// 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 gzip implements reading and writing of gzip format compressed files, // as specified in RFC 1952.
package gzip import ( ) const ( gzipID1 = 0x1f gzipID2 = 0x8b gzipDeflate = 8 flagText = 1 << 0 flagHdrCrc = 1 << 1 flagExtra = 1 << 2 flagName = 1 << 3 flagComment = 1 << 4 ) var ( // ErrChecksum is returned when reading GZIP data that has an invalid checksum. ErrChecksum = errors.New("gzip: invalid checksum") // ErrHeader is returned when reading GZIP data that has an invalid header. ErrHeader = errors.New("gzip: invalid header") ) var le = binary.LittleEndian // noEOF converts io.EOF to io.ErrUnexpectedEOF. func noEOF( error) error { if == io.EOF { return io.ErrUnexpectedEOF } return } // The gzip file stores a header giving metadata about the compressed file. // That header is exposed as the fields of the Writer and Reader structs. // // Strings must be UTF-8 encoded and may only contain Unicode code points // U+0001 through U+00FF, due to limitations of the GZIP file format. type Header struct { Comment string // comment Extra []byte // "extra data" ModTime time.Time // modification time Name string // file name OS byte // operating system type } // A Reader is an io.Reader that can be read to retrieve // uncompressed data from a gzip-format compressed file. // // In general, a gzip file can be a concatenation of gzip files, // each with its own header. Reads from the Reader // return the concatenation of the uncompressed data of each. // Only the first header is recorded in the Reader fields. // // Gzip files store a length and checksum of the uncompressed data. // The Reader will return an ErrChecksum when Read // reaches the end of the uncompressed data if it does not // have the expected length or checksum. Clients should treat data // returned by Read as tentative until they receive the io.EOF // marking the end of the data. type Reader struct { Header // valid after NewReader or Reader.Reset r flate.Reader decompressor io.ReadCloser digest uint32 // CRC-32, IEEE polynomial (section 8) size uint32 // Uncompressed size (section 2.3.1) buf [512]byte err error multistream bool } // NewReader creates a new Reader reading the given reader. // If r does not also implement io.ByteReader, // the decompressor may read more data than necessary from r. // // It is the caller's responsibility to call Close on the Reader when done. // // The Reader.Header fields will be valid in the Reader returned. func ( io.Reader) (*Reader, error) { := new(Reader) if := .Reset(); != nil { return nil, } return , nil } // Reset discards the Reader z's state and makes it equivalent to the // result of its original state from NewReader, but reading from r instead. // This permits reusing a Reader rather than allocating a new one. func ( *Reader) ( io.Reader) error { * = Reader{ decompressor: .decompressor, multistream: true, } if , := .(flate.Reader); { .r = } else { .r = bufio.NewReader() } .Header, .err = .readHeader() return .err } // Multistream controls whether the reader supports multistream files. // // If enabled (the default), the Reader expects the input to be a sequence // of individually gzipped data streams, each with its own header and // trailer, ending at EOF. The effect is that the concatenation of a sequence // of gzipped files is treated as equivalent to the gzip of the concatenation // of the sequence. This is standard behavior for gzip readers. // // Calling Multistream(false) disables this behavior; disabling the behavior // can be useful when reading file formats that distinguish individual gzip // data streams or mix gzip data streams with other data streams. // In this mode, when the Reader reaches the end of the data stream, // Read returns io.EOF. The underlying reader must implement io.ByteReader // in order to be left positioned just after the gzip stream. // To start the next stream, call z.Reset(r) followed by z.Multistream(false). // If there is no next stream, z.Reset(r) will return io.EOF. func ( *Reader) ( bool) { .multistream = } // readString reads a NUL-terminated string from z.r. // It treats the bytes read as being encoded as ISO 8859-1 (Latin-1) and // will output a string encoded using UTF-8. // This method always updates z.digest with the data read. func ( *Reader) () (string, error) { var error := false for := 0; ; ++ { if >= len(.buf) { return "", ErrHeader } .buf[], = .r.ReadByte() if != nil { return "", } if .buf[] > 0x7f { = true } if .buf[] == 0 { // Digest covers the NUL terminator. .digest = crc32.Update(.digest, crc32.IEEETable, .buf[:+1]) // Strings are ISO 8859-1, Latin-1 (RFC 1952, section 2.3.1). if { := make([]rune, 0, ) for , := range .buf[:] { = append(, rune()) } return string(), nil } return string(.buf[:]), nil } } } // readHeader reads the GZIP header according to section 2.3.1. // This method does not set z.err. func ( *Reader) () ( Header, error) { if _, = io.ReadFull(.r, .buf[:10]); != nil { // RFC 1952, section 2.2, says the following: // A gzip file consists of a series of "members" (compressed data sets). // // Other than this, the specification does not clarify whether a // "series" is defined as "one or more" or "zero or more". To err on the // side of caution, Go interprets this to mean "zero or more". // Thus, it is okay to return io.EOF here. return , } if .buf[0] != gzipID1 || .buf[1] != gzipID2 || .buf[2] != gzipDeflate { return , ErrHeader } := .buf[3] if := int64(le.Uint32(.buf[4:8])); > 0 { // Section 2.3.1, the zero value for MTIME means that the // modified time is not set. .ModTime = time.Unix(, 0) } // z.buf[8] is XFL and is currently ignored. .OS = .buf[9] .digest = crc32.ChecksumIEEE(.buf[:10]) if &flagExtra != 0 { if _, = io.ReadFull(.r, .buf[:2]); != nil { return , noEOF() } .digest = crc32.Update(.digest, crc32.IEEETable, .buf[:2]) := make([]byte, le.Uint16(.buf[:2])) if _, = io.ReadFull(.r, ); != nil { return , noEOF() } .digest = crc32.Update(.digest, crc32.IEEETable, ) .Extra = } var string if &flagName != 0 { if , = .readString(); != nil { return , } .Name = } if &flagComment != 0 { if , = .readString(); != nil { return , } .Comment = } if &flagHdrCrc != 0 { if _, = io.ReadFull(.r, .buf[:2]); != nil { return , noEOF() } := le.Uint16(.buf[:2]) if != uint16(.digest) { return , ErrHeader } } .digest = 0 if .decompressor == nil { .decompressor = flate.NewReader(.r) } else { .decompressor.(flate.Resetter).Reset(.r, nil) } return , nil } // Read implements io.Reader, reading uncompressed bytes from its underlying Reader. func ( *Reader) ( []byte) ( int, error) { if .err != nil { return 0, .err } , .err = .decompressor.Read() .digest = crc32.Update(.digest, crc32.IEEETable, [:]) .size += uint32() if .err != io.EOF { // In the normal case we return here. return , .err } // Finished file; check checksum and size. if , := io.ReadFull(.r, .buf[:8]); != nil { .err = noEOF() return , .err } := le.Uint32(.buf[:4]) := le.Uint32(.buf[4:8]) if != .digest || != .size { .err = ErrChecksum return , .err } .digest, .size = 0, 0 // File is ok; check if there is another. if !.multistream { return , io.EOF } .err = nil // Remove io.EOF if _, .err = .readHeader(); .err != nil { return , .err } // Read from next file, if necessary. if > 0 { return , nil } return .() } // Close closes the Reader. It does not close the underlying io.Reader. // In order for the GZIP checksum to be verified, the reader must be // fully consumed until the io.EOF. func ( *Reader) () error { return .decompressor.Close() }