archive/tar: properly handle header-only "files" in Reader

Certain special type-flags, specifically 1, 2, 3, 4, 5, 6,
do not have a data section. Thus, regardless of what the size field
says, we should not attempt to read any data for these special types.

The relevant PAX and USTAR specification says:
<<<
If the typeflag field is set to specify a file to be of type 1 (a link)
or 2 (a symbolic link), the size field shall be specified as zero.
If the typeflag field is set to specify a file of type 5 (directory),
the size field shall be interpreted as described under the definition
of that record type. No data logical records are stored for types 1, 2, or 5.
If the typeflag field is set to 3 (character special file),
4 (block special file), or 6 (FIFO), the meaning of the size field is
unspecified by this volume of POSIX.1-2008, and no data logical records shall
be stored on the medium.
Additionally, for type 6, the size field shall be ignored when reading.
If the typeflag field is set to any other value, the number of logical
records written following the header shall be (size+511)/512, ignoring
any fraction in the result of the division.
>>>

Contrary to the specification, we do not assert that the size field
is zero for type 1 and 2 since we liberally accept non-conforming formats.

Change-Id: I666b601597cb9d7a50caa081813d90ca9cfc52ed
Reviewed-on: https://go-review.googlesource.com/16614
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Run-TryBot: Brad Fitzpatrick <bradfitz@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>

archive/tar/reader.go

@@ -769,97 +769,77 @@ func (tr *Reader) readOldGNUSparseMap(header []byte) []sparseEntry {
 	return sp
 }
 
-// readGNUSparseMap1x0 reads the sparse map as stored in GNU's PAX sparse format version 1.0.
+// readGNUSparseMap1x0 reads the sparse map as stored in GNU's PAX sparse format
-// The sparse map is stored just before the file data and padded out to the nearest block boundary.
+// version 1.0. The format of the sparse map consists of a series of
+// newline-terminated numeric fields. The first field is the number of entries
+// and is always present. Following this are the entries, consisting of two
+// fields (offset, numBytes). This function must stop reading at the end
+// boundary of the block containing the last newline.
+//
+// Note that the GNU manual says that numeric values should be encoded in octal
+// format. However, the GNU tar utility itself outputs these values in decimal.
+// As such, this library treats values as being encoded in decimal.
 func readGNUSparseMap1x0(r io.Reader) ([]sparseEntry, error) {
-	buf := make([]byte, 2*blockSize)
+	var cntNewline int64
-	sparseHeader := buf[:blockSize]
+	var buf bytes.Buffer
+	var blk = make([]byte, blockSize)
 
-	// readDecimal is a helper function to read a decimal integer from the sparse map
+	// feedTokens copies data in numBlock chunks from r into buf until there are
-	// while making sure to read from the file in blocks of size blockSize
+	// at least cnt newlines in buf. It will not read more blocks than needed.
-	readDecimal := func() (int64, error) {
+	var feedTokens = func(cnt int64) error {
-		// Look for newline
+		for cntNewline < cnt {
-		nl := bytes.IndexByte(sparseHeader, '\n')
+			if _, err := io.ReadFull(r, blk); err != nil {
-		if nl == -1 {
+				if err == io.EOF {
-			if len(sparseHeader) >= blockSize {
+					err = io.ErrUnexpectedEOF
-				// This is an error
-				return 0, ErrHeader
 				}
-			oldLen := len(sparseHeader)
+				return err
-			newLen := oldLen + blockSize
+			}
-			if cap(sparseHeader) < newLen {
+			buf.Write(blk)
-				// There's more header, but we need to make room for the next block
+			for _, c := range blk {
-				copy(buf, sparseHeader)
+				if c == '\n' {
-				sparseHeader = buf[:newLen]
+					cntNewline++
-			} else {
+				}
-				// There's more header, and we can just reslice
+			}
-				sparseHeader = sparseHeader[:newLen]
+		}
+		return nil
 	}
 
-			// Now that sparseHeader is large enough, read next block
+	// nextToken gets the next token delimited by a newline. This assumes that
-			if _, err := io.ReadFull(r, sparseHeader[oldLen:newLen]); err != nil {
+	// at least one newline exists in the buffer.
-				return 0, err
+	var nextToken = func() string {
-			}
+		cntNewline--
-			// leaving this function for io.Reader makes it more testable
+		tok, _ := buf.ReadString('\n')
-			if tr, ok := r.(*Reader); ok && tr.RawAccounting {
+		return tok[:len(tok)-1] // Cut off newline
-				if _, err := tr.rawBytes.Write(sparseHeader[oldLen:newLen]); err != nil {
-					return 0, err
-				}
 	}
 
-			// Look for a newline in the new data
+	// Parse for the number of entries.
-			nl = bytes.IndexByte(sparseHeader[oldLen:newLen], '\n')
+	// Use integer overflow resistant math to check this.
-			if nl == -1 {
+	if err := feedTokens(1); err != nil {
-				// 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
-	}
-	// leaving this function for io.Reader makes it more testable
-	if tr, ok := r.(*Reader); ok && tr.RawAccounting {
-		if _, err := tr.rawBytes.Write(sparseHeader); err != nil {
 		return nil, err
 	}
+	numEntries, err := strconv.ParseInt(nextToken(), 10, 0) // Intentionally parse as native int
+	if err != nil || numEntries < 0 || int(2*numEntries) < int(numEntries) {
+		return nil, ErrHeader
 	}
 
-	// The first line contains the number of entries
+	// Parse for all member entries.
-	numEntries, err := readDecimal()
+	// numEntries is trusted after this since a potential attacker must have
-	if err != nil {
+	// committed resources proportional to what this library used.
+	if err := feedTokens(2 * numEntries); 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 := strconv.ParseInt(nextToken(), 10, 64)
-		offset, err := readDecimal()
 		if err != nil {
-			return nil, err
+			return nil, ErrHeader
 		}
-		// Read numBytes
+		numBytes, err := strconv.ParseInt(nextToken(), 10, 64)
-		numBytes, err := readDecimal()
 		if err != nil {
-			return nil, err
+			return nil, ErrHeader
 		}
-
 		sp = append(sp, sparseEntry{offset: offset, numBytes: numBytes})
 	}
-
 	return sp, nil
 }
 

archive/tar/reader_test.go

@@ -719,35 +719,82 @@ func TestReadGNUSparseMap0x1(t *testing.T) {
 }
 
 func TestReadGNUSparseMap1x0(t *testing.T) {
-	// This test uses lots of holes so the sparse header takes up more than two blocks
+	var sp = []sparseEntry{{1, 2}, {3, 4}}
-	numEntries := 100
+	for i := 0; i < 98; i++ {
-	expected := make([]sparseEntry, 0, numEntries)
+		sp = append(sp, sparseEntry{54321, 12345})
-	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
+	var vectors = []struct {
-	headerBlocks := (sparseMap.Len() + blockSize - 1) / blockSize
+		input     string        // Input data
-	bufLen := blockSize * headerBlocks
+		sparseMap []sparseEntry // Expected sparse entries to be outputted
-	buf := make([]byte, bufLen)
+		cnt       int           // Expected number of bytes read
-	copy(buf, sparseMap.Bytes())
+		err       error         // Expected errors that may be raised
+	}{{
+		input: "",
+		cnt:   0,
+		err:   io.ErrUnexpectedEOF,
+	}, {
+		input: "ab",
+		cnt:   2,
+		err:   io.ErrUnexpectedEOF,
+	}, {
+		input: strings.Repeat("\x00", 512),
+		cnt:   512,
+		err:   io.ErrUnexpectedEOF,
+	}, {
+		input: strings.Repeat("\x00", 511) + "\n",
+		cnt:   512,
+		err:   ErrHeader,
+	}, {
+		input: strings.Repeat("\n", 512),
+		cnt:   512,
+		err:   ErrHeader,
+	}, {
+		input:     "0\n" + strings.Repeat("\x00", 510) + strings.Repeat("a", 512),
+		sparseMap: []sparseEntry{},
+		cnt:       512,
+	}, {
+		input:     strings.Repeat("0", 512) + "0\n" + strings.Repeat("\x00", 510),
+		sparseMap: []sparseEntry{},
+		cnt:       1024,
+	}, {
+		input:     strings.Repeat("0", 1024) + "1\n2\n3\n" + strings.Repeat("\x00", 506),
+		sparseMap: []sparseEntry{{2, 3}},
+		cnt:       1536,
+	}, {
+		input: strings.Repeat("0", 1024) + "1\n2\n\n" + strings.Repeat("\x00", 509),
+		cnt:   1536,
+		err:   ErrHeader,
+	}, {
+		input: strings.Repeat("0", 1024) + "1\n2\n" + strings.Repeat("\x00", 508),
+		cnt:   1536,
+		err:   io.ErrUnexpectedEOF,
+	}, {
+		input: "-1\n2\n\n" + strings.Repeat("\x00", 506),
+		cnt:   512,
+		err:   ErrHeader,
+	}, {
+		input: "1\nk\n2\n" + strings.Repeat("\x00", 506),
+		cnt:   512,
+		err:   ErrHeader,
+	}, {
+		input:     "100\n1\n2\n3\n4\n" + strings.Repeat("54321\n0000000000000012345\n", 98) + strings.Repeat("\x00", 512),
+		cnt:       2560,
+		sparseMap: sp,
+	}}
 
-	// Get an reader to read the sparse map
+	for i, v := range vectors {
-	r := bytes.NewReader(buf)
+		r := strings.NewReader(v.input)
-
-	// Read the sparse map
 		sp, err := readGNUSparseMap1x0(r)
-	if err != nil {
+		if !reflect.DeepEqual(sp, v.sparseMap) && !(len(sp) == 0 && len(v.sparseMap) == 0) {
-		t.Errorf("Unexpected error: %v", err)
+			t.Errorf("test %d, readGNUSparseMap1x0(...): got %v, want %v", i, sp, v.sparseMap)
+		}
+		if numBytes := len(v.input) - r.Len(); numBytes != v.cnt {
+			t.Errorf("test %d, bytes read: got %v, want %v", i, numBytes, v.cnt)
+		}
+		if err != v.err {
+			t.Errorf("test %d, unexpected error: got %v, want %v", i, err, v.err)
 		}
-	if !reflect.DeepEqual(sp, expected) {
-		t.Errorf("Incorrect sparse map: got %v, wanted %v", sp, expected)
 	}
 }