.err != nil {
+ return tw.err
+ }
+
+ // a map to hold pax header records, if any are needed
+ paxHeaders := make(map[string]string)
+
+ // TODO(shanemhansen): we might want to use PAX headers for
+ // subsecond time resolution, but for now let's just capture
+ // too long fields or non ascii characters
+
+ var header []byte
+
+ // We need to select which scratch buffer to use carefully,
+ // since this method is called recursively to write PAX headers.
+ // If allowPax is true, this is the non-recursive call, and we will use hdrBuff.
+ // If allowPax is false, we are being called by writePAXHeader, and hdrBuff is
+ // already being used by the non-recursive call, so we must use paxHdrBuff.
+ header = tw.hdrBuff[:]
+ if !allowPax {
+ header = tw.paxHdrBuff[:]
+ }
+ copy(header, zeroBlock)
+ s := slicer(header)
+
+ // keep a reference to the filename to allow to overwrite it later if we detect that we can use ustar longnames instead of pax
+ pathHeaderBytes := s.next(fileNameSize)
+
+ tw.cString(pathHeaderBytes, hdr.Name, true, paxPath, paxHeaders)
+
+ // Handle out of range ModTime carefully.
+ var modTime int64
+ if !hdr.ModTime.Before(minTime) && !hdr.ModTime.After(maxTime) {
+ modTime = hdr.ModTime.Unix()
+ }
+
+ tw.octal(s.next(8), hdr.Mode) // 100:108
+ tw.numeric(s.next(8), int64(hdr.Uid), true, paxUid, paxHeaders) // 108:116
+ tw.numeric(s.next(8), int64(hdr.Gid), true, paxGid, paxHeaders) // 116:124
+ tw.numeric(s.next(12), hdr.Size, true, paxSize, paxHeaders) // 124:136
+ tw.numeric(s.next(12), modTime, false, paxNone, nil) // 136:148 --- consider using pax for finer granularity
+ s.next(8) // chksum (148:156)
+ s.next(1)[0] = hdr.Typeflag // 156:157
+
+ tw.cString(s.next(100), hdr.Linkname, true, paxLinkpath, paxHeaders)
+
+ copy(s.next(8), []byte("ustar\x0000")) // 257:265
+ tw.cString(s.next(32), hdr.Uname, true, paxUname, paxHeaders) // 265:297
+ tw.cString(s.next(32), hdr.Gname, true, paxGname, paxHeaders) // 297:329
+ tw.numeric(s.next(8), hdr.Devmajor, false, paxNone, nil) // 329:337
+ tw.numeric(s.next(8), hdr.Devminor, false, paxNone, nil) // 337:345
+
+ // keep a reference to the prefix to allow to overwrite it later if we detect that we can use ustar longnames instead of pax
+ prefixHeaderBytes := s.next(155)
+ tw.cString(prefixHeaderBytes, "", false, paxNone, nil) // 345:500 prefix
+
+ // Use the GNU magic instead of POSIX magic if we used any GNU extensions.
+ if tw.usedBinary {
+ copy(header[257:265], []byte("ustar \x00"))
+ }
+
+ _, paxPathUsed := paxHeaders[paxPath]
+ // try to use a ustar header when only the name is too long
+ if !tw.preferPax && len(paxHeaders) == 1 && paxPathUsed {
+ suffix := hdr.Name
+ prefix := ""
+ if len(hdr.Name) > fileNameSize && isASCII(hdr.Name) {
+ var err error
+ prefix, suffix, err = tw.splitUSTARLongName(hdr.Name)
+ if err == nil {
+ // ok we can use a ustar long name instead of pax, now correct the fields
+
+ // remove the path field from the pax header. this will suppress the pax header
+ delete(paxHeaders, paxPath)
+
+ // update the path fields
+ tw.cString(pathHeaderBytes, suffix, false, paxNone, nil)
+ tw.cString(prefixHeaderBytes, prefix, false, paxNone, nil)
+
+ // Use the ustar magic if we used ustar long names.
+ if len(prefix) > 0 && !tw.usedBinary {
+ copy(header[257:265], []byte("ustar\x00"))
+ }
+ }
+ }
+ }
+
+ // The chksum field is terminated by a NUL and a space.
+ // This is different from the other octal fields.
+ chksum, _ := checksum(header)
+ tw.octal(header[148:155], chksum)
+ header[155] = ' '
+
+ if tw.err != nil {
+ // problem with header; probably integer too big for a field.
+ return tw.err
+ }
+
+ if allowPax {
+ for k, v := range hdr.Xattrs {
+ paxHeaders[paxXattr+k] = v
+ }
+ }
+
+ if len(paxHeaders) > 0 {
+ if !allowPax {
+ return errInvalidHeader
+ }
+ if err := tw.writePAXHeader(hdr, paxHeaders); err != nil {
+ return err
+ }
+ }
+ tw.nb = int64(hdr.Size)
+ tw.pad = (blockSize - (tw.nb % blockSize)) % blockSize
+
+ _, tw.err = tw.w.Write(header)
+ return tw.err
+}
+
+// writeUSTARLongName splits a USTAR long name hdr.Name.
+// name must be < 256 characters. errNameTooLong is returned
+// if hdr.Name can't be split. The splitting heuristic
+// is compatible with gnu tar.
+func (tw *Writer) splitUSTARLongName(name string) (prefix, suffix string, err error) {
+ length := len(name)
+ if length > fileNamePrefixSize+1 {
+ length = fileNamePrefixSize + 1
+ } else if name[length-1] == '/' {
+ length--
+ }
+ i := strings.LastIndex(name[:length], "/")
+ // nlen contains the resulting length in the name field.
+ // plen contains the resulting length in the prefix field.
+ nlen := len(name) - i - 1
+ plen := i
+ if i <= 0 || nlen > fileNameSize || nlen == 0 || plen > fileNamePrefixSize {
+ err = errNameTooLong
+ return
+ }
+ prefix, suffix = name[:i], name[i+1:]
+ return
+}
+
+// writePaxHeader writes an extended pax header to the
+// archive.
+func (tw *Writer) writePAXHeader(hdr *Header, paxHeaders map[string]string) error {
+ // Prepare extended header
+ ext := new(Header)
+ ext.Typeflag = TypeXHeader
+ // Setting ModTime is required for reader parsing to
+ // succeed, and seems harmless enough.
+ ext.ModTime = hdr.ModTime
+ // The spec asks that we namespace our pseudo files
+ // with the current pid.
+ pid := os.Getpid()
+ dir, file := path.Split(hdr.Name)
+ fullName := path.Join(dir,
+ fmt.Sprintf("PaxHeaders.%d", pid), file)
+
+ ascii := toASCII(fullName)
+ if len(ascii) > 100 {
+ ascii = ascii[:100]
+ }
+ ext.Name = ascii
+ // Construct the body
+ var buf bytes.Buffer
+
+ for k, v := range paxHeaders {
+ fmt.Fprint(&buf, paxHeader(k+"="+v))
+ }
+
+ ext.Size = int64(len(buf.Bytes()))
+ if err := tw.writeHeader(ext, false); err != nil {
+ return err
+ }
+ if _, err := tw.Write(buf.Bytes()); err != nil {
+ return err
+ }
+ if err := tw.Flush(); err != nil {
+ return err
+ }
+ return nil
+}
+
+// paxHeader formats a single pax record, prefixing it with the appropriate length
+func paxHeader(msg string) string {
+ const padding = 2 // Extra padding for space and newline
+ size := len(msg) + padding
+ size += len(strconv.Itoa(size))
+ record := fmt.Sprintf("%d %s\n", size, msg)
+ if len(record) != size {
+ // Final adjustment if adding size increased
+ // the number of digits in size
+ size = len(record)
+ record = fmt.Sprintf("%d %s\n", size, msg)
+ }
+ return record
+}
+
+// Write writes to the current entry in the tar archive.
+// Write returns the error ErrWriteTooLong if more than
+// hdr.Size bytes are written after WriteHeader.
+func (tw *Writer) Write(b []byte) (n int, err error) {
+ if tw.closed {
+ err = ErrWriteTooLong
+ return
+ }
+ overwrite := false
+ if int64(len(b)) > tw.nb {
+ b = b[0:tw.nb]
+ overwrite = true
+ }
+ n, err = tw.w.Write(b)
+ tw.nb -= int64(n)
+ if err == nil && overwrite {
+ err = ErrWriteTooLong
+ return
+ }
+ tw.err = err
+ return
+}
+
+// Close closes the tar archive, flushing any unwritten
+// data to the underlying writer.
+func (tw *Writer) Close() error {
+ if tw.err != nil || tw.closed {
+ return tw.err
+ }
+ tw.Flush()
+ tw.closed = true
+ if tw.err != nil {
+ return tw.err
+ }
+
+ // trailer: two zero blocks
+ for i := 0; i < 2; i++ {
+ _, tw.err = tw.w.Write(zeroBlock)
+ if tw.err != nil {
+ break
+ }
+ }
+ return tw.err
+}
diff --git a/src/archive/tar/writer_test.go b/src/archive/tar/writer_test.go
new file mode 100644
index 000000000..5e42e322f
--- /dev/null
+++ b/src/archive/tar/writer_test.go
@@ -0,0 +1,491 @@
+// Copyright 2009 The Go Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+package tar
+
+import (
+ "bytes"
+ "fmt"
+ "io"
+ "io/ioutil"
+ "os"
+ "reflect"
+ "strings"
+ "testing"
+ "testing/iotest"
+ "time"
+)
+
+type writerTestEntry struct {
+ header *Header
+ contents string
+}
+
+type writerTest struct {
+ file string // filename of expected output
+ entries []*writerTestEntry
+}
+
+var writerTests = []*writerTest{
+ // The writer test file was produced with this command:
+ // tar (GNU tar) 1.26
+ // ln -s small.txt link.txt
+ // tar -b 1 --format=ustar -c -f writer.tar small.txt small2.txt link.txt
+ {
+ file: "testdata/writer.tar",
+ entries: []*writerTestEntry{
+ {
+ header: &Header{
+ Name: "small.txt",
+ Mode: 0640,
+ Uid: 73025,
+ Gid: 5000,
+ Size: 5,
+ ModTime: time.Unix(1246508266, 0),
+ Typeflag: '0',
+ Uname: "dsymonds",
+ Gname: "eng",
+ },
+ contents: "Kilts",
+ },
+ {
+ header: &Header{
+ Name: "small2.txt",
+ Mode: 0640,
+ Uid: 73025,
+ Gid: 5000,
+ Size: 11,
+ ModTime: time.Unix(1245217492, 0),
+ Typeflag: '0',
+ Uname: "dsymonds",
+ Gname: "eng",
+ },
+ contents: "Google.com\n",
+ },
+ {
+ header: &Header{
+ Name: "link.txt",
+ Mode: 0777,
+ Uid: 1000,
+ Gid: 1000,
+ Size: 0,
+ ModTime: time.Unix(1314603082, 0),
+ Typeflag: '2',
+ Linkname: "small.txt",
+ Uname: "strings",
+ Gname: "strings",
+ },
+ // no contents
+ },
+ },
+ },
+ // The truncated test file was produced using these commands:
+ // dd if=/dev/zero bs=1048576 count=16384 > /tmp/16gig.txt
+ // tar -b 1 -c -f- /tmp/16gig.txt | dd bs=512 count=8 > writer-big.tar
+ {
+ file: "testdata/writer-big.tar",
+ entries: []*writerTestEntry{
+ {
+ header: &Header{
+ Name: "tmp/16gig.txt",
+ Mode: 0640,
+ Uid: 73025,
+ Gid: 5000,
+ Size: 16 << 30,
+ ModTime: time.Unix(1254699560, 0),
+ Typeflag: '0',
+ Uname: "dsymonds",
+ Gname: "eng",
+ },
+ // fake contents
+ contents: strings.Repeat("\x00", 4<<10),
+ },
+ },
+ },
+ // The truncated test file was produced using these commands:
+ // dd if=/dev/zero bs=1048576 count=16384 > (longname/)*15 /16gig.txt
+ // tar -b 1 -c -f- (longname/)*15 /16gig.txt | dd bs=512 count=8 > writer-big-long.tar
+ {
+ file: "testdata/writer-big-long.tar",
+ entries: []*writerTestEntry{
+ {
+ header: &Header{
+ Name: strings.Repeat("longname/", 15) + "16gig.txt",
+ Mode: 0644,
+ Uid: 1000,
+ Gid: 1000,
+ Size: 16 << 30,
+ ModTime: time.Unix(1399583047, 0),
+ Typeflag: '0',
+ Uname: "guillaume",
+ Gname: "guillaume",
+ },
+ // fake contents
+ contents: strings.Repeat("\x00", 4<<10),
+ },
+ },
+ },
+ // This file was produced using gnu tar 1.17
+ // gnutar -b 4 --format=ustar (longname/)*15 + file.txt
+ {
+ file: "testdata/ustar.tar",
+ entries: []*writerTestEntry{
+ {
+ header: &Header{
+ Name: strings.Repeat("longname/", 15) + "file.txt",
+ Mode: 0644,
+ Uid: 0765,
+ Gid: 024,
+ Size: 06,
+ ModTime: time.Unix(1360135598, 0),
+ Typeflag: '0',
+ Uname: "shane",
+ Gname: "staff",
+ },
+ contents: "hello\n",
+ },
+ },
+ },
+}
+
+// Render byte array in a two-character hexadecimal string, spaced for easy visual inspection.
+func bytestr(offset int, b []byte) string {
+ const rowLen = 32
+ s := fmt.Sprintf("%04x ", offset)
+ for _, ch := range b {
+ switch {
+ case '0' <= ch && ch <= '9', 'A' <= ch && ch <= 'Z', 'a' <= ch && ch <= 'z':
+ s += fmt.Sprintf(" %c", ch)
+ default:
+ s += fmt.Sprintf(" %02x", ch)
+ }
+ }
+ return s
+}
+
+// Render a pseudo-diff between two blocks of bytes.
+func bytediff(a []byte, b []byte) string {
+ const rowLen = 32
+ s := fmt.Sprintf("(%d bytes vs. %d bytes)\n", len(a), len(b))
+ for offset := 0; len(a)+len(b) > 0; offset += rowLen {
+ na, nb := rowLen, rowLen
+ if na > len(a) {
+ na = len(a)
+ }
+ if nb > len(b) {
+ nb = len(b)
+ }
+ sa := bytestr(offset, a[0:na])
+ sb := bytestr(offset, b[0:nb])
+ if sa != sb {
+ s += fmt.Sprintf("-%v\n+%v\n", sa, sb)
+ }
+ a = a[na:]
+ b = b[nb:]
+ }
+ return s
+}
+
+func TestWriter(t *testing.T) {
+testLoop:
+ for i, test := range writerTests {
+ expected, err := ioutil.ReadFile(test.file)
+ if err != nil {
+ t.Errorf("test %d: Unexpected error: %v", i, err)
+ continue
+ }
+
+ buf := new(bytes.Buffer)
+ tw := NewWriter(iotest.TruncateWriter(buf, 4<<10)) // only catch the first 4 KB
+ big := false
+ for j, entry := range test.entries {
+ big = big || entry.header.Size > 1<<10
+ if err := tw.WriteHeader(entry.header); err != nil {
+ t.Errorf("test %d, entry %d: Failed writing header: %v", i, j, err)
+ continue testLoop
+ }
+ if _, err := io.WriteString(tw, entry.contents); err != nil {
+ t.Errorf("test %d, entry %d: Failed writing contents: %v", i, j, err)
+ continue testLoop
+ }
+ }
+ // Only interested in Close failures for the small tests.
+ if err := tw.Close(); err != nil && !big {
+ t.Errorf("test %d: Failed closing archive: %v", i, err)
+ continue testLoop
+ }
+
+ actual := buf.Bytes()
+ if !bytes.Equal(expected, actual) {
+ t.Errorf("test %d: Incorrect result: (-=expected, +=actual)\n%v",
+ i, bytediff(expected, actual))
+ }
+ if testing.Short() { // The second test is expensive.
+ break
+ }
+ }
+}
+
+func TestPax(t *testing.T) {
+ // Create an archive with a large name
+ fileinfo, err := os.Stat("testdata/small.txt")
+ if err != nil {
+ t.Fatal(err)
+ }
+ hdr, err := FileInfoHeader(fileinfo, "")
+ if err != nil {
+ t.Fatalf("os.Stat: %v", err)
+ }
+ // Force a PAX long name to be written
+ longName := strings.Repeat("ab", 100)
+ contents := strings.Repeat(" ", int(hdr.Size))
+ hdr.Name = longName
+ var buf bytes.Buffer
+ writer := NewWriter(&buf)
+ if err := writer.WriteHeader(hdr); err != nil {
+ t.Fatal(err)
+ }
+ if _, err = writer.Write([]byte(contents)); err != nil {
+ t.Fatal(err)
+ }
+ if err := writer.Close(); err != nil {
+ t.Fatal(err)
+ }
+ // Simple test to make sure PAX extensions are in effect
+ if !bytes.Contains(buf.Bytes(), []byte("PaxHeaders.")) {
+ t.Fatal("Expected at least one PAX header to be written.")
+ }
+ // Test that we can get a long name back out of the archive.
+ reader := NewReader(&buf)
+ hdr, err = reader.Next()
+ if err != nil {
+ t.Fatal(err)
+ }
+ if hdr.Name != longName {
+ t.Fatal("Couldn't recover long file name")
+ }
+}
+
+func TestPaxSymlink(t *testing.T) {
+ // Create an archive with a large linkname
+ fileinfo, err := os.Stat("testdata/small.txt")
+ if err != nil {
+ t.Fatal(err)
+ }
+ hdr, err := FileInfoHeader(fileinfo, "")
+ hdr.Typeflag = TypeSymlink
+ if err != nil {
+ t.Fatalf("os.Stat:1 %v", err)
+ }
+ // Force a PAX long linkname to be written
+ longLinkname := strings.Repeat("1234567890/1234567890", 10)
+ hdr.Linkname = longLinkname
+
+ hdr.Size = 0
+ var buf bytes.Buffer
+ writer := NewWriter(&buf)
+ if err := writer.WriteHeader(hdr); err != nil {
+ t.Fatal(err)
+ }
+ if err := writer.Close(); err != nil {
+ t.Fatal(err)
+ }
+ // Simple test to make sure PAX extensions are in effect
+ if !bytes.Contains(buf.Bytes(), []byte("PaxHeaders.")) {
+ t.Fatal("Expected at least one PAX header to be written.")
+ }
+ // Test that we can get a long name back out of the archive.
+ reader := NewReader(&buf)
+ hdr, err = reader.Next()
+ if err != nil {
+ t.Fatal(err)
+ }
+ if hdr.Linkname != longLinkname {
+ t.Fatal("Couldn't recover long link name")
+ }
+}
+
+func TestPaxNonAscii(t *testing.T) {
+ // Create an archive with non ascii. These should trigger a pax header
+ // because pax headers have a defined utf-8 encoding.
+ fileinfo, err := os.Stat("testdata/small.txt")
+ if err != nil {
+ t.Fatal(err)
+ }
+
+ hdr, err := FileInfoHeader(fileinfo, "")
+ if err != nil {
+ t.Fatalf("os.Stat:1 %v", err)
+ }
+
+ // some sample data
+ chineseFilename := "文件名"
+ chineseGroupname := "組"
+ chineseUsername := "用戶名"
+
+ hdr.Name = chineseFilename
+ hdr.Gname = chineseGroupname
+ hdr.Uname = chineseUsername
+
+ contents := strings.Repeat(" ", int(hdr.Size))
+
+ var buf bytes.Buffer
+ writer := NewWriter(&buf)
+ if err := writer.WriteHeader(hdr); err != nil {
+ t.Fatal(err)
+ }
+ if _, err = writer.Write([]byte(contents)); err != nil {
+ t.Fatal(err)
+ }
+ if err := writer.Close(); err != nil {
+ t.Fatal(err)
+ }
+ // Simple test to make sure PAX extensions are in effect
+ if !bytes.Contains(buf.Bytes(), []byte("PaxHeaders.")) {
+ t.Fatal("Expected at least one PAX header to be written.")
+ }
+ // Test that we can get a long name back out of the archive.
+ reader := NewReader(&buf)
+ hdr, err = reader.Next()
+ if err != nil {
+ t.Fatal(err)
+ }
+ if hdr.Name != chineseFilename {
+ t.Fatal("Couldn't recover unicode name")
+ }
+ if hdr.Gname != chineseGroupname {
+ t.Fatal("Couldn't recover unicode group")
+ }
+ if hdr.Uname != chineseUsername {
+ t.Fatal("Couldn't recover unicode user")
+ }
+}
+
+func TestPaxXattrs(t *testing.T) {
+ xattrs := map[string]string{
+ "user.key": "value",
+ }
+
+ // Create an archive with an xattr
+ fileinfo, err := os.Stat("testdata/small.txt")
+ if err != nil {
+ t.Fatal(err)
+ }
+ hdr, err := FileInfoHeader(fileinfo, "")
+ if err != nil {
+ t.Fatalf("os.Stat: %v", err)
+ }
+ contents := "Kilts"
+ hdr.Xattrs = xattrs
+ var buf bytes.Buffer
+ writer := NewWriter(&buf)
+ if err := writer.WriteHeader(hdr); err != nil {
+ t.Fatal(err)
+ }
+ if _, err = writer.Write([]byte(contents)); err != nil {
+ t.Fatal(err)
+ }
+ if err := writer.Close(); err != nil {
+ t.Fatal(err)
+ }
+ // Test that we can get the xattrs back out of the archive.
+ reader := NewReader(&buf)
+ hdr, err = reader.Next()
+ if err != nil {
+ t.Fatal(err)
+ }
+ if !reflect.DeepEqual(hdr.Xattrs, xattrs) {
+ t.Fatalf("xattrs did not survive round trip: got %+v, want %+v",
+ hdr.Xattrs, xattrs)
+ }
+}
+
+func TestPAXHeader(t *testing.T) {
+ medName := strings.Repeat("CD", 50)
+ longName := strings.Repeat("AB", 100)
+ paxTests := [][2]string{
+ {paxPath + "=/etc/hosts", "19 path=/etc/hosts\n"},
+ {"a=b", "6 a=b\n"}, // Single digit length
+ {"a=names", "11 a=names\n"}, // Test case involving carries
+ {paxPath + "=" + longName, fmt.Sprintf("210 path=%s\n", longName)},
+ {paxPath + "=" + medName, fmt.Sprintf("110 path=%s\n", medName)}}
+
+ for _, test := range paxTests {
+ key, expected := test[0], test[1]
+ if result := paxHeader(key); result != expected {
+ t.Fatalf("paxHeader: got %s, expected %s", result, expected)
+ }
+ }
+}
+
+func TestUSTARLongName(t *testing.T) {
+ // Create an archive with a path that failed to split with USTAR extension in previous versions.
+ fileinfo, err := os.Stat("testdata/small.txt")
+ if err != nil {
+ t.Fatal(err)
+ }
+ hdr, err := FileInfoHeader(fileinfo, "")
+ hdr.Typeflag = TypeDir
+ if err != nil {
+ t.Fatalf("os.Stat:1 %v", err)
+ }
+ // Force a PAX long name to be written. The name was taken from a practical example
+ // that fails and replaced ever char through numbers to anonymize the sample.
+ longName := "/0000_0000000/00000-000000000/0000_0000000/00000-0000000000000/0000_0000000/00000-0000000-00000000/0000_0000000/00000000/0000_0000000/000/0000_0000000/00000000v00/0000_0000000/000000/0000_0000000/0000000/0000_0000000/00000y-00/0000/0000/00000000/0x000000/"
+ hdr.Name = longName
+
+ hdr.Size = 0
+ var buf bytes.Buffer
+ writer := NewWriter(&buf)
+ if err := writer.WriteHeader(hdr); err != nil {
+ t.Fatal(err)
+ }
+ if err := writer.Close(); err != nil {
+ t.Fatal(err)
+ }
+ // Test that we can get a long name back out of the archive.
+ reader := NewReader(&buf)
+ hdr, err = reader.Next()
+ if err != nil {
+ t.Fatal(err)
+ }
+ if hdr.Name != longName {
+ t.Fatal("Couldn't recover long name")
+ }
+}
+
+func TestValidTypeflagWithPAXHeader(t *testing.T) {
+ var buffer bytes.Buffer
+ tw := NewWriter(&buffer)
+
+ fileName := strings.Repeat("ab", 100)
+
+ hdr := &Header{
+ Name: fileName,
+ Size: 4,
+ Typeflag: 0,
+ }
+ if err := tw.WriteHeader(hdr); err != nil {
+ t.Fatalf("Failed to write header: %s", err)
+ }
+ if _, err := tw.Write([]byte("fooo")); err != nil {
+ t.Fatalf("Failed to write the file's data: %s", err)
+ }
+ tw.Close()
+
+ tr := NewReader(&buffer)
+
+ for {
+ header, err := tr.Next()
+ if err == io.EOF {
+ break
+ }
+ if err != nil {
+ t.Fatalf("Failed to read header: %s", err)
+ }
+ if header.Typeflag != 0 {
+ t.Fatalf("Typeflag should've been 0, found %d", header.Typeflag)
+ }
+ }
+}
diff --git a/src/archive/zip/example_test.go b/src/archive/zip/example_test.go
new file mode 100644
index 000000000..c2ed9e79c
--- /dev/null
+++ b/src/archive/zip/example_test.go
@@ -0,0 +1,75 @@
+// Copyright 2012 The Go Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+package zip_test
+
+import (
+ "archive/zip"
+ "bytes"
+ "fmt"
+ "io"
+ "log"
+ "os"
+)
+
+func ExampleWriter() {
+ // Create a buffer to write our archive to.
+ buf := new(bytes.Buffer)
+
+ // Create a new zip archive.
+ w := zip.NewWriter(buf)
+
+ // Add some files to the archive.
+ var files = []struct {
+ Name, Body string
+ }{
+ {"readme.txt", "This archive contains some text files."},
+ {"gopher.txt", "Gopher names:\nGeorge\nGeoffrey\nGonzo"},
+ {"todo.txt", "Get animal handling licence.\nWrite more examples."},
+ }
+ for _, file := range files {
+ f, err := w.Create(file.Name)
+ if err != nil {
+ log.Fatal(err)
+ }
+ _, err = f.Write([]byte(file.Body))
+ if err != nil {
+ log.Fatal(err)
+ }
+ }
+
+ // Make sure to check the error on Close.
+ err := w.Close()
+ if err != nil {
+ log.Fatal(err)
+ }
+}
+
+func ExampleReader() {
+ // Open a zip archive for reading.
+ r, err := zip.OpenReader("testdata/readme.zip")
+ if err != nil {
+ log.Fatal(err)
+ }
+ defer r.Close()
+
+ // Iterate through the files in the archive,
+ // printing some of their contents.
+ for _, f := range r.File {
+ fmt.Printf("Contents of %s:\n", f.Name)
+ rc, err := f.Open()
+ if err != nil {
+ log.Fatal(err)
+ }
+ _, err = io.CopyN(os.Stdout, rc, 68)
+ if err != nil {
+ log.Fatal(err)
+ }
+ rc.Close()
+ fmt.Println()
+ }
+ // Output:
+ // Contents of README:
+ // This is the source code repository for the Go programming language.
+}
diff --git a/src/archive/zip/reader.go b/src/archive/zip/reader.go
new file mode 100644
index 000000000..8136b840d
--- /dev/null
+++ b/src/archive/zip/reader.go
@@ -0,0 +1,453 @@
+// Copyright 2010 The Go Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+package zip
+
+import (
+ "bufio"
+ "encoding/binary"
+ "errors"
+ "hash"
+ "hash/crc32"
+ "io"
+ "os"
+)
+
+var (
+ ErrFormat = errors.New("zip: not a valid zip file")
+ ErrAlgorithm = errors.New("zip: unsupported compression algorithm")
+ ErrChecksum = errors.New("zip: checksum error")
+)
+
+type Reader struct {
+ r io.ReaderAt
+ File []*File
+ Comment string
+}
+
+type ReadCloser struct {
+ f *os.File
+ Reader
+}
+
+type File struct {
+ FileHeader
+ zipr io.ReaderAt
+ zipsize int64
+ headerOffset int64
+}
+
+func (f *File) hasDataDescriptor() bool {
+ return f.Flags&0x8 != 0
+}
+
+// OpenReader will open the Zip file specified by name and return a ReadCloser.
+func OpenReader(name string) (*ReadCloser, error) {
+ f, err := os.Open(name)
+ if err != nil {
+ return nil, err
+ }
+ fi, err := f.Stat()
+ if err != nil {
+ f.Close()
+ return nil, err
+ }
+ r := new(ReadCloser)
+ if err := r.init(f, fi.Size()); err != nil {
+ f.Close()
+ return nil, err
+ }
+ r.f = f
+ return r, nil
+}
+
+// NewReader returns a new Reader reading from r, which is assumed to
+// have the given size in bytes.
+func NewReader(r io.ReaderAt, size int64) (*Reader, error) {
+ zr := new(Reader)
+ if err := zr.init(r, size); err != nil {
+ return nil, err
+ }
+ return zr, nil
+}
+
+func (z *Reader) init(r io.ReaderAt, size int64) error {
+ end, err := readDirectoryEnd(r, size)
+ if err != nil {
+ return err
+ }
+ z.r = r
+ z.File = make([]*File, 0, end.directoryRecords)
+ z.Comment = end.comment
+ rs := io.NewSectionReader(r, 0, size)
+ if _, err = rs.Seek(int64(end.directoryOffset), os.SEEK_SET); err != nil {
+ return err
+ }
+ buf := bufio.NewReader(rs)
+
+ // The count of files inside a zip is truncated to fit in a uint16.
+ // Gloss over this by reading headers until we encounter
+ // a bad one, and then only report a ErrFormat or UnexpectedEOF if
+ // the file count modulo 65536 is incorrect.
+ for {
+ f := &File{zipr: r, zipsize: size}
+ err = readDirectoryHeader(f, buf)
+ if err == ErrFormat || err == io.ErrUnexpectedEOF {
+ break
+ }
+ if err != nil {
+ return err
+ }
+ z.File = append(z.File, f)
+ }
+ if uint16(len(z.File)) != uint16(end.directoryRecords) { // only compare 16 bits here
+ // Return the readDirectoryHeader error if we read
+ // the wrong number of directory entries.
+ return err
+ }
+ return nil
+}
+
+// Close closes the Zip file, rendering it unusable for I/O.
+func (rc *ReadCloser) Close() error {
+ return rc.f.Close()
+}
+
+// DataOffset returns the offset of the file's possibly-compressed
+// data, relative to the beginning of the zip file.
+//
+// Most callers should instead use Open, which transparently
+// decompresses data and verifies checksums.
+func (f *File) DataOffset() (offset int64, err error) {
+ bodyOffset, err := f.findBodyOffset()
+ if err != nil {
+ return
+ }
+ return f.headerOffset + bodyOffset, nil
+}
+
+// Open returns a ReadCloser that provides access to the File's contents.
+// Multiple files may be read concurrently.
+func (f *File) Open() (rc io.ReadCloser, err error) {
+ bodyOffset, err := f.findBodyOffset()
+ if err != nil {
+ return
+ }
+ size := int64(f.CompressedSize64)
+ r := io.NewSectionReader(f.zipr, f.headerOffset+bodyOffset, size)
+ dcomp := decompressor(f.Method)
+ if dcomp == nil {
+ err = ErrAlgorithm
+ return
+ }
+ rc = dcomp(r)
+ var desr io.Reader
+ if f.hasDataDescriptor() {
+ desr = io.NewSectionReader(f.zipr, f.headerOffset+bodyOffset+size, dataDescriptorLen)
+ }
+ rc = &checksumReader{rc, crc32.NewIEEE(), f, desr, nil}
+ return
+}
+
+type checksumReader struct {
+ rc io.ReadCloser
+ hash hash.Hash32
+ f *File
+ desr io.Reader // if non-nil, where to read the data descriptor
+ err error // sticky error
+}
+
+func (r *checksumReader) Read(b []byte) (n int, err error) {
+ if r.err != nil {
+ return 0, r.err
+ }
+ n, err = r.rc.Read(b)
+ r.hash.Write(b[:n])
+ if err == nil {
+ return
+ }
+ if err == io.EOF {
+ if r.desr != nil {
+ if err1 := readDataDescriptor(r.desr, r.f); err1 != nil {
+ err = err1
+ } else if r.hash.Sum32() != r.f.CRC32 {
+ err = ErrChecksum
+ }
+ } else {
+ // If there's not a data descriptor, we still compare
+ // the CRC32 of what we've read against the file header
+ // or TOC's CRC32, if it seems like it was set.
+ if r.f.CRC32 != 0 && r.hash.Sum32() != r.f.CRC32 {
+ err = ErrChecksum
+ }
+ }
+ }
+ r.err = err
+ return
+}
+
+func (r *checksumReader) Close() error { return r.rc.Close() }
+
+// findBodyOffset does the minimum work to verify the file has a header
+// and returns the file body offset.
+func (f *File) findBodyOffset() (int64, error) {
+ var buf [fileHeaderLen]byte
+ if _, err := f.zipr.ReadAt(buf[:], f.headerOffset); err != nil {
+ return 0, err
+ }
+ b := readBuf(buf[:])
+ if sig := b.uint32(); sig != fileHeaderSignature {
+ return 0, ErrFormat
+ }
+ b = b[22:] // skip over most of the header
+ filenameLen := int(b.uint16())
+ extraLen := int(b.uint16())
+ return int64(fileHeaderLen + filenameLen + extraLen), nil
+}
+
+// readDirectoryHeader attempts to read a directory header from r.
+// It returns io.ErrUnexpectedEOF if it cannot read a complete header,
+// and ErrFormat if it doesn't find a valid header signature.
+func readDirectoryHeader(f *File, r io.Reader) error {
+ var buf [directoryHeaderLen]byte
+ if _, err := io.ReadFull(r, buf[:]); err != nil {
+ return err
+ }
+ b := readBuf(buf[:])
+ if sig := b.uint32(); sig != directoryHeaderSignature {
+ return ErrFormat
+ }
+ f.CreatorVersion = b.uint16()
+ f.ReaderVersion = b.uint16()
+ f.Flags = b.uint16()
+ f.Method = b.uint16()
+ f.ModifiedTime = b.uint16()
+ f.ModifiedDate = b.uint16()
+ f.CRC32 = b.uint32()
+ f.CompressedSize = b.uint32()
+ f.UncompressedSize = b.uint32()
+ f.CompressedSize64 = uint64(f.CompressedSize)
+ f.UncompressedSize64 = uint64(f.UncompressedSize)
+ filenameLen := int(b.uint16())
+ extraLen := int(b.uint16())
+ commentLen := int(b.uint16())
+ b = b[4:] // skipped start disk number and internal attributes (2x uint16)
+ f.ExternalAttrs = b.uint32()
+ f.headerOffset = int64(b.uint32())
+ d := make([]byte, filenameLen+extraLen+commentLen)
+ if _, err := io.ReadFull(r, d); err != nil {
+ return err
+ }
+ f.Name = string(d[:filenameLen])
+ f.Extra = d[filenameLen : filenameLen+extraLen]
+ f.Comment = string(d[filenameLen+extraLen:])
+
+ if len(f.Extra) > 0 {
+ b := readBuf(f.Extra)
+ for len(b) >= 4 { // need at least tag and size
+ tag := b.uint16()
+ size := b.uint16()
+ if int(size) > len(b) {
+ return ErrFormat
+ }
+ if tag == zip64ExtraId {
+ // update directory values from the zip64 extra block
+ eb := readBuf(b[:size])
+ if len(eb) >= 8 {
+ f.UncompressedSize64 = eb.uint64()
+ }
+ if len(eb) >= 8 {
+ f.CompressedSize64 = eb.uint64()
+ }
+ if len(eb) >= 8 {
+ f.headerOffset = int64(eb.uint64())
+ }
+ }
+ b = b[size:]
+ }
+ // Should have consumed the whole header.
+ // But popular zip & JAR creation tools are broken and
+ // may pad extra zeros at the end, so accept those
+ // too. See golang.org/issue/8186.
+ for _, v := range b {
+ if v != 0 {
+ return ErrFormat
+ }
+ }
+ }
+ return nil
+}
+
+func readDataDescriptor(r io.Reader, f *File) error {
+ var buf [dataDescriptorLen]byte
+
+ // The spec says: "Although not originally assigned a
+ // signature, the value 0x08074b50 has commonly been adopted
+ // as a signature value for the data descriptor record.
+ // Implementers should be aware that ZIP files may be
+ // encountered with or without this signature marking data
+ // descriptors and should account for either case when reading
+ // ZIP files to ensure compatibility."
+ //
+ // dataDescriptorLen includes the size of the signature but
+ // first read just those 4 bytes to see if it exists.
+ if _, err := io.ReadFull(r, buf[:4]); err != nil {
+ return err
+ }
+ off := 0
+ maybeSig := readBuf(buf[:4])
+ if maybeSig.uint32() != dataDescriptorSignature {
+ // No data descriptor signature. Keep these four
+ // bytes.
+ off += 4
+ }
+ if _, err := io.ReadFull(r, buf[off:12]); err != nil {
+ return err
+ }
+ b := readBuf(buf[:12])
+ if b.uint32() != f.CRC32 {
+ return ErrChecksum
+ }
+
+ // The two sizes that follow here can be either 32 bits or 64 bits
+ // but the spec is not very clear on this and different
+ // interpretations has been made causing incompatibilities. We
+ // already have the sizes from the central directory so we can
+ // just ignore these.
+
+ return nil
+}
+
+func readDirectoryEnd(r io.ReaderAt, size int64) (dir *directoryEnd, err error) {
+ // look for directoryEndSignature in the last 1k, then in the last 65k
+ var buf []byte
+ var directoryEndOffset int64
+ for i, bLen := range []int64{1024, 65 * 1024} {
+ if bLen > size {
+ bLen = size
+ }
+ buf = make([]byte, int(bLen))
+ if _, err := r.ReadAt(buf, size-bLen); err != nil && err != io.EOF {
+ return nil, err
+ }
+ if p := findSignatureInBlock(buf); p >= 0 {
+ buf = buf[p:]
+ directoryEndOffset = size - bLen + int64(p)
+ break
+ }
+ if i == 1 || bLen == size {
+ return nil, ErrFormat
+ }
+ }
+
+ // read header into struct
+ b := readBuf(buf[4:]) // skip signature
+ d := &directoryEnd{
+ diskNbr: uint32(b.uint16()),
+ dirDiskNbr: uint32(b.uint16()),
+ dirRecordsThisDisk: uint64(b.uint16()),
+ directoryRecords: uint64(b.uint16()),
+ directorySize: uint64(b.uint32()),
+ directoryOffset: uint64(b.uint32()),
+ commentLen: b.uint16(),
+ }
+ l := int(d.commentLen)
+ if l > len(b) {
+ return nil, errors.New("zip: invalid comment length")
+ }
+ d.comment = string(b[:l])
+
+ p, err := findDirectory64End(r, directoryEndOffset)
+ if err == nil && p >= 0 {
+ err = readDirectory64End(r, p, d)
+ }
+ if err != nil {
+ return nil, err
+ }
+
+ // Make sure directoryOffset points to somewhere in our file.
+ if o := int64(d.directoryOffset); o < 0 || o >= size {
+ return nil, ErrFormat
+ }
+ return d, nil
+}
+
+// findDirectory64End tries to read the zip64 locator just before the
+// directory end and returns the offset of the zip64 directory end if
+// found.
+func findDirectory64End(r io.ReaderAt, directoryEndOffset int64) (int64, error) {
+ locOffset := directoryEndOffset - directory64LocLen
+ if locOffset < 0 {
+ return -1, nil // no need to look for a header outside the file
+ }
+ buf := make([]byte, directory64LocLen)
+ if _, err := r.ReadAt(buf, locOffset); err != nil {
+ return -1, err
+ }
+ b := readBuf(buf)
+ if sig := b.uint32(); sig != directory64LocSignature {
+ return -1, nil
+ }
+ b = b[4:] // skip number of the disk with the start of the zip64 end of central directory
+ p := b.uint64() // relative offset of the zip64 end of central directory record
+ return int64(p), nil
+}
+
+// readDirectory64End reads the zip64 directory end and updates the
+// directory end with the zip64 directory end values.
+func readDirectory64End(r io.ReaderAt, offset int64, d *directoryEnd) (err error) {
+ buf := make([]byte, directory64EndLen)
+ if _, err := r.ReadAt(buf, offset); err != nil {
+ return err
+ }
+
+ b := readBuf(buf)
+ if sig := b.uint32(); sig != directory64EndSignature {
+ return ErrFormat
+ }
+
+ b = b[12:] // skip dir size, version and version needed (uint64 + 2x uint16)
+ d.diskNbr = b.uint32() // number of this disk
+ d.dirDiskNbr = b.uint32() // number of the disk with the start of the central directory
+ d.dirRecordsThisDisk = b.uint64() // total number of entries in the central directory on this disk
+ d.directoryRecords = b.uint64() // total number of entries in the central directory
+ d.directorySize = b.uint64() // size of the central directory
+ d.directoryOffset = b.uint64() // offset of start of central directory with respect to the starting disk number
+
+ return nil
+}
+
+func findSignatureInBlock(b []byte) int {
+ for i := len(b) - directoryEndLen; i >= 0; i-- {
+ // defined from directoryEndSignature in struct.go
+ if b[i] == 'P' && b[i+1] == 'K' && b[i+2] == 0x05 && b[i+3] == 0x06 {
+ // n is length of comment
+ n := int(b[i+directoryEndLen-2]) | int(b[i+directoryEndLen-1])<<8
+ if n+directoryEndLen+i <= len(b) {
+ return i
+ }
+ }
+ }
+ return -1
+}
+
+type readBuf []byte
+
+func (b *readBuf) uint16() uint16 {
+ v := binary.LittleEndian.Uint16(*b)
+ *b = (*b)[2:]
+ return v
+}
+
+func (b *readBuf) uint32() uint32 {
+ v := binary.LittleEndian.Uint32(*b)
+ *b = (*b)[4:]
+ return v
+}
+
+func (b *readBuf) uint64() uint64 {
+ v := binary.LittleEndian.Uint64(*b)
+ *b = (*b)[8:]
+ return v
+}
diff --git a/src/archive/zip/reader_test.go b/src/archive/zip/reader_test.go
new file mode 100644
index 000000000..29d0652dc
--- /dev/null
+++ b/src/archive/zip/reader_test.go
@@ -0,0 +1,533 @@
+// Copyright 2010 The Go Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+package zip
+
+import (
+ "bytes"
+ "encoding/binary"
+ "encoding/hex"
+ "io"
+ "io/ioutil"
+ "os"
+ "path/filepath"
+ "regexp"
+ "strings"
+ "testing"
+ "time"
+)
+
+type ZipTest struct {
+ Name string
+ Source func() (r io.ReaderAt, size int64) // if non-nil, used instead of testdata/<Name> file
+ Comment string
+ File []ZipTestFile
+ Error error // the error that Opening this file should return
+}
+
+type ZipTestFile struct {
+ Name string
+ Content []byte // if blank, will attempt to compare against File
+ ContentErr error
+ File string // name of file to compare to (relative to testdata/)
+ Mtime string // modified time in format "mm-dd-yy hh:mm:ss"
+ Mode os.FileMode
+}
+
+// Caution: The Mtime values found for the test files should correspond to
+// the values listed with unzip -l <zipfile>. However, the values
+// listed by unzip appear to be off by some hours. When creating
+// fresh test files and testing them, this issue is not present.
+// The test files were created in Sydney, so there might be a time
+// zone issue. The time zone information does have to be encoded
+// somewhere, because otherwise unzip -l could not provide a different
+// time from what the archive/zip package provides, but there appears
+// to be no documentation about this.
+
+var tests = []ZipTest{
+ {
+ Name: "test.zip",
+ Comment: "This is a zipfile comment.",
+ File: []ZipTestFile{
+ {
+ Name: "test.txt",
+ Content: []byte("This is a test text file.\n"),
+ Mtime: "09-05-10 12:12:02",
+ Mode: 0644,
+ },
+ {
+ Name: "gophercolor16x16.png",
+ File: "gophercolor16x16.png",
+ Mtime: "09-05-10 15:52:58",
+ Mode: 0644,
+ },
+ },
+ },
+ {
+ Name: "test-trailing-junk.zip",
+ Comment: "This is a zipfile comment.",
+ File: []ZipTestFile{
+ {
+ Name: "test.txt",
+ Content: []byte("This is a test text file.\n"),
+ Mtime: "09-05-10 12:12:02",
+ Mode: 0644,
+ },
+ {
+ Name: "gophercolor16x16.png",
+ File: "gophercolor16x16.png",
+ Mtime: "09-05-10 15:52:58",
+ Mode: 0644,
+ },
+ },
+ },
+ {
+ Name: "r.zip",
+ Source: returnRecursiveZip,
+ File: []ZipTestFile{
+ {
+ Name: "r/r.zip",
+ Content: rZipBytes(),
+ Mtime: "03-04-10 00:24:16",
+ Mode: 0666,
+ },
+ },
+ },
+ {
+ Name: "symlink.zip",
+ File: []ZipTestFile{
+ {
+ Name: "symlink",
+ Content: []byte("../target"),
+ Mode: 0777 | os.ModeSymlink,
+ },
+ },
+ },
+ {
+ Name: "readme.zip",
+ },
+ {
+ Name: "readme.notzip",
+ Error: ErrFormat,
+ },
+ {
+ Name: "dd.zip",
+ File: []ZipTestFile{
+ {
+ Name: "filename",
+ Content: []byte("This is a test textfile.\n"),
+ Mtime: "02-02-11 13:06:20",
+ Mode: 0666,
+ },
+ },
+ },
+ {
+ // created in windows XP file manager.
+ Name: "winxp.zip",
+ File: crossPlatform,
+ },
+ {
+ // created by Zip 3.0 under Linux
+ Name: "unix.zip",
+ File: crossPlatform,
+ },
+ {
+ // created by Go, before we wrote the "optional" data
+ // descriptor signatures (which are required by OS X)
+ Name: "go-no-datadesc-sig.zip",
+ File: []ZipTestFile{
+ {
+ Name: "foo.txt",
+ Content: []byte("foo\n"),
+ Mtime: "03-08-12 16:59:10",
+ Mode: 0644,
+ },
+ {
+ Name: "bar.txt",
+ Content: []byte("bar\n"),
+ Mtime: "03-08-12 16:59:12",
+ Mode: 0644,
+ },
+ },
+ },
+ {
+ // created by Go, after we wrote the "optional" data
+ // descriptor signatures (which are required by OS X)
+ Name: "go-with-datadesc-sig.zip",
+ File: []ZipTestFile{
+ {
+ Name: "foo.txt",
+ Content: []byte("foo\n"),
+ Mode: 0666,
+ },
+ {
+ Name: "bar.txt",
+ Content: []byte("bar\n"),
+ Mode: 0666,
+ },
+ },
+ },
+ {
+ Name: "Bad-CRC32-in-data-descriptor",
+ Source: returnCorruptCRC32Zip,
+ File: []ZipTestFile{
+ {
+ Name: "foo.txt",
+ Content: []byte("foo\n"),
+ Mode: 0666,
+ ContentErr: ErrChecksum,
+ },
+ {
+ Name: "bar.txt",
+ Content: []byte("bar\n"),
+ Mode: 0666,
+ },
+ },
+ },
+ // Tests that we verify (and accept valid) crc32s on files
+ // with crc32s in their file header (not in data descriptors)
+ {
+ Name: "crc32-not-streamed.zip",
+ File: []ZipTestFile{
+ {
+ Name: "foo.txt",
+ Content: []byte("foo\n"),
+ Mtime: "03-08-12 16:59:10",
+ Mode: 0644,
+ },
+ {
+ Name: "bar.txt",
+ Content: []byte("bar\n"),
+ Mtime: "03-08-12 16:59:12",
+ Mode: 0644,
+ },
+ },
+ },
+ // Tests that we verify (and reject invalid) crc32s on files
+ // with crc32s in their file header (not in data descriptors)
+ {
+ Name: "crc32-not-streamed.zip",
+ Source: returnCorruptNotStreamedZip,
+ File: []ZipTestFile{
+ {
+ Name: "foo.txt",
+ Content: []byte("foo\n"),
+ Mtime: "03-08-12 16:59:10",
+ Mode: 0644,
+ ContentErr: ErrChecksum,
+ },
+ {
+ Name: "bar.txt",
+ Content: []byte("bar\n"),
+ Mtime: "03-08-12 16:59:12",
+ Mode: 0644,
+ },
+ },
+ },
+ {
+ Name: "zip64.zip",
+ File: []ZipTestFile{
+ {
+ Name: "README",
+ Content: []byte("This small file is in ZIP64 format.\n"),
+ Mtime: "08-10-12 14:33:32",
+ Mode: 0644,
+ },
+ },
+ },
+ // Another zip64 file with different Extras fields. (golang.org/issue/7069)
+ {
+ Name: "zip64-2.zip",
+ File: []ZipTestFile{
+ {
+ Name: "README",
+ Content: []byte("This small file is in ZIP64 format.\n"),
+ Mtime: "08-10-12 14:33:32",
+ Mode: 0644,
+ },
+ },
+ },
+}
+
+var crossPlatform = []ZipTestFile{
+ {
+ Name: "hello",
+ Content: []byte("world \r\n"),
+ Mode: 0666,
+ },
+ {
+ Name: "dir/bar",
+ Content: []byte("foo \r\n"),
+ Mode: 0666,
+ },
+ {
+ Name: "dir/empty/",
+ Content: []byte{},
+ Mode: os.ModeDir | 0777,
+ },
+ {
+ Name: "readonly",
+ Content: []byte("important \r\n"),
+ Mode: 0444,
+ },
+}
+
+func TestReader(t *testing.T) {
+ for _, zt := range tests {
+ readTestZip(t, zt)
+ }
+}
+
+func readTestZip(t *testing.T, zt ZipTest) {
+ var z *Reader
+ var err error
+ if zt.Source != nil {
+ rat, size := zt.Source()
+ z, err = NewReader(rat, size)
+ } else {
+ var rc *ReadCloser
+ rc, err = OpenReader(filepath.Join("testdata", zt.Name))
+ if err == nil {
+ defer rc.Close()
+ z = &rc.Reader
+ }
+ }
+ if err != zt.Error {
+ t.Errorf("%s: error=%v, want %v", zt.Name, err, zt.Error)
+ return
+ }
+
+ // bail if file is not zip
+ if err == ErrFormat {
+ return
+ }
+
+ // bail here if no Files expected to be tested
+ // (there may actually be files in the zip, but we don't care)
+ if zt.File == nil {
+ return
+ }
+
+ if z.Comment != zt.Comment {
+ t.Errorf("%s: comment=%q, want %q", zt.Name, z.Comment, zt.Comment)
+ }
+ if len(z.File) != len(zt.File) {
+ t.Fatalf("%s: file count=%d, want %d", zt.Name, len(z.File), len(zt.File))
+ }
+
+ // test read of each file
+ for i, ft := range zt.File {
+ readTestFile(t, zt, ft, z.File[i])
+ }
+
+ // test simultaneous reads
+ n := 0
+ done := make(chan bool)
+ for i := 0; i < 5; i++ {
+ for j, ft := range zt.File {
+ go func(j int, ft ZipTestFile) {
+ readTestFile(t, zt, ft, z.File[j])
+ done <- true
+ }(j, ft)
+ n++
+ }
+ }
+ for ; n > 0; n-- {
+ <-done
+ }
+}
+
+func readTestFile(t *testing.T, zt ZipTest, ft ZipTestFile, f *File) {
+ if f.Name != ft.Name {
+ t.Errorf("%s: name=%q, want %q", zt.Name, f.Name, ft.Name)
+ }
+
+ if ft.Mtime != "" {
+ mtime, err := time.Parse("01-02-06 15:04:05", ft.Mtime)
+ if err != nil {
+ t.Error(err)
+ return
+ }
+ if ft := f.ModTime(); !ft.Equal(mtime) {
+ t.Errorf("%s: %s: mtime=%s, want %s", zt.Name, f.Name, ft, mtime)
+ }
+ }
+
+ testFileMode(t, zt.Name, f, ft.Mode)
+
+ var b bytes.Buffer
+ r, err := f.Open()
+ if err != nil {
+ t.Errorf("%s: %v", zt.Name, err)
+ return
+ }
+
+ _, err = io.Copy(&b, r)
+ if err != ft.ContentErr {
+ t.Errorf("%s: copying contents: %v (want %v)", zt.Name, err, ft.ContentErr)
+ }
+ if err != nil {
+ return
+ }
+ r.Close()
+
+ size := uint64(f.UncompressedSize)
+ if size == uint32max {
+ size = f.UncompressedSize64
+ }
+ if g := uint64(b.Len()); g != size {
+ t.Errorf("%v: read %v bytes but f.UncompressedSize == %v", f.Name, g, size)
+ }
+
+ var c []byte
+ if ft.Content != nil {
+ c = ft.Content
+ } else if c, err = ioutil.ReadFile("testdata/" + ft.File); err != nil {
+ t.Error(err)
+ return
+ }
+
+ if b.Len() != len(c) {
+ t.Errorf("%s: len=%d, want %d", f.Name, b.Len(), len(c))
+ return
+ }
+
+ for i, b := range b.Bytes() {
+ if b != c[i] {
+ t.Errorf("%s: content[%d]=%q want %q", f.Name, i, b, c[i])
+ return
+ }
+ }
+}
+
+func testFileMode(t *testing.T, zipName string, f *File, want os.FileMode) {
+ mode := f.Mode()
+ if want == 0 {
+ t.Errorf("%s: %s mode: got %v, want none", zipName, f.Name, mode)
+ } else if mode != want {
+ t.Errorf("%s: %s mode: want %v, got %v", zipName, f.Name, want, mode)
+ }
+}
+
+func TestInvalidFiles(t *testing.T) {
+ const size = 1024 * 70 // 70kb
+ b := make([]byte, size)
+
+ // zeroes
+ _, err := NewReader(bytes.NewReader(b), size)
+ if err != ErrFormat {
+ t.Errorf("zeroes: error=%v, want %v", err, ErrFormat)
+ }
+
+ // repeated directoryEndSignatures
+ sig := make([]byte, 4)
+ binary.LittleEndian.PutUint32(sig, directoryEndSignature)
+ for i := 0; i < size-4; i += 4 {
+ copy(b[i:i+4], sig)
+ }
+ _, err = NewReader(bytes.NewReader(b), size)
+ if err != ErrFormat {
+ t.Errorf("sigs: error=%v, want %v", err, ErrFormat)
+ }
+}
+
+func messWith(fileName string, corrupter func(b []byte)) (r io.ReaderAt, size int64) {
+ data, err := ioutil.ReadFile(filepath.Join("testdata", fileName))
+ if err != nil {
+ panic("Error reading " + fileName + ": " + err.Error())
+ }
+ corrupter(data)
+ return bytes.NewReader(data), int64(len(data))
+}
+
+func returnCorruptCRC32Zip() (r io.ReaderAt, size int64) {
+ return messWith("go-with-datadesc-sig.zip", func(b []byte) {
+ // Corrupt one of the CRC32s in the data descriptor:
+ b[0x2d]++
+ })
+}
+
+func returnCorruptNotStreamedZip() (r io.ReaderAt, size int64) {
+ return messWith("crc32-not-streamed.zip", func(b []byte) {
+ // Corrupt foo.txt's final crc32 byte, in both
+ // the file header and TOC. (0x7e -> 0x7f)
+ b[0x11]++
+ b[0x9d]++
+
+ // TODO(bradfitz): add a new test that only corrupts
+ // one of these values, and verify that that's also an
+ // error. Currently, the reader code doesn't verify the
+ // fileheader and TOC's crc32 match if they're both
+ // non-zero and only the second line above, the TOC,
+ // is what matters.
+ })
+}
+
+// rZipBytes returns the bytes of a recursive zip file, without
+// putting it on disk and triggering certain virus scanners.
+func rZipBytes() []byte {
+ s := `
+0000000 50 4b 03 04 14 00 00 00 08 00 08 03 64 3c f9 f4
+0000010 89 64 48 01 00 00 b8 01 00 00 07 00 00 00 72 2f
+0000020 72 2e 7a 69 70 00 25 00 da ff 50 4b 03 04 14 00
+0000030 00 00 08 00 08 03 64 3c f9 f4 89 64 48 01 00 00
+0000040 b8 01 00 00 07 00 00 00 72 2f 72 2e 7a 69 70 00
+0000050 2f 00 d0 ff 00 25 00 da ff 50 4b 03 04 14 00 00
+0000060 00 08 00 08 03 64 3c f9 f4 89 64 48 01 00 00 b8
+0000070 01 00 00 07 00 00 00 72 2f 72 2e 7a 69 70 00 2f
+0000080 00 d0 ff c2 54 8e 57 39 00 05 00 fa ff c2 54 8e
+0000090 57 39 00 05 00 fa ff 00 05 00 fa ff 00 14 00 eb
+00000a0 ff c2 54 8e 57 39 00 05 00 fa ff 00 05 00 fa ff
+00000b0 00 14 00 eb ff 42 88 21 c4 00 00 14 00 eb ff 42
+00000c0 88 21 c4 00 00 14 00 eb ff 42 88 21 c4 00 00 14
+00000d0 00 eb ff 42 88 21 c4 00 00 14 00 eb ff 42 88 21
+00000e0 c4 00 00 00 00 ff ff 00 00 00 ff ff 00 34 00 cb
+00000f0 ff 42 88 21 c4 00 00 00 00 ff ff 00 00 00 ff ff
+0000100 00 34 00 cb ff 42 e8 21 5e 0f 00 00 00 ff ff 0a
+0000110 f0 66 64 12 61 c0 15 dc e8 a0 48 bf 48 af 2a b3
+0000120 20 c0 9b 95 0d c4 67 04 42 53 06 06 06 40 00 06
+0000130 00 f9 ff 6d 01 00 00 00 00 42 e8 21 5e 0f 00 00
+0000140 00 ff ff 0a f0 66 64 12 61 c0 15 dc e8 a0 48 bf
+0000150 48 af 2a b3 20 c0 9b 95 0d c4 67 04 42 53 06 06
+0000160 06 40 00 06 00 f9 ff 6d 01 00 00 00 00 50 4b 01
+0000170 02 14 00 14 00 00 00 08 00 08 03 64 3c f9 f4 89
+0000180 64 48 01 00 00 b8 01 00 00 07 00 00 00 00 00 00
+0000190 00 00 00 00 00 00 00 00 00 00 00 72 2f 72 2e 7a
+00001a0 69 70 50 4b 05 06 00 00 00 00 01 00 01 00 35 00
+00001b0 00 00 6d 01 00 00 00 00`
+ s = regexp.MustCompile(`[0-9a-f]{7}`).ReplaceAllString(s, "")
+ s = regexp.MustCompile(`\s+`).ReplaceAllString(s, "")
+ b, err := hex.DecodeString(s)
+ if err != nil {
+ panic(err)
+ }
+ return b
+}
+
+func returnRecursiveZip() (r io.ReaderAt, size int64) {
+ b := rZipBytes()
+ return bytes.NewReader(b), int64(len(b))
+}
+
+func TestIssue8186(t *testing.T) {
+ // Directory headers & data found in the TOC of a JAR file.
+ dirEnts := []string{
+ "PK\x01\x02\n\x00\n\x00\x00\b\x00\x004\x9d3?\xaa\x1b\x06\xf0\x81\x02\x00\x00\x81\x02\x00\x00-\x00\x05\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00res/drawable-xhdpi-v4/ic_actionbar_accept.png\xfe\xca\x00\x00\x00",
+ "PK\x01\x02\n\x00\n\x00\x00\b\x00\x004\x9d3?\x90K\x89\xc7t\n\x00\x00t\n\x00\x00\x0e\x00\x03\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\xd1\x02\x00\x00resources.arsc\x00\x00\x00",
+ "PK\x01\x02\x14\x00\x14\x00\b\b\b\x004\x9d3?\xff$\x18\xed3\x03\x00\x00\xb4\b\x00\x00\x13\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00t\r\x00\x00AndroidManifest.xml",
+ "PK\x01\x02\x14\x00\x14\x00\b\b\b\x004\x9d3?\x14\xc5K\xab\x192\x02\x00\xc8\xcd\x04\x00\v\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\xe8\x10\x00\x00classes.dex",
+ "PK\x01\x02\x14\x00\x14\x00\b\b\b\x004\x9d3?E\x96\nD\xac\x01\x00\x00P\x03\x00\x00&\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00:C\x02\x00res/layout/actionbar_set_wallpaper.xml",
+ "PK\x01\x02\x14\x00\x14\x00\b\b\b\x004\x9d3?Ļ\x14\xe3\xd8\x01\x00\x00\xd8\x03\x00\x00 \x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00:E\x02\x00res/layout/wallpaper_cropper.xml",
+ "PK\x01\x02\x14\x00\x14\x00\b\b\b\x004\x9d3?}\xc1\x15\x9eZ\x01\x00\x00!\x02\x00\x00\x14\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00`G\x02\x00META-INF/MANIFEST.MF",
+ "PK\x01\x02\x14\x00\x14\x00\b\b\b\x004\x9d3?\xe6\x98Ьo\x01\x00\x00\x84\x02\x00\x00\x10\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\xfcH\x02\x00META-INF/CERT.SF",
+ "PK\x01\x02\x14\x00\x14\x00\b\b\b\x004\x9d3?\xbfP\x96b\x86\x04\x00\x00\xb2\x06\x00\x00\x11\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\xa9J\x02\x00META-INF/CERT.RSA",
+ }
+ for i, s := range dirEnts {
+ var f File
+ err := readDirectoryHeader(&f, strings.NewReader(s))
+ if err != nil {
+ t.Errorf("error reading #%d: %v", i, err)
+ }
+ }
+}
diff --git a/src/archive/zip/register.go b/src/archive/zip/register.go
new file mode 100644
index 000000000..4211ec7af
--- /dev/null
+++ b/src/archive/zip/register.go
@@ -0,0 +1,110 @@
+// Copyright 2010 The Go Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+package zip
+
+import (
+ "compress/flate"
+ "errors"
+ "io"
+ "io/ioutil"
+ "sync"
+)
+
+// A Compressor returns a compressing writer, writing to the
+// provided writer. On Close, any pending data should be flushed.
+type Compressor func(io.Writer) (io.WriteCloser, error)
+
+// Decompressor is a function that wraps a Reader with a decompressing Reader.
+// The decompressed ReadCloser is returned to callers who open files from
+// within the archive. These callers are responsible for closing this reader
+// when they're finished reading.
+type Decompressor func(io.Reader) io.ReadCloser
+
+var flateWriterPool sync.Pool
+
+func newFlateWriter(w io.Writer) io.WriteCloser {
+ fw, ok := flateWriterPool.Get().(*flate.Writer)
+ if ok {
+ fw.Reset(w)
+ } else {
+ fw, _ = flate.NewWriter(w, 5)
+ }
+ return &pooledFlateWriter{fw: fw}
+}
+
+type pooledFlateWriter struct {
+ mu sync.Mutex // guards Close and Write
+ fw *flate.Writer
+}
+
+func (w *pooledFlateWriter) Write(p []byte) (n int, err error) {
+ w.mu.Lock()
+ defer w.mu.Unlock()
+ if w.fw == nil {
+ return 0, errors.New("Write after Close")
+ }
+ return w.fw.Write(p)
+}
+
+func (w *pooledFlateWriter) Close() error {
+ w.mu.Lock()
+ defer w.mu.Unlock()
+ var err error
+ if w.fw != nil {
+ err = w.fw.Close()
+ flateWriterPool.Put(w.fw)
+ w.fw = nil
+ }
+ return err
+}
+
+var (
+ mu sync.RWMutex // guards compressor and decompressor maps
+
+ compressors = map[uint16]Compressor{
+ Store: func(w io.Writer) (io.WriteCloser, error) { return &nopCloser{w}, nil },
+ Deflate: func(w io.Writer) (io.WriteCloser, error) { return newFlateWriter(w), nil },
+ }
+
+ decompressors = map[uint16]Decompressor{
+ Store: ioutil.NopCloser,
+ Deflate: flate.NewReader,
+ }
+)
+
+// RegisterDecompressor allows custom decompressors for a specified method ID.
+func RegisterDecompressor(method uint16, d Decompressor) {
+ mu.Lock()
+ defer mu.Unlock()
+
+ if _, ok := decompressors[method]; ok {
+ panic("decompressor already registered")
+ }
+ decompressors[method] = d
+}
+
+// RegisterCompressor registers custom compressors for a specified method ID.
+// The common methods Store and Deflate are built in.
+func RegisterCompressor(method uint16, comp Compressor) {
+ mu.Lock()
+ defer mu.Unlock()
+
+ if _, ok := compressors[method]; ok {
+ panic("compressor already registered")
+ }
+ compressors[method] = comp
+}
+
+func compressor(method uint16) Compressor {
+ mu.RLock()
+ defer mu.RUnlock()
+ return compressors[method]
+}
+
+func decompressor(method uint16) Decompressor {
+ mu.RLock()
+ defer mu.RUnlock()
+ return decompressors[method]
+}
diff --git a/src/archive/zip/struct.go b/src/archive/zip/struct.go
new file mode 100644
index 000000000..cb28e8324
--- /dev/null
+++ b/src/archive/zip/struct.go
@@ -0,0 +1,313 @@
+// Copyright 2010 The Go Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+/*
+Package zip provides support for reading and writing ZIP archives.
+
+See: http://www.pkware.com/documents/casestudies/APPNOTE.TXT
+
+This package does not support disk spanning.
+
+A note about ZIP64:
+
+To be backwards compatible the FileHeader has both 32 and 64 bit Size
+fields. The 64 bit fields will always contain the correct value and
+for normal archives both fields will be the same. For files requiring
+the ZIP64 format the 32 bit fields will be 0xffffffff and the 64 bit
+fields must be used instead.
+*/
+package zip
+
+import (
+ "os"
+ "path"
+ "time"
+)
+
+// Compression methods.
+const (
+ Store uint16 = 0
+ Deflate uint16 = 8
+)
+
+const (
+ fileHeaderSignature = 0x04034b50
+ directoryHeaderSignature = 0x02014b50
+ directoryEndSignature = 0x06054b50
+ directory64LocSignature = 0x07064b50
+ directory64EndSignature = 0x06064b50
+ dataDescriptorSignature = 0x08074b50 // de-facto standard; required by OS X Finder
+ fileHeaderLen = 30 // + filename + extra
+ directoryHeaderLen = 46 // + filename + extra + comment
+ directoryEndLen = 22 // + comment
+ dataDescriptorLen = 16 // four uint32: descriptor signature, crc32, compressed size, size
+ dataDescriptor64Len = 24 // descriptor with 8 byte sizes
+ directory64LocLen = 20 //
+ directory64EndLen = 56 // + extra
+
+ // Constants for the first byte in CreatorVersion
+ creatorFAT = 0
+ creatorUnix = 3
+ creatorNTFS = 11
+ creatorVFAT = 14
+ creatorMacOSX = 19
+
+ // version numbers
+ zipVersion20 = 20 // 2.0
+ zipVersion45 = 45 // 4.5 (reads and writes zip64 archives)
+
+ // limits for non zip64 files
+ uint16max = (1 << 16) - 1
+ uint32max = (1 << 32) - 1
+
+ // extra header id's
+ zip64ExtraId = 0x0001 // zip64 Extended Information Extra Field
+)
+
+// FileHeader describes a file within a zip file.
+// See the zip spec for details.
+type FileHeader struct {
+ // Name is the name of the file.
+ // It must be a relative path: it must not start with a drive
+ // letter (e.g. C:) or leading slash, and only forward slashes
+ // are allowed.
+ Name string
+
+ CreatorVersion uint16
+ ReaderVersion uint16
+ Flags uint16
+ Method uint16
+ ModifiedTime uint16 // MS-DOS time
+ ModifiedDate uint16 // MS-DOS date
+ CRC32 uint32
+ CompressedSize uint32 // deprecated; use CompressedSize64
+ UncompressedSize uint32 // deprecated; use UncompressedSize64
+ CompressedSize64 uint64
+ UncompressedSize64 uint64
+ Extra []byte
+ ExternalAttrs uint32 // Meaning depends on CreatorVersion
+ Comment string
+}
+
+// FileInfo returns an os.FileInfo for the FileHeader.
+func (h *FileHeader) FileInfo() os.FileInfo {
+ return headerFileInfo{h}
+}
+
+// headerFileInfo implements os.FileInfo.
+type headerFileInfo struct {
+ fh *FileHeader
+}
+
+func (fi headerFileInfo) Name() string { return path.Base(fi.fh.Name) }
+func (fi headerFileInfo) Size() int64 {
+ if fi.fh.UncompressedSize64 > 0 {
+ return int64(fi.fh.UncompressedSize64)
+ }
+ return int64(fi.fh.UncompressedSize)
+}
+func (fi headerFileInfo) IsDir() bool { return fi.Mode().IsDir() }
+func (fi headerFileInfo) ModTime() time.Time { return fi.fh.ModTime() }
+func (fi headerFileInfo) Mode() os.FileMode { return fi.fh.Mode() }
+func (fi headerFileInfo) Sys() interface{} { return fi.fh }
+
+// FileInfoHeader creates a partially-populated FileHeader from an
+// os.FileInfo.
+// Because os.FileInfo's Name method returns only the base name of
+// the file it describes, it may be necessary to modify the Name field
+// of the returned header to provide the full path name of the file.
+func FileInfoHeader(fi os.FileInfo) (*FileHeader, error) {
+ size := fi.Size()
+ fh := &FileHeader{
+ Name: fi.Name(),
+ UncompressedSize64: uint64(size),
+ }
+ fh.SetModTime(fi.ModTime())
+ fh.SetMode(fi.Mode())
+ if fh.UncompressedSize64 > uint32max {
+ fh.UncompressedSize = uint32max
+ } else {
+ fh.UncompressedSize = uint32(fh.UncompressedSize64)
+ }
+ return fh, nil
+}
+
+type directoryEnd struct {
+ diskNbr uint32 // unused
+ dirDiskNbr uint32 // unused
+ dirRecordsThisDisk uint64 // unused
+ directoryRecords uint64
+ directorySize uint64
+ directoryOffset uint64 // relative to file
+ commentLen uint16
+ comment string
+}
+
+// msDosTimeToTime converts an MS-DOS date and time into a time.Time.
+// The resolution is 2s.
+// See: http://msdn.microsoft.com/en-us/library/ms724247(v=VS.85).aspx
+func msDosTimeToTime(dosDate, dosTime uint16) time.Time {
+ return time.Date(
+ // date bits 0-4: day of month; 5-8: month; 9-15: years since 1980
+ int(dosDate>>9+1980),
+ time.Month(dosDate>>5&0xf),
+ int(dosDate&0x1f),
+
+ // time bits 0-4: second/2; 5-10: minute; 11-15: hour
+ int(dosTime>>11),
+ int(dosTime>>5&0x3f),
+ int(dosTime&0x1f*2),
+ 0, // nanoseconds
+
+ time.UTC,
+ )
+}
+
+// timeToMsDosTime converts a time.Time to an MS-DOS date and time.
+// The resolution is 2s.
+// See: http://msdn.microsoft.com/en-us/library/ms724274(v=VS.85).aspx
+func timeToMsDosTime(t time.Time) (fDate uint16, fTime uint16) {
+ t = t.In(time.UTC)
+ fDate = uint16(t.Day() + int(t.Month())<<5 + (t.Year()-1980)<<9)
+ fTime = uint16(t.Second()/2 + t.Minute()<<5 + t.Hour()<<11)
+ return
+}
+
+// ModTime returns the modification time in UTC.
+// The resolution is 2s.
+func (h *FileHeader) ModTime() time.Time {
+ return msDosTimeToTime(h.ModifiedDate, h.ModifiedTime)
+}
+
+// SetModTime sets the ModifiedTime and ModifiedDate fields to the given time in UTC.
+// The resolution is 2s.
+func (h *FileHeader) SetModTime(t time.Time) {
+ h.ModifiedDate, h.ModifiedTime = timeToMsDosTime(t)
+}
+
+const (
+ // Unix constants. The specification doesn't mention them,
+ // but these seem to be the values agreed on by tools.
+ s_IFMT = 0xf000
+ s_IFSOCK = 0xc000
+ s_IFLNK = 0xa000
+ s_IFREG = 0x8000
+ s_IFBLK = 0x6000
+ s_IFDIR = 0x4000
+ s_IFCHR = 0x2000
+ s_IFIFO = 0x1000
+ s_ISUID = 0x800
+ s_ISGID = 0x400
+ s_ISVTX = 0x200
+
+ msdosDir = 0x10
+ msdosReadOnly = 0x01
+)
+
+// Mode returns the permission and mode bits for the FileHeader.
+func (h *FileHeader) Mode() (mode os.FileMode) {
+ switch h.CreatorVersion >> 8 {
+ case creatorUnix, creatorMacOSX:
+ mode = unixModeToFileMode(h.ExternalAttrs >> 16)
+ case creatorNTFS, creatorVFAT, creatorFAT:
+ mode = msdosModeToFileMode(h.ExternalAttrs)
+ }
+ if len(h.Name) > 0 && h.Name[len(h.Name)-1] == '/' {
+ mode |= os.ModeDir
+ }
+ return mode
+}
+
+// SetMode changes the permission and mode bits for the FileHeader.
+func (h *FileHeader) SetMode(mode os.FileMode) {
+ h.CreatorVersion = h.CreatorVersion&0xff | creatorUnix<<8
+ h.ExternalAttrs = fileModeToUnixMode(mode) << 16
+
+ // set MSDOS attributes too, as the original zip does.
+ if mode&os.ModeDir != 0 {
+ h.ExternalAttrs |= msdosDir
+ }
+ if mode&0200 == 0 {
+ h.ExternalAttrs |= msdosReadOnly
+ }
+}
+
+// isZip64 returns true if the file size exceeds the 32 bit limit
+func (fh *FileHeader) isZip64() bool {
+ return fh.CompressedSize64 > uint32max || fh.UncompressedSize64 > uint32max
+}
+
+func msdosModeToFileMode(m uint32) (mode os.FileMode) {
+ if m&msdosDir != 0 {
+ mode = os.ModeDir | 0777
+ } else {
+ mode = 0666
+ }
+ if m&msdosReadOnly != 0 {
+ mode &^= 0222
+ }
+ return mode
+}
+
+func fileModeToUnixMode(mode os.FileMode) uint32 {
+ var m uint32
+ switch mode & os.ModeType {
+ default:
+ m = s_IFREG
+ case os.ModeDir:
+ m = s_IFDIR
+ case os.ModeSymlink:
+ m = s_IFLNK
+ case os.ModeNamedPipe:
+ m = s_IFIFO
+ case os.ModeSocket:
+ m = s_IFSOCK
+ case os.ModeDevice:
+ if mode&os.ModeCharDevice != 0 {
+ m = s_IFCHR
+ } else {
+ m = s_IFBLK
+ }
+ }
+ if mode&os.ModeSetuid != 0 {
+ m |= s_ISUID
+ }
+ if mode&os.ModeSetgid != 0 {
+ m |= s_ISGID
+ }
+ if mode&os.ModeSticky != 0 {
+ m |= s_ISVTX
+ }
+ return m | uint32(mode&0777)
+}
+
+func unixModeToFileMode(m uint32) os.FileMode {
+ mode := os.FileMode(m & 0777)
+ switch m & s_IFMT {
+ case s_IFBLK:
+ mode |= os.ModeDevice
+ case s_IFCHR:
+ mode |= os.ModeDevice | os.ModeCharDevice
+ case s_IFDIR:
+ mode |= os.ModeDir
+ case s_IFIFO:
+ mode |= os.ModeNamedPipe
+ case s_IFLNK:
+ mode |= os.ModeSymlink
+ case s_IFREG:
+ // nothing to do
+ case s_IFSOCK:
+ mode |= os.ModeSocket
+ }
+ if m&s_ISGID != 0 {
+ mode |= os.ModeSetgid
+ }
+ if m&s_ISUID != 0 {
+ mode |= os.ModeSetuid
+ }
+ if m&s_ISVTX != 0 {
+ mode |= os.ModeSticky
+ }
+ return mode
+}
diff --git a/src/archive/zip/testdata/crc32-not-streamed.zip b/src/archive/zip/testdata/crc32-not-streamed.zip
new file mode 100644
index 000000000..f268d8873
--- /dev/null
+++ b/src/archive/zip/testdata/crc32-not-streamed.zip
Binary files differ
diff --git a/src/archive/zip/testdata/dd.zip b/src/archive/zip/testdata/dd.zip
new file mode 100644
index 000000000..e53378b0b
--- /dev/null
+++ b/src/archive/zip/testdata/dd.zip
Binary files differ
diff --git a/src/archive/zip/testdata/go-no-datadesc-sig.zip b/src/archive/zip/testdata/go-no-datadesc-sig.zip
new file mode 100644
index 000000000..c3d593f44
--- /dev/null
+++ b/src/archive/zip/testdata/go-no-datadesc-sig.zip
Binary files differ
diff --git a/src/archive/zip/testdata/go-with-datadesc-sig.zip b/src/archive/zip/testdata/go-with-datadesc-sig.zip
new file mode 100644
index 000000000..bcfe121bb
--- /dev/null
+++ b/src/archive/zip/testdata/go-with-datadesc-sig.zip
Binary files differ
diff --git a/src/archive/zip/testdata/gophercolor16x16.png b/src/archive/zip/testdata/gophercolor16x16.png
new file mode 100644
index 000000000..48854ff3b
--- /dev/null
+++ b/src/archive/zip/testdata/gophercolor16x16.png
Binary files differ
diff --git a/src/archive/zip/testdata/readme.notzip b/src/archive/zip/testdata/readme.notzip
new file mode 100644
index 000000000..06668c4c1
--- /dev/null
+++ b/src/archive/zip/testdata/readme.notzip
Binary files differ
diff --git a/src/archive/zip/testdata/readme.zip b/src/archive/zip/testdata/readme.zip
new file mode 100644
index 000000000..db3bb900e
--- /dev/null
+++ b/src/archive/zip/testdata/readme.zip
Binary files differ
diff --git a/src/archive/zip/testdata/symlink.zip b/src/archive/zip/testdata/symlink.zip
new file mode 100644
index 000000000..af846938c
--- /dev/null
+++ b/src/archive/zip/testdata/symlink.zip
Binary files differ
diff --git a/src/archive/zip/testdata/test-trailing-junk.zip b/src/archive/zip/testdata/test-trailing-junk.zip
new file mode 100644
index 000000000..42281b4e3
--- /dev/null
+++ b/src/archive/zip/testdata/test-trailing-junk.zip
Binary files differ
diff --git a/src/archive/zip/testdata/test.zip b/src/archive/zip/testdata/test.zip
new file mode 100644
index 000000000..03890c05d
--- /dev/null
+++ b/src/archive/zip/testdata/test.zip
Binary files differ
diff --git a/src/archive/zip/testdata/unix.zip b/src/archive/zip/testdata/unix.zip
new file mode 100644
index 000000000..ce1a981b2
--- /dev/null
+++ b/src/archive/zip/testdata/unix.zip
Binary files differ
diff --git a/src/archive/zip/testdata/winxp.zip b/src/archive/zip/testdata/winxp.zip
new file mode 100644
index 000000000..3919322f0
--- /dev/null
+++ b/src/archive/zip/testdata/winxp.zip
Binary files differ
diff --git a/src/archive/zip/testdata/zip64-2.zip b/src/archive/zip/testdata/zip64-2.zip
new file mode 100644
index 000000000..f844e3537
--- /dev/null
+++ b/src/archive/zip/testdata/zip64-2.zip
Binary files differ
diff --git a/src/archive/zip/testdata/zip64.zip b/src/archive/zip/testdata/zip64.zip
new file mode 100644
index 000000000..a2ee1fa33
--- /dev/null
+++ b/src/archive/zip/testdata/zip64.zip
Binary files differ
diff --git a/src/archive/zip/writer.go b/src/archive/zip/writer.go
new file mode 100644
index 000000000..170beec0e
--- /dev/null
+++ b/src/archive/zip/writer.go
@@ -0,0 +1,357 @@
+// Copyright 2011 The Go Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+package zip
+
+import (
+ "bufio"
+ "encoding/binary"
+ "errors"
+ "hash"
+ "hash/crc32"
+ "io"
+)
+
+// TODO(adg): support zip file comments
+// TODO(adg): support specifying deflate level
+
+// Writer implements a zip file writer.
+type Writer struct {
+ cw *countWriter
+ dir []*header
+ last *fileWriter
+ closed bool
+}
+
+type header struct {
+ *FileHeader
+ offset uint64
+}
+
+// NewWriter returns a new Writer writing a zip file to w.
+func NewWriter(w io.Writer) *Writer {
+ return &Writer{cw: &countWriter{w: bufio.NewWriter(w)}}
+}
+
+// Flush flushes any buffered data to the underlying writer.
+// Calling Flush is not normally necessary; calling Close is sufficient.
+func (w *Writer) Flush() error {
+ return w.cw.w.(*bufio.Writer).Flush()
+}
+
+// Close finishes writing the zip file by writing the central directory.
+// It does not (and can not) close the underlying writer.
+func (w *Writer) Close() error {
+ if w.last != nil && !w.last.closed {
+ if err := w.last.close(); err != nil {
+ return err
+ }
+ w.last = nil
+ }
+ if w.closed {
+ return errors.New("zip: writer closed twice")
+ }
+ w.closed = true
+
+ // write central directory
+ start := w.cw.count
+ for _, h := range w.dir {
+ var buf [directoryHeaderLen]byte
+ b := writeBuf(buf[:])
+ b.uint32(uint32(directoryHeaderSignature))
+ b.uint16(h.CreatorVersion)
+ b.uint16(h.ReaderVersion)
+ b.uint16(h.Flags)
+ b.uint16(h.Method)
+ b.uint16(h.ModifiedTime)
+ b.uint16(h.ModifiedDate)
+ b.uint32(h.CRC32)
+ if h.isZip64() || h.offset > uint32max {
+ // the file needs a zip64 header. store maxint in both
+ // 32 bit size fields (and offset later) to signal that the
+ // zip64 extra header should be used.
+ b.uint32(uint32max) // compressed size
+ b.uint32(uint32max) // uncompressed size
+
+ // append a zip64 extra block to Extra
+ var buf [28]byte // 2x uint16 + 3x uint64
+ eb := writeBuf(buf[:])
+ eb.uint16(zip64ExtraId)
+ eb.uint16(24) // size = 3x uint64
+ eb.uint64(h.UncompressedSize64)
+ eb.uint64(h.CompressedSize64)
+ eb.uint64(h.offset)
+ h.Extra = append(h.Extra, buf[:]...)
+ } else {
+ b.uint32(h.CompressedSize)
+ b.uint32(h.UncompressedSize)
+ }
+ b.uint16(uint16(len(h.Name)))
+ b.uint16(uint16(len(h.Extra)))
+ b.uint16(uint16(len(h.Comment)))
+ b = b[4:] // skip disk number start and internal file attr (2x uint16)
+ b.uint32(h.ExternalAttrs)
+ if h.offset > uint32max {
+ b.uint32(uint32max)
+ } else {
+ b.uint32(uint32(h.offset))
+ }
+ if _, err := w.cw.Write(buf[:]); err != nil {
+ return err
+ }
+ if _, err := io.WriteString(w.cw, h.Name); err != nil {
+ return err
+ }
+ if _, err := w.cw.Write(h.Extra); err != nil {
+ return err
+ }
+ if _, err := io.WriteString(w.cw, h.Comment); err != nil {
+ return err
+ }
+ }
+ end := w.cw.count
+
+ records := uint64(len(w.dir))
+ size := uint64(end - start)
+ offset := uint64(start)
+
+ if records > uint16max || size > uint32max || offset > uint32max {
+ var buf [directory64EndLen + directory64LocLen]byte
+ b := writeBuf(buf[:])
+
+ // zip64 end of central directory record
+ b.uint32(directory64EndSignature)
+ b.uint64(directory64EndLen)
+ b.uint16(zipVersion45) // version made by
+ b.uint16(zipVersion45) // version needed to extract
+ b.uint32(0) // number of this disk
+ b.uint32(0) // number of the disk with the start of the central directory
+ b.uint64(records) // total number of entries in the central directory on this disk
+ b.uint64(records) // total number of entries in the central directory
+ b.uint64(size) // size of the central directory
+ b.uint64(offset) // offset of start of central directory with respect to the starting disk number
+
+ // zip64 end of central directory locator
+ b.uint32(directory64LocSignature)
+ b.uint32(0) // number of the disk with the start of the zip64 end of central directory
+ b.uint64(uint64(end)) // relative offset of the zip64 end of central directory record
+ b.uint32(1) // total number of disks
+
+ if _, err := w.cw.Write(buf[:]); err != nil {
+ return err
+ }
+
+ // store max values in the regular end record to signal that
+ // that the zip64 values should be used instead
+ records = uint16max
+ size = uint32max
+ offset = uint32max
+ }
+
+ // write end record
+ var buf [directoryEndLen]byte
+ b := writeBuf(buf[:])
+ b.uint32(uint32(directoryEndSignature))
+ b = b[4:] // skip over disk number and first disk number (2x uint16)
+ b.uint16(uint16(records)) // number of entries this disk
+ b.uint16(uint16(records)) // number of entries total
+ b.uint32(uint32(size)) // size of directory
+ b.uint32(uint32(offset)) // start of directory
+ // skipped size of comment (always zero)
+ if _, err := w.cw.Write(buf[:]); err != nil {
+ return err
+ }
+
+ return w.cw.w.(*bufio.Writer).Flush()
+}
+
+// Create adds a file to the zip file using the provided name.
+// It returns a Writer to which the file contents should be written.
+// The name must be a relative path: it must not start with a drive
+// letter (e.g. C:) or leading slash, and only forward slashes are
+// allowed.
+// The file's contents must be written to the io.Writer before the next
+// call to Create, CreateHeader, or Close.
+func (w *Writer) Create(name string) (io.Writer, error) {
+ header := &FileHeader{
+ Name: name,
+ Method: Deflate,
+ }
+ return w.CreateHeader(header)
+}
+
+// CreateHeader adds a file to the zip file using the provided FileHeader
+// for the file metadata.
+// It returns a Writer to which the file contents should be written.
+// The file's contents must be written to the io.Writer before the next
+// call to Create, CreateHeader, or Close.
+func (w *Writer) CreateHeader(fh *FileHeader) (io.Writer, error) {
+ if w.last != nil && !w.last.closed {
+ if err := w.last.close(); err != nil {
+ return nil, err
+ }
+ }
+
+ fh.Flags |= 0x8 // we will write a data descriptor
+
+ fh.CreatorVersion = fh.CreatorVersion&0xff00 | zipVersion20 // preserve compatibility byte
+ fh.ReaderVersion = zipVersion20
+
+ fw := &fileWriter{
+ zipw: w.cw,
+ compCount: &countWriter{w: w.cw},
+ crc32: crc32.NewIEEE(),
+ }
+ comp := compressor(fh.Method)
+ if comp == nil {
+ return nil, ErrAlgorithm
+ }
+ var err error
+ fw.comp, err = comp(fw.compCount)
+ if err != nil {
+ return nil, err
+ }
+ fw.rawCount = &countWriter{w: fw.comp}
+
+ h := &header{
+ FileHeader: fh,
+ offset: uint64(w.cw.count),
+ }
+ w.dir = append(w.dir, h)
+ fw.header = h
+
+ if err := writeHeader(w.cw, fh); err != nil {
+ return nil, err
+ }
+
+ w.last = fw
+ return fw, nil
+}
+
+func writeHeader(w io.Writer, h *FileHeader) error {
+ var buf [fileHeaderLen]byte
+ b := writeBuf(buf[:])
+ b.uint32(uint32(fileHeaderSignature))
+ b.uint16(h.ReaderVersion)
+ b.uint16(h.Flags)
+ b.uint16(h.Method)
+ b.uint16(h.ModifiedTime)
+ b.uint16(h.ModifiedDate)
+ b.uint32(0) // since we are writing a data descriptor crc32,
+ b.uint32(0) // compressed size,
+ b.uint32(0) // and uncompressed size should be zero
+ b.uint16(uint16(len(h.Name)))
+ b.uint16(uint16(len(h.Extra)))
+ if _, err := w.Write(buf[:]); err != nil {
+ return err
+ }
+ if _, err := io.WriteString(w, h.Name); err != nil {
+ return err
+ }
+ _, err := w.Write(h.Extra)
+ return err
+}
+
+type fileWriter struct {
+ *header
+ zipw io.Writer
+ rawCount *countWriter
+ comp io.WriteCloser
+ compCount *countWriter
+ crc32 hash.Hash32
+ closed bool
+}
+
+func (w *fileWriter) Write(p []byte) (int, error) {
+ if w.closed {
+ return 0, errors.New("zip: write to closed file")
+ }
+ w.crc32.Write(p)
+ return w.rawCount.Write(p)
+}
+
+func (w *fileWriter) close() error {
+ if w.closed {
+ return errors.New("zip: file closed twice")
+ }
+ w.closed = true
+ if err := w.comp.Close(); err != nil {
+ return err
+ }
+
+ // update FileHeader
+ fh := w.header.FileHeader
+ fh.CRC32 = w.crc32.Sum32()
+ fh.CompressedSize64 = uint64(w.compCount.count)
+ fh.UncompressedSize64 = uint64(w.rawCount.count)
+
+ if fh.isZip64() {
+ fh.CompressedSize = uint32max
+ fh.UncompressedSize = uint32max
+ fh.ReaderVersion = zipVersion45 // requires 4.5 - File uses ZIP64 format extensions
+ } else {
+ fh.CompressedSize = uint32(fh.CompressedSize64)
+ fh.UncompressedSize = uint32(fh.UncompressedSize64)
+ }
+
+ // Write data descriptor. This is more complicated than one would
+ // think, see e.g. comments in zipfile.c:putextended() and
+ // http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=7073588.
+ // The approach here is to write 8 byte sizes if needed without
+ // adding a zip64 extra in the local header (too late anyway).
+ var buf []byte
+ if fh.isZip64() {
+ buf = make([]byte, dataDescriptor64Len)
+ } else {
+ buf = make([]byte, dataDescriptorLen)
+ }
+ b := writeBuf(buf)
+ b.uint32(dataDescriptorSignature) // de-facto standard, required by OS X
+ b.uint32(fh.CRC32)
+ if fh.isZip64() {
+ b.uint64(fh.CompressedSize64)
+ b.uint64(fh.UncompressedSize64)
+ } else {
+ b.uint32(fh.CompressedSize)
+ b.uint32(fh.UncompressedSize)
+ }
+ _, err := w.zipw.Write(buf)
+ return err
+}
+
+type countWriter struct {
+ w io.Writer
+ count int64
+}
+
+func (w *countWriter) Write(p []byte) (int, error) {
+ n, err := w.w.Write(p)
+ w.count += int64(n)
+ return n, err
+}
+
+type nopCloser struct {
+ io.Writer
+}
+
+func (w nopCloser) Close() error {
+ return nil
+}
+
+type writeBuf []byte
+
+func (b *writeBuf) uint16(v uint16) {
+ binary.LittleEndian.PutUint16(*b, v)
+ *b = (*b)[2:]
+}
+
+func (b *writeBuf) uint32(v uint32) {
+ binary.LittleEndian.PutUint32(*b, v)
+ *b = (*b)[4:]
+}
+
+func (b *writeBuf) uint64(v uint64) {
+ binary.LittleEndian.PutUint64(*b, v)
+ *b = (*b)[8:]
+}
diff --git a/src/archive/zip/writer_test.go b/src/archive/zip/writer_test.go
new file mode 100644
index 000000000..184a7d96a
--- /dev/null
+++ b/src/archive/zip/writer_test.go
@@ -0,0 +1,164 @@
+// Copyright 2011 The Go Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+package zip
+
+import (
+ "bytes"
+ "io"
+ "io/ioutil"
+ "math/rand"
+ "os"
+ "testing"
+)
+
+// TODO(adg): a more sophisticated test suite
+
+type WriteTest struct {
+ Name string
+ Data []byte
+ Method uint16
+ Mode os.FileMode
+}
+
+var writeTests = []WriteTest{
+ {
+ Name: "foo",
+ Data: []byte("Rabbits, guinea pigs, gophers, marsupial rats, and quolls."),
+ Method: Store,
+ Mode: 0666,
+ },
+ {
+ Name: "bar",
+ Data: nil, // large data set in the test
+ Method: Deflate,
+ Mode: 0644,
+ },
+ {
+ Name: "setuid",
+ Data: []byte("setuid file"),
+ Method: Deflate,
+ Mode: 0755 | os.ModeSetuid,
+ },
+ {
+ Name: "setgid",
+ Data: []byte("setgid file"),
+ Method: Deflate,
+ Mode: 0755 | os.ModeSetgid,
+ },
+ {
+ Name: "symlink",
+ Data: []byte("../link/target"),
+ Method: Deflate,
+ Mode: 0755 | os.ModeSymlink,
+ },
+}
+
+func TestWriter(t *testing.T) {
+ largeData := make([]byte, 1<<17)
+ for i := range largeData {
+ largeData[i] = byte(rand.Int())
+ }
+ writeTests[1].Data = largeData
+ defer func() {
+ writeTests[1].Data = nil
+ }()
+
+ // write a zip file
+ buf := new(bytes.Buffer)
+ w := NewWriter(buf)
+
+ for _, wt := range writeTests {
+ testCreate(t, w, &wt)
+ }
+
+ if err := w.Close(); err != nil {
+ t.Fatal(err)
+ }
+
+ // read it back
+ r, err := NewReader(bytes.NewReader(buf.Bytes()), int64(buf.Len()))
+ if err != nil {
+ t.Fatal(err)
+ }
+ for i, wt := range writeTests {
+ testReadFile(t, r.File[i], &wt)
+ }
+}
+
+func TestWriterFlush(t *testing.T) {
+ var buf bytes.Buffer
+ w := NewWriter(struct{ io.Writer }{&buf})
+ _, err := w.Create("foo")
+ if err != nil {
+ t.Fatal(err)
+ }
+ if buf.Len() > 0 {
+ t.Fatalf("Unexpected %d bytes already in buffer", buf.Len())
+ }
+ if err := w.Flush(); err != nil {
+ t.Fatal(err)
+ }
+ if buf.Len() == 0 {
+ t.Fatal("No bytes written after Flush")
+ }
+}
+
+func testCreate(t *testing.T, w *Writer, wt *WriteTest) {
+ header := &FileHeader{
+ Name: wt.Name,
+ Method: wt.Method,
+ }
+ if wt.Mode != 0 {
+ header.SetMode(wt.Mode)
+ }
+ f, err := w.CreateHeader(header)
+ if err != nil {
+ t.Fatal(err)
+ }
+ _, err = f.Write(wt.Data)
+ if err != nil {
+ t.Fatal(err)
+ }
+}
+
+func testReadFile(t *testing.T, f *File, wt *WriteTest) {
+ if f.Name != wt.Name {
+ t.Fatalf("File name: got %q, want %q", f.Name, wt.Name)
+ }
+ testFileMode(t, wt.Name, f, wt.Mode)
+ rc, err := f.Open()
+ if err != nil {
+ t.Fatal("opening:", err)
+ }
+ b, err := ioutil.ReadAll(rc)
+ if err != nil {
+ t.Fatal("reading:", err)
+ }
+ err = rc.Close()
+ if err != nil {
+ t.Fatal("closing:", err)
+ }
+ if !bytes.Equal(b, wt.Data) {
+ t.Errorf("File contents %q, want %q", b, wt.Data)
+ }
+}
+
+func BenchmarkCompressedZipGarbage(b *testing.B) {
+ b.ReportAllocs()
+ var buf bytes.Buffer
+ bigBuf := bytes.Repeat([]byte("a"), 1<<20)
+ for i := 0; i < b.N; i++ {
+ buf.Reset()
+ zw := NewWriter(&buf)
+ for j := 0; j < 3; j++ {
+ w, _ := zw.CreateHeader(&FileHeader{
+ Name: "foo",
+ Method: Deflate,
+ })
+ w.Write(bigBuf)
+ }
+ zw.Close()
+ }
+}
diff --git a/src/archive/zip/zip_test.go b/src/archive/zip/zip_test.go
new file mode 100644
index 000000000..32a16a79e
--- /dev/null
+++ b/src/archive/zip/zip_test.go
@@ -0,0 +1,395 @@
+// Copyright 2011 The Go Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+// Tests that involve both reading and writing.
+
+package zip
+
+import (
+ "bytes"
+ "fmt"
+ "hash"
+ "io"
+ "io/ioutil"
+ "sort"
+ "strings"
+ "testing"
+ "time"
+)
+
+func TestOver65kFiles(t *testing.T) {
+ buf := new(bytes.Buffer)
+ w := NewWriter(buf)
+ const nFiles = (1 << 16) + 42
+ for i := 0; i < nFiles; i++ {
+ _, err := w.CreateHeader(&FileHeader{
+ Name: fmt.Sprintf("%d.dat", i),
+ Method: Store, // avoid Issue 6136 and Issue 6138
+ })
+ if err != nil {
+ t.Fatalf("creating file %d: %v", i, err)
+ }
+ }
+ if err := w.Close(); err != nil {
+ t.Fatalf("Writer.Close: %v", err)
+ }
+ s := buf.String()
+ zr, err := NewReader(strings.NewReader(s), int64(len(s)))
+ if err != nil {
+ t.Fatalf("NewReader: %v", err)
+ }
+ if got := len(zr.File); got != nFiles {
+ t.Fatalf("File contains %d files, want %d", got, nFiles)
+ }
+ for i := 0; i < nFiles; i++ {
+ want := fmt.Sprintf("%d.dat", i)
+ if zr.File[i].Name != want {
+ t.Fatalf("File(%d) = %q, want %q", i, zr.File[i].Name, want)
+ }
+ }
+}
+
+func TestModTime(t *testing.T) {
+ var testTime = time.Date(2009, time.November, 10, 23, 45, 58, 0, time.UTC)
+ fh := new(FileHeader)
+ fh.SetModTime(testTime)
+ outTime := fh.ModTime()
+ if !outTime.Equal(testTime) {
+ t.Errorf("times don't match: got %s, want %s", outTime, testTime)
+ }
+}
+
+func testHeaderRoundTrip(fh *FileHeader, wantUncompressedSize uint32, wantUncompressedSize64 uint64, t *testing.T) {
+ fi := fh.FileInfo()
+ fh2, err := FileInfoHeader(fi)
+ if err != nil {
+ t.Fatal(err)
+ }
+ if got, want := fh2.Name, fh.Name; got != want {
+ t.Errorf("Name: got %s, want %s\n", got, want)
+ }
+ if got, want := fh2.UncompressedSize, wantUncompressedSize; got != want {
+ t.Errorf("UncompressedSize: got %d, want %d\n", got, want)
+ }
+ if got, want := fh2.UncompressedSize64, wantUncompressedSize64; got != want {
+ t.Errorf("UncompressedSize64: got %d, want %d\n", got, want)
+ }
+ if got, want := fh2.ModifiedTime, fh.ModifiedTime; got != want {
+ t.Errorf("ModifiedTime: got %d, want %d\n", got, want)
+ }
+ if got, want := fh2.ModifiedDate, fh.ModifiedDate; got != want {
+ t.Errorf("ModifiedDate: got %d, want %d\n", got, want)
+ }
+
+ if sysfh, ok := fi.Sys().(*FileHeader); !ok && sysfh != fh {
+ t.Errorf("Sys didn't return original *FileHeader")
+ }
+}
+
+func TestFileHeaderRoundTrip(t *testing.T) {
+ fh := &FileHeader{
+ Name: "foo.txt",
+ UncompressedSize: 987654321,
+ ModifiedTime: 1234,
+ ModifiedDate: 5678,
+ }
+ testHeaderRoundTrip(fh, fh.UncompressedSize, uint64(fh.UncompressedSize), t)
+}
+
+func TestFileHeaderRoundTrip64(t *testing.T) {
+ fh := &FileHeader{
+ Name: "foo.txt",
+ UncompressedSize64: 9876543210,
+ ModifiedTime: 1234,
+ ModifiedDate: 5678,
+ }
+ testHeaderRoundTrip(fh, uint32max, fh.UncompressedSize64, t)
+}
+
+type repeatedByte struct {
+ off int64
+ b byte
+ n int64
+}
+
+// rleBuffer is a run-length-encoded byte buffer.
+// It's an io.Writer (like a bytes.Buffer) and also an io.ReaderAt,
+// allowing random-access reads.
+type rleBuffer struct {
+ buf []repeatedByte
+}
+
+func (r *rleBuffer) Size() int64 {
+ if len(r.buf) == 0 {
+ return 0
+ }
+ last := &r.buf[len(r.buf)-1]
+ return last.off + last.n
+}
+
+func (r *rleBuffer) Write(p []byte) (n int, err error) {
+ var rp *repeatedByte
+ if len(r.buf) > 0 {
+ rp = &r.buf[len(r.buf)-1]
+ // Fast path, if p is entirely the same byte repeated.
+ if lastByte := rp.b; len(p) > 0 && p[0] == lastByte {
+ all := true
+ for _, b := range p {
+ if b != lastByte {
+ all = false
+ break
+ }
+ }
+ if all {
+ rp.n += int64(len(p))
+ return len(p), nil
+ }
+ }
+ }
+
+ for _, b := range p {
+ if rp == nil || rp.b != b {
+ r.buf = append(r.buf, repeatedByte{r.Size(), b, 1})
+ rp = &r.buf[len(r.buf)-1]
+ } else {
+ rp.n++
+ }
+ }
+ return len(p), nil
+}
+
+func (r *rleBuffer) ReadAt(p []byte, off int64) (n int, err error) {
+ if len(p) == 0 {
+ return
+ }
+ skipParts := sort.Search(len(r.buf), func(i int) bool {
+ part := &r.buf[i]
+ return part.off+part.n > off
+ })
+ parts := r.buf[skipParts:]
+ if len(parts) > 0 {
+ skipBytes := off - parts[0].off
+ for len(parts) > 0 {
+ part := parts[0]
+ for i := skipBytes; i < part.n; i++ {
+ if n == len(p) {
+ return
+ }
+ p[n] = part.b
+ n++
+ }
+ parts = parts[1:]
+ skipBytes = 0
+ }
+ }
+ if n != len(p) {
+ err = io.ErrUnexpectedEOF
+ }
+ return
+}
+
+// Just testing the rleBuffer used in the Zip64 test above. Not used by the zip code.
+func TestRLEBuffer(t *testing.T) {
+ b := new(rleBuffer)
+ var all []byte
+ writes := []string{"abcdeee", "eeeeeee", "eeeefghaaiii"}
+ for _, w := range writes {
+ b.Write([]byte(w))
+ all = append(all, w...)
+ }
+ if len(b.buf) != 10 {
+ t.Fatalf("len(b.buf) = %d; want 10", len(b.buf))
+ }
+
+ for i := 0; i < len(all); i++ {
+ for j := 0; j < len(all)-i; j++ {
+ buf := make([]byte, j)
+ n, err := b.ReadAt(buf, int64(i))
+ if err != nil || n != len(buf) {
+ t.Errorf("ReadAt(%d, %d) = %d, %v; want %d, nil", i, j, n, err, len(buf))
+ }
+ if !bytes.Equal(buf, all[i:i+j]) {
+ t.Errorf("ReadAt(%d, %d) = %q; want %q", i, j, buf, all[i:i+j])
+ }
+ }
+ }
+}
+
+// fakeHash32 is a dummy Hash32 that always returns 0.
+type fakeHash32 struct {
+ hash.Hash32
+}
+
+func (fakeHash32) Write(p []byte) (int, error) { return len(p), nil }
+func (fakeHash32) Sum32() uint32 { return 0 }
+
+func TestZip64(t *testing.T) {
+ if testing.Short() {
+ t.Skip("slow test; skipping")
+ }
+ const size = 1 << 32 // before the "END\n" part
+ testZip64(t, size)
+}
+
+func testZip64(t testing.TB, size int64) {
+ const chunkSize = 1024
+ chunks := int(size / chunkSize)
+ // write 2^32 bytes plus "END\n" to a zip file
+ buf := new(rleBuffer)
+ w := NewWriter(buf)
+ f, err := w.CreateHeader(&FileHeader{
+ Name: "huge.txt",
+ Method: Store,
+ })
+ if err != nil {
+ t.Fatal(err)
+ }
+ f.(*fileWriter).crc32 = fakeHash32{}
+ chunk := make([]byte, chunkSize)
+ for i := range chunk {
+ chunk[i] = '.'
+ }
+ for i := 0; i < chunks; i++ {
+ _, err := f.Write(chunk)
+ if err != nil {
+ t.Fatal("write chunk:", err)
+ }
+ }
+ end := []byte("END\n")
+ _, err = f.Write(end)
+ if err != nil {
+ t.Fatal("write end:", err)
+ }
+ if err := w.Close(); err != nil {
+ t.Fatal(err)
+ }
+
+ // read back zip file and check that we get to the end of it
+ r, err := NewReader(buf, int64(buf.Size()))
+ if err != nil {
+ t.Fatal("reader:", err)
+ }
+ f0 := r.File[0]
+ rc, err := f0.Open()
+ if err != nil {
+ t.Fatal("opening:", err)
+ }
+ rc.(*checksumReader).hash = fakeHash32{}
+ for i := 0; i < chunks; i++ {
+ _, err := io.ReadFull(rc, chunk)
+ if err != nil {
+ t.Fatal("read:", err)
+ }
+ }
+ gotEnd, err := ioutil.ReadAll(rc)
+ if err != nil {
+ t.Fatal("read end:", err)
+ }
+ if !bytes.Equal(gotEnd, end) {
+ t.Errorf("End of zip64 archive %q, want %q", gotEnd, end)
+ }
+ err = rc.Close()
+ if err != nil {
+ t.Fatal("closing:", err)
+ }
+ if size == 1<<32 {
+ if got, want := f0.UncompressedSize, uint32(uint32max); got != want {
+ t.Errorf("UncompressedSize %d, want %d", got, want)
+ }
+ }
+
+ if got, want := f0.UncompressedSize64, uint64(size)+uint64(len(end)); got != want {
+ t.Errorf("UncompressedSize64 %d, want %d", got, want)
+ }
+}
+
+func testInvalidHeader(h *FileHeader, t *testing.T) {
+ var buf bytes.Buffer
+ z := NewWriter(&buf)
+
+ f, err := z.CreateHeader(h)
+ if err != nil {
+ t.Fatalf("error creating header: %v", err)
+ }
+ if _, err := f.Write([]byte("hi")); err != nil {
+ t.Fatalf("error writing content: %v", err)
+ }
+ if err := z.Close(); err != nil {
+ t.Fatalf("error closing zip writer: %v", err)
+ }
+
+ b := buf.Bytes()
+ if _, err = NewReader(bytes.NewReader(b), int64(len(b))); err != ErrFormat {
+ t.Fatalf("got %v, expected ErrFormat", err)
+ }
+}
+
+func testValidHeader(h *FileHeader, t *testing.T) {
+ var buf bytes.Buffer
+ z := NewWriter(&buf)
+
+ f, err := z.CreateHeader(h)
+ if err != nil {
+ t.Fatalf("error creating header: %v", err)
+ }
+ if _, err := f.Write([]byte("hi")); err != nil {
+ t.Fatalf("error writing content: %v", err)
+ }
+ if err := z.Close(); err != nil {
+ t.Fatalf("error closing zip writer: %v", err)
+ }
+
+ b := buf.Bytes()
+ if _, err = NewReader(bytes.NewReader(b), int64(len(b))); err != nil {
+ t.Fatalf("got %v, expected nil", err)
+ }
+}
+
+// Issue 4302.
+func TestHeaderInvalidTagAndSize(t *testing.T) {
+ const timeFormat = "20060102T150405.000.txt"
+
+ ts := time.Now()
+ filename := ts.Format(timeFormat)
+
+ h := FileHeader{
+ Name: filename,
+ Method: Deflate,
+ Extra: []byte(ts.Format(time.RFC3339Nano)), // missing tag and len
+ }
+ h.SetModTime(ts)
+
+ testInvalidHeader(&h, t)
+}
+
+func TestHeaderTooShort(t *testing.T) {
+ h := FileHeader{
+ Name: "foo.txt",
+ Method: Deflate,
+ Extra: []byte{zip64ExtraId}, // missing size
+ }
+ testInvalidHeader(&h, t)
+}
+
+// Issue 4393. It is valid to have an extra data header
+// which contains no body.
+func TestZeroLengthHeader(t *testing.T) {
+ h := FileHeader{
+ Name: "extadata.txt",
+ Method: Deflate,
+ Extra: []byte{
+ 85, 84, 5, 0, 3, 154, 144, 195, 77, // tag 21589 size 5
+ 85, 120, 0, 0, // tag 30805 size 0
+ },
+ }
+ testValidHeader(&h, t)
+}
+
+// Just benchmarking how fast the Zip64 test above is. Not related to
+// our zip performance, since the test above disabled CRC32 and flate.
+func BenchmarkZip64Test(b *testing.B) {
+ for i := 0; i < b.N; i++ {
+ testZip64(b, 1<<26)
+ }
+}