File:    APPNOTE.TXT - .ZIP File Format Specification
Version: 6.3.2
Revised: September 28, 2007
Copyright (c) 1989 - 2007 PKWARE Inc., All Rights Reserved.

The use of certain technological aspects disclosed in the current
APPNOTE is available pursuant to the below section entitled
"Incorporating PKWARE Proprietary Technology into Your Product".

I. Purpose
----------

This specification is intended to define a cross-platform,
interoperable file storage and transfer format.  Since its
first publication in 1989, PKWARE has remained committed to
ensuring the interoperability of the .ZIP file format through
publication and maintenance of this specification.  We trust that
all .ZIP compatible vendors and application developers that have
adopted and benefited from this format will share and support
this commitment to interoperability.

II. Contacting PKWARE
---------------------

     PKWARE, Inc.
     648 N. Plankinton Avenue, Suite 220
     Milwaukee, WI 53203
     +1-414-289-9788
     +1-414-289-9789 FAX
     zipformat@pkware.com

III. Disclaimer
---------------

Although PKWARE will attempt to supply current and accurate
information relating to its file formats, algorithms, and the
subject programs, the possibility of error or omission cannot
be eliminated. PKWARE therefore expressly disclaims any warranty
that the information contained in the associated materials relating
to the subject programs and/or the format of the files created or
accessed by the subject programs and/or the algorithms used by
the subject programs, or any other matter, is current, correct or
accurate as delivered.  Any risk of damage due to any possible
inaccurate information is assumed by the user of the information.
Furthermore, the information relating to the subject programs
and/or the file formats created or accessed by the subject
programs and/or the algorithms used by the subject programs is
subject to change without notice.

If the version of this file is marked as a NOTIFICATION OF CHANGE,
the content defines an Early Feature Specification (EFS) change
to the .ZIP file format that may be subject to modification prior
to publication of the Final Feature Specification (FFS).  This
document may also contain information on Planned Feature
Specifications (PFS) defining recognized future extensions.

IV. Change Log
--------------

Version       Change Description                        Date
-------       ------------------                       ----------
5.2           -Single Password Symmetric Encryption    06/02/2003
               storage

6.1.0         -Smartcard compatibility                 01/20/2004
              -Documentation on certificate storage

6.2.0         -Introduction of Central Directory       04/26/2004
               Encryption for encrypting metadata
              -Added OS/X to Version Made By values

6.2.1         -Added Extra Field placeholder for       04/01/2005
               POSZIP using ID 0x4690

              -Clarified size field on
               "zip64 end of central directory record"

6.2.2         -Documented Final Feature Specification  01/06/2006
               for Strong Encryption

              -Clarifications and typographical
               corrections

6.3.0         -Added tape positioning storage          09/29/2006
               parameters

              -Expanded list of supported hash algorithms

              -Expanded list of supported compression
               algorithms

              -Expanded list of supported encryption
               algorithms

              -Added option for Unicode filename
               storage

              -Clarifications for consistent use
               of Data Descriptor records

              -Added additional "Extra Field"
               definitions

6.3.1         -Corrected standard hash values for      04/11/2007
               SHA-256/384/512

6.3.2         -Added compression method 97             09/28/2007

              -Documented InfoZIP "Extra Field"
               values for UTF-8 file name and
               file comment storage

V. General Format of a .ZIP file
--------------------------------

  Files stored in arbitrary order.  Large .ZIP files can span multiple
  volumes or be split into user-defined segment sizes. All values
  are stored in little-endian byte order unless otherwise specified.

  Overall .ZIP file format:

    [local file header 1]
    [file data 1]
    [data descriptor 1]
    .
    .
    .
    [local file header n]
    [file data n]
    [data descriptor n]
    [archive decryption header]
    [archive extra data record]
    [central directory]
    [zip64 end of central directory record]
    [zip64 end of central directory locator]
    [end of central directory record]


  A.  Local file header:

        local file header signature     4 bytes  (0x04034b50)
        version needed to extract       2 bytes
        general purpose bit flag        2 bytes
        compression method              2 bytes
        last mod file time              2 bytes
        last mod file date              2 bytes
        crc-32                          4 bytes
        compressed size                 4 bytes
        uncompressed size               4 bytes
        file name length                2 bytes
        extra field length              2 bytes

        file name (variable size)
        extra field (variable size)

  B.  File data

      Immediately following the local header for a file
      is the compressed or stored data for the file.
      The series of [local file header][file data][data
      descriptor] repeats for each file in the .ZIP archive.

  C.  Data descriptor:

        crc-32                          4 bytes
        compressed size                 4 bytes
        uncompressed size               4 bytes

      This descriptor exists only if bit 3 of the general
      purpose bit flag is set (see below).  It is byte aligned
      and immediately follows the last byte of compressed data.
      This descriptor is used only when it was not possible to
      seek in the output .ZIP file, e.g., when the output .ZIP file
      was standard output or a non-seekable device.  For ZIP64(tm) format
      archives, the compressed and uncompressed sizes are 8 bytes each.

      When compressing files, compressed and uncompressed sizes
      should be stored in ZIP64 format (as 8 byte values) when a
      files size exceeds 0xFFFFFFFF.   However ZIP64 format may be
      used regardless of the size of a file.  When extracting, if
      the zip64 extended information extra field is present for
      the file the compressed and uncompressed sizes will be 8
      byte values. 

      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.
      When writing ZIP files, it is recommended to include the
      signature value marking the data descriptor record.  When
      the signature is used, the fields currently defined for
      the data descriptor record will immediately follow the
      signature.

      An extensible data descriptor will be released in a future
      version of this APPNOTE.  This new record is intended to
      resolve conflicts with the use of this record going forward,
      and to provide better support for streamed file processing.

      When the Central Directory Encryption method is used, the data
      descriptor record is not required, but may be used.  If present,
      and bit 3 of the general purpose bit field is set to indicate
      its presence, the values in fields of the data descriptor
      record should be set to binary zeros.

  D.  Archive decryption header: 

      The Archive Decryption Header is introduced in version 6.2
      of the ZIP format specification.  This record exists in support
      of the Central Directory Encryption Feature implemented as part of
      the Strong Encryption Specification as described in this document.
      When the Central Directory Structure is encrypted, this decryption
      header will precede the encrypted data segment.  The encrypted
      data segment will consist of the Archive extra data record (if
      present) and the encrypted Central Directory Structure data.
      The format of this data record is identical to the Decryption
      header record preceding compressed file data.  If the central
      directory structure is encrypted, the location of the start of
      this data record is determined using the Start of Central Directory
      field in the Zip64 End of Central Directory record.  Refer to the
      section on the Strong Encryption Specification for information
      on the fields used in the Archive Decryption Header record.


  E.  Archive extra data record:

        archive extra data signature    4 bytes  (0x08064b50)
        extra field length              4 bytes
        extra field data                (variable size)

      The Archive Extra Data Record is introduced in version 6.2
      of the ZIP format specification.  This record exists in support
      of the Central Directory Encryption Feature implemented as part of
      the Strong Encryption Specification as described in this document.
      When present, this record immediately precedes the central
      directory data structure.  The size of this data record will be
      included in the Size of the Central Directory field in the
      End of Central Directory record.  If the central directory structure
      is compressed, but not encrypted, the location of the start of
      this data record is determined using the Start of Central Directory
      field in the Zip64 End of Central Directory record. 


  F.  Central directory structure:

      [file header 1]
      .
      .
      .
      [file header n]
      [digital signature]

      File header:

        central file header signature   4 bytes  (0x02014b50)
        version made by                 2 bytes
        version needed to extract       2 bytes
        general purpose bit flag        2 bytes
        compression method              2 bytes
        last mod file time              2 bytes
        last mod file date              2 bytes
        crc-32                          4 bytes
        compressed size                 4 bytes
        uncompressed size               4 bytes
        file name length                2 bytes
        extra field length              2 bytes
        file comment length             2 bytes
        disk number start               2 bytes
        internal file attributes        2 bytes
        external file attributes        4 bytes
        relative offset of local header 4 bytes

        file name (variable size)
        extra field (variable size)
        file comment (variable size)

      Digital signature:

        header signature                4 bytes  (0x05054b50)
        size of data                    2 bytes
        signature data (variable size)

      With the introduction of the Central Directory Encryption
      feature in version 6.2 of this specification, the Central
      Directory Structure may be stored both compressed and encrypted.
      Although not required, it is assumed when encrypting the
      Central Directory Structure, that it will be compressed
      for greater storage efficiency.  Information on the
      Central Directory Encryption feature can be found in the section
      describing the Strong Encryption Specification. The Digital
      Signature record will be neither compressed nor encrypted.

  G.  Zip64 end of central directory record

        zip64 end of central dir
        signature                       4 bytes  (0x06064b50)
        size of zip64 end of central
        directory record                8 bytes
        version made by                 2 bytes
        version needed to extract       2 bytes
        number of this disk             4 bytes
        number of the disk with the
        start of the central directory  4 bytes
        total number of entries in the
        central directory on this disk  8 bytes
        total number of entries in the
        central directory               8 bytes
        size of the central directory   8 bytes
        offset of start of central
        directory with respect to
        the starting disk number        8 bytes
        zip64 extensible data sector    (variable size)

        The value stored into the "size of zip64 end of central
        directory record" should be the size of the remaining
        record and should not include the leading 12 bytes.
 
        Size = SizeOfFixedFields + SizeOfVariableData - 12.

        The above record structure defines Version 1 of the
        zip64 end of central directory record. Version 1 was
        implemented in versions of this specification preceding
        6.2 in support of the ZIP64 large file feature. The
        introduction of the Central Directory Encryption feature
        implemented in version 6.2 as part of the Strong Encryption
        Specification defines Version 2 of this record structure.
        Refer to the section describing the Strong Encryption
        Specification for details on the version 2 format for
        this record.

        Special purpose data may reside in the zip64 extensible data
        sector field following either a V1 or V2 version of this
        record.  To ensure identification of this special purpose data
        it must include an identifying header block consisting of the
        following:

           Header ID  -  2 bytes
           Data Size  -  4 bytes

        The Header ID field indicates the type of data that is in the
        data block that follows.

        Data Size identifies the number of bytes that follow for this
        data block type.

        Multiple special purpose data blocks may be present, but each
        must be preceded by a Header ID and Data Size field.  Current
        mappings of Header ID values supported in this field are as
        defined in APPENDIX C.

  H.  Zip64 end of central directory locator

        zip64 end of central dir locator
        signature                       4 bytes  (0x07064b50)
        number of the disk with the
        start of the zip64 end of
        central directory               4 bytes
        relative offset of the zip64
        end of central directory record 8 bytes
        total number of disks           4 bytes
       
  I.  End of central directory record:

        end of central dir signature    4 bytes  (0x06054b50)
        number of this disk             2 bytes
        number of the disk with the
        start of the central directory  2 bytes
        total number of entries in the
        central directory on this disk  2 bytes
        total number of entries in
        the central directory           2 bytes
        size of the central directory   4 bytes
        offset of start of central
        directory with respect to
        the starting disk number        4 bytes
        .ZIP file comment length        2 bytes
        .ZIP file comment       (variable size)

  J.  Explanation of fields:

      version made by (2 bytes)

          The upper byte indicates the compatibility of the file
          attribute information.  If the external file attributes
          are compatible with MS-DOS and can be read by PKZIP for
          DOS version 2.04g then this value will be zero.  If these
          attributes are not compatible, then this value will
          identify the host system on which the attributes are
          compatible.  Software can use this information to determine
          the line record format for text files etc.  The current
          mappings are:

          0 - MS-DOS and OS/2 (FAT / VFAT / FAT32 file systems)
          1 - Amiga                     2 - OpenVMS
          3 - UNIX                      4 - VM/CMS
          5 - Atari ST                  6 - OS/2 H.P.F.S.
          7 - Macintosh                 8 - Z-System
          9 - CP/M                     10 - Windows NTFS
         11 - MVS (OS/390 - Z/OS)      12 - VSE
         13 - Acorn Risc               14 - VFAT
         15 - alternate MVS            16 - BeOS
         17 - Tandem                   18 - OS/400
         19 - OS/X (Darwin)            20 thru 255 - unused

          The lower byte indicates the ZIP specification version
          (the version of this document) supported by the software
          used to encode the file.  The value/10 indicates the major
          version number, and the value mod 10 is the minor version
          number. 

      version needed to extract (2 bytes)

          The minimum supported ZIP specification version needed to
          extract the file, mapped as above.  This value is based on
          the specific format features a ZIP program must support to
          be able to extract the file.  If multiple features are
          applied to a file, the minimum version should be set to the
          feature having the highest value. New features or feature
          changes affecting the published format specification will be
          implemented using higher version numbers than the last
          published value to avoid conflict.

          Current minimum feature versions are as defined below:

          1.0 - Default value
          1.1 - File is a volume label
          2.0 - File is a folder (directory)
          2.0 - File is compressed using Deflate compression
          2.0 - File is encrypted using traditional PKWARE encryption
          2.1 - File is compressed using Deflate64(tm)
          2.5 - File is compressed using PKWARE DCL Implode
          2.7 - File is a patch data set
          4.5 - File uses ZIP64 format extensions
          4.6 - File is compressed using BZIP2 compression*
          5.0 - File is encrypted using DES
          5.0 - File is encrypted using 3DES
          5.0 - File is encrypted using original RC2 encryption
          5.0 - File is encrypted using RC4 encryption
          5.1 - File is encrypted using AES encryption
          5.1 - File is encrypted using corrected RC2 encryption**
          5.2 - File is encrypted using corrected RC2-64 encryption**
          6.1 - File is encrypted using non-OAEP key wrapping***
          6.2 - Central directory encryption
          6.3 - File is compressed using LZMA
          6.3 - File is compressed using PPMd+
          6.3 - File is encrypted using Blowfish
          6.3 - File is encrypted using Twofish


          * Early 7.x (pre-7.2) versions of PKZIP incorrectly set the
          version needed to extract for BZIP2 compression to be 50
          when it should have been 46.

          ** Refer to the section on Strong Encryption Specification
          for additional information regarding RC2 corrections.

          *** Certificate encryption using non-OAEP key wrapping is the
          intended mode of operation for all versions beginning with 6.1.
          Support for OAEP key wrapping should only be used for
          backward compatibility when sending ZIP files to be opened by
          versions of PKZIP older than 6.1 (5.0 or 6.0).

          + Files compressed using PPMd should set the version
          needed to extract field to 6.3, however, not all ZIP
          programs enforce this and may be unable to decompress
          data files compressed using PPMd if this value is set.

          When using ZIP64 extensions, the corresponding value in the
          zip64 end of central directory record should also be set. 
          This field should be set appropriately to indicate whether
          Version 1 or Version 2 format is in use.

      general purpose bit flag: (2 bytes)

          Bit 0: If set, indicates that the file is encrypted.

          (For Method 6 - Imploding)
          Bit 1: If the compression method used was type 6,
                 Imploding, then this bit, if set, indicates
                 an 8K sliding dictionary was used.  If clear,
                 then a 4K sliding dictionary was used.
          Bit 2: If the compression method used was type 6,
                 Imploding, then this bit, if set, indicates
                 3 Shannon-Fano trees were used to encode the
                 sliding dictionary output.  If clear, then 2
                 Shannon-Fano trees were used.

          (For Methods 8 and 9 - Deflating)
          Bit 2  Bit 1
            0      0    Normal (-en) compression option was used.
            0      1    Maximum (-exx/-ex) compression option was used.
            1      0    Fast (-ef) compression option was used.
            1      1    Super Fast (-es) compression option was used.

          (For Method 14 - LZMA)
          Bit 1: If the compression method used was type 14,
                 LZMA, then this bit, if set, indicates
                 an end-of-stream (EOS) marker is used to
                 mark the end of the compressed data stream.
                 If clear, then an EOS marker is not present
                 and the compressed data size must be known
                 to extract.

          Note:  Bits 1 and 2 are undefined if the compression
                 method is any other.

          Bit 3: If this bit is set, the fields crc-32, compressed
                 size and uncompressed size are set to zero in the
                 local header.  The correct values are put in the
                 data descriptor immediately following the compressed
                 data.  (Note: PKZIP version 2.04g for DOS only
                 recognizes this bit for method 8 compression, newer
                 versions of PKZIP recognize this bit for any
                 compression method.)

          Bit 4: Reserved for use with method 8, for enhanced
                 deflating.

          Bit 5: If this bit is set, this indicates that the file is
                 compressed patched data.  (Note: Requires PKZIP
                 version 2.70 or greater)

          Bit 6: Strong encryption.  If this bit is set, you should
                 set the version needed to extract value to at least
                 50 and you must also set bit 0.  If AES encryption
                 is used, the version needed to extract value must
                 be at least 51.

          Bit 7: Currently unused.

          Bit 8: Currently unused.

          Bit 9: Currently unused.

          Bit 10: Currently unused.

          Bit 11: Language encoding flag (EFS).  If this bit is set,
                  the filename and comment fields for this file
                  must be encoded using UTF-8. (see APPENDIX D)

          Bit 12: Reserved by PKWARE for enhanced compression.

          Bit 13: Used when encrypting the Central Directory to indicate
                  selected data values in the Local Header are masked to
                  hide their actual values.  See the section describing
                  the Strong Encryption Specification for details.

          Bit 14: Reserved by PKWARE.

          Bit 15: Reserved by PKWARE.

      compression method: (2 bytes)

          (see accompanying documentation for algorithm
          descriptions)

          0 - The file is stored (no compression)
          1 - The file is Shrunk
          2 - The file is Reduced with compression factor 1
          3 - The file is Reduced with compression factor 2
          4 - The file is Reduced with compression factor 3
          5 - The file is Reduced with compression factor 4
          6 - The file is Imploded
          7 - Reserved for Tokenizing compression algorithm
          8 - The file is Deflated
          9 - Enhanced Deflating using Deflate64(tm)
         10 - PKWARE Data Compression Library Imploding (old IBM TERSE)
         11 - Reserved by PKWARE
         12 - File is compressed using BZIP2 algorithm
         13 - Reserved by PKWARE
         14 - LZMA (EFS)
         15 - Reserved by PKWARE
         16 - Reserved by PKWARE
         17 - Reserved by PKWARE
         18 - File is compressed using IBM TERSE (new)
         19 - IBM LZ77 z Architecture (PFS)
         97 - WavPack compressed data
         98 - PPMd version I, Rev 1

      date and time fields: (2 bytes each)

          The date and time are encoded in standard MS-DOS format.
          If input came from standard input, the date and time are
          those at which compression was started for this data.
          If encrypting the central directory and general purpose bit
          flag 13 is set indicating masking, the value stored in the
          Local Header will be zero.

      CRC-32: (4 bytes)

          The CRC-32 algorithm was generously contributed by
          David Schwaderer and can be found in his excellent
          book "C Programmers Guide to NetBIOS" published by
          Howard W. Sams & Co. Inc.  The 'magic number' for
          the CRC is 0xdebb20e3.  The proper CRC pre and post
          conditioning is used, meaning that the CRC register
          is pre-conditioned with all ones (a starting value
          of 0xffffffff) and the value is post-conditioned by
          taking the one's complement of the CRC residual.
          If bit 3 of the general purpose flag is set, this
          field is set to zero in the local header and the correct
          value is put in the data descriptor and in the central
          directory. When encrypting the central directory, if the
          local header is not in ZIP64 format and general purpose
          bit flag 13 is set indicating masking, the value stored
          in the Local Header will be zero.

      compressed size: (4 bytes)
      uncompressed size: (4 bytes)

          The size of the file compressed and uncompressed,
          respectively.  When a decryption header is present it will
          be placed in front of the file data and the value of the
          compressed file size will include the bytes of the decryption
          header.  If bit 3 of the general purpose bit flag is set,
          these fields are set to zero in the local header and the
          correct values are put in the data descriptor and
          in the central directory.  If an archive is in ZIP64 format
          and the value in this field is 0xFFFFFFFF, the size will be
          in the corresponding 8 byte ZIP64 extended information
          extra field.  When encrypting the central directory, if the
          local header is not in ZIP64 format and general purpose bit
          flag 13 is set indicating masking, the value stored for the
          uncompressed size in the Local Header will be zero.

      file name length: (2 bytes)
      extra field length: (2 bytes)
      file comment length: (2 bytes)

          The length of the file name, extra field, and comment
          fields respectively.  The combined length of any
          directory record and these three fields should not
          generally exceed 65,535 bytes.  If input came from standard
          input, the file name length is set to zero. 

      disk number start: (2 bytes)

          The number of the disk on which this file begins.  If an
          archive is in ZIP64 format and the value in this field is
          0xFFFF, the size will be in the corresponding 4 byte zip64
          extended information extra field.

      internal file attributes: (2 bytes)

          Bits 1 and 2 are reserved for use by PKWARE.

          The lowest bit of this field indicates, if set, that
          the file is apparently an ASCII or text file.  If not
          set, that the file apparently contains binary data.
          The remaining bits are unused in version 1.0.

          The 0x0002 bit of this field indicates, if set, that a
          4 byte variable record length control field precedes each
          logical record indicating the length of the record. The
          record length control field is stored in little-endian byte
          order.  This flag is independent of text control characters,
          and if used in conjunction with text data, includes any
          control characters in the total length of the record. This
          value is provided for mainframe data transfer support.

      external file attributes: (4 bytes)

          The mapping of the external attributes is
          host-system dependent (see 'version made by').  For
          MS-DOS, the low order byte is the MS-DOS directory
          attribute byte.  If input came from standard input, this
          field is set to zero.

      relative offset of local header: (4 bytes)

          This is the offset from the start of the first disk on
          which this file appears, to where the local header should
          be found.  If an archive is in ZIP64 format and the value
          in this field is 0xFFFFFFFF, the size will be in the
          corresponding 8 byte zip64 extended information extra field.

      file name: (Variable)

          The name of the file, with optional relative path.
          The path stored should not contain a drive or
          device letter, or a leading slash.  All slashes
          should be forward slashes '/' as opposed to
          backwards slashes '\' for compatibility with Amiga
          and UNIX file systems etc.  If input came from standard
          input, there is no file name field.  If encrypting
          the central directory and general purpose bit flag 13 is set
          indicating masking, the file name stored in the Local Header
          will not be the actual file name.  A masking value consisting
          of a unique hexadecimal value will be stored.  This value will
          be sequentially incremented for each file in the archive. See
          the section on the Strong Encryption Specification for details
          on retrieving the encrypted file name.

      extra field: (Variable)

          This is for expansion.  If additional information
          needs to be stored for special needs or for specific
          platforms, it should be stored here.  Earlier versions
          of the software can then safely skip this file, and
          find the next file or header.  This field will be 0
          length in version 1.0.

          In order to allow different programs and different types
          of information to be stored in the 'extra' field in .ZIP
          files, the following structure should be used for all
          programs storing data in this field:

          header1+data1 + header2+data2 . . .

          Each header should consist of:

            Header ID - 2 bytes
            Data Size - 2 bytes

          Note: all fields stored in Intel low-byte/high-byte order.

          The Header ID field indicates the type of data that is in
          the following data block.

          Header ID's of 0 thru 31 are reserved for use by PKWARE.
          The remaining ID's can be used by third party vendors for
          proprietary usage.

          The current Header ID mappings defined by PKWARE are:

          0x0001        Zip64 extended information extra field
          0x0007        AV Info
          0x0008        Reserved for extended language encoding data (PFS)
                        (see APPENDIX D)
          0x0009        OS/2
          0x000a        NTFS
          0x000c        OpenVMS
          0x000d        UNIX
          0x000e        Reserved for file stream and fork descriptors
          0x000f        Patch Descriptor
          0x0014        PKCS#7 Store for X.509 Certificates
          0x0015        X.509 Certificate ID and Signature for
                        individual file
          0x0016        X.509 Certificate ID for Central Directory
          0x0017        Strong Encryption Header
          0x0018        Record Management Controls
          0x0019        PKCS#7 Encryption Recipient Certificate List
          0x0065        IBM S/390 (Z390), AS/400 (I400) attributes
                        - uncompressed
          0x0066        Reserved for IBM S/390 (Z390), AS/400 (I400)
                        attributes - compressed
          0x4690        POSZIP 4690 (reserved)

          Third party mappings commonly used are:


          0x07c8        Macintosh
          0x2605        ZipIt Macintosh
          0x2705        ZipIt Macintosh 1.3.5+
          0x2805        ZipIt Macintosh 1.3.5+
          0x334d        Info-ZIP Macintosh
          0x4341        Acorn/SparkFS
          0x4453        Windows NT security descriptor (binary ACL)
          0x4704        VM/CMS
          0x470f        MVS
          0x4b46        FWKCS MD5 (see below)
          0x4c41        OS/2 access control list (text ACL)
          0x4d49        Info-ZIP OpenVMS
          0x4f4c        Xceed original location extra field
          0x5356        AOS/VS (ACL)
          0x5455        extended timestamp
          0x554e        Xceed unicode extra field
          0x5855        Info-ZIP UNIX (original, also OS/2, NT, etc)
          0x6375        Info-ZIP Unicode Comment Extra Field
          0x6542        BeOS/BeBox
          0x7075        Info-ZIP Unicode Path Extra Field
          0x756e        ASi UNIX
          0x7855        Info-ZIP UNIX (new)
          0xa220        Microsoft Open Packaging Growth Hint
          0xfd4a        SMS/QDOS

          Detailed descriptions of Extra Fields defined by third
          party mappings will be documented as information on
          these data structures is made available to PKWARE. 
          PKWARE does not guarantee the accuracy of any published
          third party data.

          The Data Size field indicates the size of the following
          data block. Programs can use this value to skip to the
          next header block, passing over any data blocks that are
          not of interest.

          Note: As stated above, the size of the entire .ZIP file
                header, including the file name, comment, and extra
                field should not exceed 64K in size.

          In case two different programs should appropriate the same
          Header ID value, it is strongly recommended that each
          program place a unique signature of at least two bytes in
          size (and preferably 4 bytes or bigger) at the start of
          each data area.  Every program should verify that its
          unique signature is present, in addition to the Header ID
          value being correct, before assuming that it is a block of
          known type.

         -Zip64 Extended Information Extra Field (0x0001):

          The following is the layout of the zip64 extended
          information "extra" block. If one of the size or
          offset fields in the Local or Central directory
          record is too small to hold the required data,
          a Zip64 extended information record is created.
          The order of the fields in the zip64 extended
          information record is fixed, but the fields will
          only appear if the corresponding Local or Central
          directory record field is set to 0xFFFF or 0xFFFFFFFF.

          Note: all fields stored in Intel low-byte/high-byte order.

          Value      Size       Description
          -----      ----       -----------
  (ZIP64) 0x0001     2 bytes    Tag for this "extra" block type
          Size       2 bytes    Size of this "extra" block
          Original
          Size       8 bytes    Original uncompressed file size
          Compressed
          Size       8 bytes    Size of compressed data
          Relative Header
          Offset     8 bytes    Offset of local header record
          Disk Start
          Number     4 bytes    Number of the disk on which
                                this file starts

          This entry in the Local header must include BOTH original
          and compressed file size fields. If encrypting the
          central directory and bit 13 of the general purpose bit
          flag is set indicating masking, the value stored in the
          Local Header for the original file size will be zero.


         -OS/2 Extra Field (0x0009):

          The following is the layout of the OS/2 attributes "extra"
          block.  (Last Revision  09/05/95)

          Note: all fields stored in Intel low-byte/high-byte order.

          Value       Size          Description
          -----       ----          -----------
  (OS/2)  0x0009      2 bytes       Tag for this "extra" block type
          TSize       2 bytes       Size for the following data block
          BSize       4 bytes       Uncompressed Block Size
          CType       2 bytes       Compression type
          EACRC       4 bytes       CRC value for uncompress block
          (var)       variable      Compressed block

          The OS/2 extended attribute structure (FEA2LIST) is
          compressed and then stored in it's entirety within this
          structure.  There will only ever be one "block" of data in
          VarFields[].

         -NTFS Extra Field (0x000a):

          The following is the layout of the NTFS attributes
          "extra" block. (Note: At this time the Mtime, Atime
          and Ctime values may be used on any WIN32 system.) 

          Note: all fields stored in Intel low-byte/high-byte order.

          Value      Size       Description
          -----      ----       -----------
  (NTFS)  0x000a     2 bytes    Tag for this "extra" block type
          TSize      2 bytes    Size of the total "extra" block
          Reserved   4 bytes    Reserved for future use
          Tag1       2 bytes    NTFS attribute tag value #1
          Size1      2 bytes    Size of attribute #1, in bytes
          (var.)     Size1      Attribute #1 data
          .
          .
          .
          TagN       2 bytes    NTFS attribute tag value #N
          SizeN      2 bytes    Size of attribute #N, in bytes
          (var.)     SizeN      Attribute #N data

          For NTFS, values for Tag1 through TagN are as follows:
          (currently only one set of attributes is defined for NTFS)

          Tag        Size       Description
          -----      ----       -----------
          0x0001     2 bytes    Tag for attribute #1
          Size1      2 bytes    Size of attribute #1, in bytes
          Mtime      8 bytes    File last modification time
          Atime      8 bytes    File last access time
          Ctime      8 bytes    File creation time

         -OpenVMS Extra Field (0x000c):

          The following is the layout of the OpenVMS attributes
          "extra" block.

          Note: all fields stored in Intel low-byte/high-byte order.

          Value      Size       Description
          -----      ----       -----------
  (VMS)   0x000c     2 bytes    Tag for this "extra" block type
          TSize      2 bytes    Size of the total "extra" block
          CRC        4 bytes    32-bit CRC for remainder of the block
          Tag1       2 bytes    OpenVMS attribute tag value #1
          Size1      2 bytes    Size of attribute #1, in bytes
          (var.)     Size1      Attribute #1 data
          .
          .
          .
          TagN       2 bytes    OpenVMS attribute tag value #N
          SizeN      2 bytes    Size of attribute #N, in bytes
          (var.)     SizeN      Attribute #N data

          Rules:

          1. There will be one or more of attributes present, which
             will each be preceded by the above TagX & SizeX values. 
             These values are identical to the ATR$C_XXXX and
             ATR$S_XXXX constants which are defined in ATR.H under
             OpenVMS C.  Neither of these values will ever be zero.

          2. No word alignment or padding is performed.

          3. A well-behaved PKZIP/OpenVMS program should never produce
             more than one sub-block with the same TagX value.  Also,
             there will never be more than one "extra" block of type
             0x000c in a particular directory record.

         -UNIX Extra Field (0x000d):

          The following is the layout of the UNIX "extra" block.
          Note: all fields are stored in Intel low-byte/high-byte
          order.

          Value       Size          Description
          -----       ----          -----------
  (UNIX)  0x000d      2 bytes       Tag for this "extra" block type
          TSize       2 bytes       Size for the following data block
          Atime       4 bytes       File last access time
          Mtime       4 bytes       File last modification time
          Uid         2 bytes       File user ID
          Gid         2 bytes       File group ID
          (var)       variable      Variable length data field

          The variable length data field will contain file type
          specific data.  Currently the only values allowed are
          the original "linked to" file names for hard or symbolic
          links, and the major and minor device node numbers for
          character and block device nodes.  Since device nodes
          cannot be either symbolic or hard links, only one set of
          variable length data is stored.  Link files will have the
          name of the original file stored.  This name is NOT NULL
          terminated.  Its size can be determined by checking TSize -
          12.  Device entries will have eight bytes stored as two 4
          byte entries (in little endian format).  The first entry
          will be the major device number, and the second the minor
          device number.
         
         -PATCH Descriptor Extra Field (0x000f):

          The following is the layout of the Patch Descriptor "extra"
          block.

          Note: all fields stored in Intel low-byte/high-byte order.

          Value     Size     Description
          -----     ----     -----------
  (Patch) 0x000f    2 bytes  Tag for this "extra" block type
          TSize     2 bytes  Size of the total "extra" block
          Version   2 bytes  Version of the descriptor
          Flags     4 bytes  Actions and reactions (see below)
          OldSize   4 bytes  Size of the file about to be patched
          OldCRC    4 bytes  32-bit CRC of the file to be patched
          NewSize   4 bytes  Size of the resulting file
          NewCRC    4 bytes  32-bit CRC of the resulting file

          Actions and reactions

          Bits          Description
          ----          ----------------
          0             Use for auto detection
          1             Treat as a self-patch
          2-3           RESERVED
          4-5           Action (see below)
          6-7           RESERVED
          8-9           Reaction (see below) to absent file
          10-11         Reaction (see below) to newer file
          12-13         Reaction (see below) to unknown file
          14-15         RESERVED
          16-31         RESERVED

          Actions

          Action       Value
          ------       -----
          none         0
          add          1
          delete       2
          patch        3

          Reactions
 
          Reaction     Value
          --------     -----
          ask          0
          skip         1
          ignore       2
          fail         3

          Patch support is provided by PKPatchMaker(tm) technology and is
          covered under U.S. Patents and Patents Pending. The use or
          implementation in a product of certain technological aspects set
          forth in the current APPNOTE, including those with regard to
          strong encryption, patching, or extended tape operations requires
          a license from PKWARE.  Please contact PKWARE with regard to
          acquiring a license.

         -PKCS#7 Store for X.509 Certificates (0x0014):

          This field contains information about each of the certificates
          files may be signed with. When the Central Directory Encryption
          feature is enabled for a ZIP file, this record will appear in
          the Archive Extra Data Record, otherwise it will appear in the
          first central directory record and will be ignored in any
          other record.
         
          Note: all fields stored in Intel low-byte/high-byte order.

          Value     Size     Description
          -----     ----     -----------
  (Store) 0x0014    2 bytes  Tag for this "extra" block type
          TSize     2 bytes  Size of the store data
          TData     TSize    Data about the store


         -X.509 Certificate ID and Signature for individual file (0x0015):

          This field contains the information about which certificate in
          the PKCS#7 store was used to sign a particular file. It also
          contains the signature data. This field can appear multiple
          times, but can only appear once per certificate.

          Note: all fields stored in Intel low-byte/high-byte order.

          Value     Size     Description
          -----     ----     -----------
  (CID)   0x0015    2 bytes  Tag for this "extra" block type
          TSize     2 bytes  Size of data that follows
          TData     TSize    Signature Data

         -X.509 Certificate ID and Signature for central directory (0x0016):

          This field contains the information about which certificate in
          the PKCS#7 store was used to sign the central directory structure.
          When the Central Directory Encryption feature is enabled for a
          ZIP file, this record will appear in the Archive Extra Data Record,
          otherwise it will appear in the first central directory record.

          Note: all fields stored in Intel low-byte/high-byte order.

          Value     Size     Description
          -----     ----     -----------
  (CDID)  0x0016    2 bytes  Tag for this "extra" block type
          TSize     2 bytes  Size of data that follows
          TData     TSize    Data

         -Strong Encryption Header (0x0017):

          Value     Size     Description
          -----     ----     -----------
          0x0017    2 bytes  Tag for this "extra" block type
          TSize     2 bytes  Size of data that follows
          Format    2 bytes  Format definition for this record
          AlgID     2 bytes  Encryption algorithm identifier
          Bitlen    2 bytes  Bit length of encryption key
          Flags     2 bytes  Processing flags
          CertData  TSize-8  Certificate decryption extra field data
                             (refer to the explanation for CertData
                              in the section describing the
                              Certificate Processing Method under
                              the Strong Encryption Specification)


         -Record Management Controls (0x0018):

          Value     Size     Description
          -----     ----     -----------
(Rec-CTL) 0x0018    2 bytes  Tag for this "extra" block type
          CSize     2 bytes  Size of total extra block data
          Tag1      2 bytes  Record control attribute 1
          Size1     2 bytes  Size of attribute 1, in bytes
          Data1     Size1    Attribute 1 data
            .
            .
            .
          TagN      2 bytes  Record control attribute N
          SizeN     2 bytes  Size of attribute N, in bytes
          DataN     SizeN    Attribute N data


         -PKCS#7 Encryption Recipient Certificate List (0x0019):

          This field contains information about each of the certificates
          used in encryption processing and it can be used to identify who is
          allowed to decrypt encrypted files.  This field should only appear
          in the archive extra data record. This field is not required and
          serves only to aide archive modifications by preserving public
          encryption key data. Individual security requirements may dictate
          that this data be omitted to deter information exposure.

          Note: all fields stored in Intel low-byte/high-byte order.

          Value     Size     Description
          -----     ----     -----------
 (CStore) 0x0019    2 bytes  Tag for this "extra" block type
          TSize     2 bytes  Size of the store data
          TData     TSize    Data about the store

          TData:

          Value     Size     Description
          -----     ----     -----------
          Version   2 bytes  Format version number - must 0x0001 at this time
          CStore    (var)    PKCS#7 data blob


         -MVS Extra Field (0x0065):

          The following is the layout of the MVS "extra" block.
          Note: Some fields are stored in Big Endian format.
          All text is in EBCDIC format unless otherwise specified.

          Value       Size          Description
          -----       ----          -----------
  (MVS)   0x0065      2 bytes       Tag for this "extra" block type
          TSize       2 bytes       Size for the following data block
          ID          4 bytes       EBCDIC "Z390" 0xE9F3F9F0 or
                                    "T4MV" for TargetFour
          (var)       TSize-4       Attribute data (see APPENDIX B)


         -OS/400 Extra Field (0x0065):

          The following is the layout of the OS/400 "extra" block.
          Note: Some fields are stored in Big Endian format.
          All text is in EBCDIC format unless otherwise specified.

          Value       Size          Description
          -----       ----          -----------
  (OS400) 0x0065      2 bytes       Tag for this "extra" block type
          TSize       2 bytes       Size for the following data block
          ID          4 bytes       EBCDIC "I400" 0xC9F4F0F0 or
                                    "T4MV" for TargetFour
          (var)       TSize-4       Attribute data (see APPENDIX A)


          Third-party Mappings:
         
         -ZipIt Macintosh Extra Field (long) (0x2605):

          The following is the layout of the ZipIt extra block
          for Macintosh. The local-header and central-header versions
          are identical. This block must be present if the file is
          stored MacBinary-encoded and it should not be used if the file
          is not stored MacBinary-encoded.

          Value         Size        Description
          -----         ----        -----------
  (Mac2)  0x2605        Short       tag for this extra block type
          TSize         Short       total data size for this block
          "ZPIT"        beLong      extra-field signature
          FnLen         Byte        length of FileName
          FileName      variable    full Macintosh filename
          FileType      Byte[4]     four-byte Mac file type string
          Creator       Byte[4]     four-byte Mac creator string


         -ZipIt Macintosh Extra Field (short, for files) (0x2705):

          The following is the layout of a shortened variant of the
          ZipIt extra block for Macintosh (without "full name" entry).
          This variant is used by ZipIt 1.3.5 and newer for entries of
          files (not directories) that do not have a MacBinary encoded
          file. The local-header and central-header versions are identical.

          Value         Size        Description
          -----         ----        -----------
  (Mac2b) 0x2705        Short       tag for this extra block type
          TSize         Short       total data size for this block (12)
          "ZPIT"        beLong      extra-field signature
          FileType      Byte[4]     four-byte Mac file type string
          Creator       Byte[4]     four-byte Mac creator string
          fdFlags       beShort     attributes from FInfo.frFlags,
                                    may be omitted
          0x0000        beShort     reserved, may be omitted


         -ZipIt Macintosh Extra Field (short, for directories) (0x2805):

          The following is the layout of a shortened variant of the
          ZipIt extra block for Macintosh used only for directory
          entries. This variant is used by ZipIt 1.3.5 and newer to
          save some optional Mac-specific information about directories.
          The local-header and central-header versions are identical.

          Value         Size        Description
          -----         ----        -----------
  (Mac2c) 0x2805        Short       tag for this extra block type
          TSize         Short       total data size for this block (12)
          "ZPIT"        beLong      extra-field signature
          frFlags       beShort     attributes from DInfo.frFlags, may
                                    be omitted
          View          beShort     ZipIt view flag, may be omitted


          The View field specifies ZipIt-internal settings as follows:

          Bits of the Flags:
              bit 0           if set, the folder is shown expanded (open)
                              when the archive contents are viewed in ZipIt.
              bits 1-15       reserved, zero;


         -FWKCS MD5 Extra Field (0x4b46):

          The FWKCS Contents_Signature System, used in
          automatically identifying files independent of file name,
          optionally adds and uses an extra field to support the
          rapid creation of an enhanced contents_signature:

              Header ID = 0x4b46
              Data Size = 0x0013
              Preface   = 'M','D','5'
              followed by 16 bytes containing the uncompressed file's
              128_bit MD5 hash(1), low byte first.

          When FWKCS revises a .ZIP file central directory to add
          this extra field for a file, it also replaces the
          central directory entry for that file's uncompressed
          file length with a measured value.

          FWKCS provides an option to strip this extra field, if
          present, from a .ZIP file central directory. In adding
          this extra field, FWKCS preserves .ZIP file Authenticity
          Verification; if stripping this extra field, FWKCS
          preserves all versions of AV through PKZIP version 2.04g.

          FWKCS, and FWKCS Contents_Signature System, are
          trademarks of Frederick W. Kantor.

          (1) R. Rivest, RFC1321.TXT, MIT Laboratory for Computer
              Science and RSA Data Security, Inc., April 1992.
              ll.76-77: "The MD5 algorithm is being placed in the
              public domain for review and possible adoption as a
              standard."


         -Info-ZIP Unicode Comment Extra Field (0x6375):

          Stores the UTF-8 version of the file comment as stored in the
          central directory header. (Last Revision 20070912)

          Value         Size        Description
          -----         ----        -----------
   (UCom) 0x6375        Short       tag for this extra block type ("uc")
          TSize         Short       total data size for this block
          Version       1 byte      version of this extra field, currently 1
          ComCRC32      4 bytes     Comment Field CRC32 Checksum
          UnicodeCom    Variable    UTF-8 version of the entry comment

          Currently Version is set to the number 1.  If there is a need
          to change this field, the version will be incremented.  Changes
          may not be backward compatible so this extra field should not be
          used if the version is not recognized.

          The ComCRC32 is the standard zip CRC32 checksum of the File Comment
          field in the central directory header.  This is used to verify that
          the comment field has not changed since the Unicode Comment extra field
          was created.  This can happen if a utility changes the File Comment
          field but does not update the UTF-8 Comment extra field.  If the CRC
          check fails, this Unicode Comment extra field should be ignored and
          the File Comment field in the header should be used instead.

          The UnicodeCom field is the UTF-8 version of the File Comment field
          in the header.  As UnicodeCom is defined to be UTF-8, no UTF-8 byte
          order mark (BOM) is used.  The length of this field is determined by
          subtracting the size of the previous fields from TSize.  If both the
          File Name and Comment fields are UTF-8, the new General Purpose Bit
          Flag, bit 11 (Language encoding flag (EFS)), can be used to indicate
          both the header File Name and Comment fields are UTF-8 and, in this
          case, the Unicode Path and Unicode Comment extra fields are not
          needed and should not be created.  Note that, for backward
          compatibility, bit 11 should only be used if the native character set
          of the paths and comments being zipped up are already in UTF-8. It is
          expected that the same file comment storage method, either general
          purpose bit 11 or extra fields, be used in both the Local and Central
          Directory Header for a file.


         -Info-ZIP Unicode Path Extra Field (0x7075):

          Stores the UTF-8 version of the file name field as stored in the
          local header and central directory header. (Last Revision 20070912)

          Value         Size        Description
          -----         ----        -----------
  (UPath) 0x7075        Short       tag for this extra block type ("up")
          TSize         Short       total data size for this block
          Version       1 byte      version of this extra field, currently 1
          NameCRC32     4 bytes     File Name Field CRC32 Checksum
          UnicodeName   Variable    UTF-8 version of the entry File Name

          Currently Version is set to the number 1.  If there is a need
          to change this field, the version will be incremented.  Changes
          may not be backward compatible so this extra field should not be
          used if the version is not recognized.

          The NameCRC32 is the standard zip CRC32 checksum of the File Name
          field in the header.  This is used to verify that the header
          File Name field has not changed since the Unicode Path extra field
          was created.  This can happen if a utility renames the File Name but
          does not update the UTF-8 path extra field.  If the CRC check fails,
          this UTF-8 Path Extra Field should be ignored and the File Name field
          in the header should be used instead.

          The UnicodeName is the UTF-8 version of the contents of the File Name
          field in the header.  As UnicodeName is defined to be UTF-8, no UTF-8
          byte order mark (BOM) is used.  The length of this field is determined
          by subtracting the size of the previous fields from TSize.  If both
          the File Name and Comment fields are UTF-8, the new General Purpose
          Bit Flag, bit 11 (Language encoding flag (EFS)), can be used to
          indicate that both the header File Name and Comment fields are UTF-8
          and, in this case, the Unicode Path and Unicode Comment extra fields
          are not needed and should not be created.  Note that, for backward
          compatibility, bit 11 should only be used if the native character set
          of the paths and comments being zipped up are already in UTF-8. It is
          expected that the same file name storage method, either general
          purpose bit 11 or extra fields, be used in both the Local and Central
          Directory Header for a file.
 

        -Microsoft Open Packaging Growth Hint (0xa220):

          Value         Size        Description
          -----         ----        -----------
          0xa220        Short       tag for this extra block type
          TSize         Short       size of Sig + PadVal + Padding
          Sig           Short       verification signature (A028)
          PadVal        Short       Initial padding value
          Padding       variable    filled with NULL characters


      file comment: (Variable)

          The comment for this file.

      number of this disk: (2 bytes)

          The number of this disk, which contains central
          directory end record. If an archive is in ZIP64 format
          and the value in this field is 0xFFFF, the size will
          be in the corresponding 4 byte zip64 end of central
          directory field.


      number of the disk with the start of the central
      directory: (2 bytes)

          The number of the disk on which the central
          directory starts. If an archive is in ZIP64 format
          and the value in this field is 0xFFFF, the size will
          be in the corresponding 4 byte zip64 end of central
          directory field.

      total number of entries in the central dir on
      this disk: (2 bytes)

          The number of central directory entries on this disk.
          If an archive is in ZIP64 format and the value in
          this field is 0xFFFF, the size will be in the
          corresponding 8 byte zip64 end of central
          directory field.

      total number of entries in the central dir: (2 bytes)

          The total number of files in the .ZIP file. If an
          archive is in ZIP64 format and the value in this field
          is 0xFFFF, the size will be in the corresponding 8 byte
          zip64 end of central directory field.

      size of the central directory: (4 bytes)

          The size (in bytes) of the entire central directory.
          If an archive is in ZIP64 format and the value in
          this field is 0xFFFFFFFF, the size will be in the
          corresponding 8 byte zip64 end of central
          directory field.

      offset of start of central directory with respect to
      the starting disk number:  (4 bytes)

          Offset of the start of the central directory on the
          disk on which the central directory starts. If an
          archive is in ZIP64 format and the value in this
          field is 0xFFFFFFFF, the size will be in the
          corresponding 8 byte zip64 end of central
          directory field.

      .ZIP file comment length: (2 bytes)

          The length of the comment for this .ZIP file.

      .ZIP file comment: (Variable)

          The comment for this .ZIP file.  ZIP file comment data
          is stored unsecured.  No encryption or data authentication
          is applied to this area at this time.  Confidential information
          should not be stored in this section.

      zip64 extensible data sector    (variable size)

          (currently reserved for use by PKWARE)


  K.  Splitting and Spanning ZIP files

          Spanning is the process of segmenting a ZIP file across
          multiple removable media. This support has typically only
          been provided for DOS formatted floppy diskettes.

          File splitting is a newer derivative of spanning. 
          Splitting follows the same segmentation process as
          spanning, however, it does not require writing each
          segment to a unique removable medium and instead supports
          placing all pieces onto local or non-removable locations
          such as file systems, local drives, folders, etc...

          A key difference between spanned and split ZIP files is
          that all pieces of a spanned ZIP file have the same name. 
          Since each piece is written to a separate volume, no name
          collisions occur and each segment can reuse the original
          .ZIP file name given to the archive.

          Sequence ordering for DOS spanned archives uses the DOS
          volume label to determine segment numbers.  Volume labels
          for each segment are written using the form PKBACK#xxx,
          where xxx is the segment number written as a decimal
          value from 001 - nnn.

          Split ZIP files are typically written to the same location
          and are subject to name collisions if the spanned name
          format is used since each segment will reside on the same
          drive. To avoid name collisions, split archives are named
          as follows.

          Segment 1   = filename.z01
          Segment n-1 = filename.z(n-1)
          Segment n   = filename.zip

          The .ZIP extension is used on the last segment to support
          quickly reading the central directory.  The segment number
          n should be a decimal value.

          Spanned ZIP files may be PKSFX Self-extracting ZIP files.
          PKSFX files may also be split, however, in this case
          the first segment must be named filename.exe.  The first
          segment of a split PKSFX archive must be large enough to
          include the entire executable program.

          Capacities for split archives are as follows.

          Maximum number of segments = 4,294,967,295 - 1
          Maximum .ZIP segment size = 4,294,967,295 bytes
          Minimum segment size = 64K
          Maximum PKSFX segment size = 2,147,483,647 bytes
         
          Segment sizes may be different however by convention, all
          segment sizes should be the same with the exception of the
          last, which may be smaller.  Local and central directory
          header records must never be split across a segment boundary.
          When writing a header record, if the number of bytes remaining
          within a segment is less than the size of the header record,
          end the current segment and write the header at the start
          of the next segment.  The central directory may span segment
          boundaries, but no single record in the central directory
          should be split across segments.

          Spanned/Split archives created using PKZIP for Windows
          (V2.50 or greater), PKZIP Command Line (V2.50 or greater),
          or PKZIP Explorer will include a special spanning
          signature as the first 4 bytes of the first segment of
          the archive.  This signature (0x08074b50) will be
          followed immediately by the local header signature for
          the first file in the archive. 

          A special spanning marker may also appear in spanned/split
          archives if the spanning or splitting process starts but
          only requires one segment.  In this case the 0x08074b50
          signature will be replaced with the temporary spanning
          marker signature of 0x30304b50.  Split archives can
          only be uncompressed by other versions of PKZIP that
          know how to create a split archive.

          The signature value 0x08074b50 is also used by some
          ZIP implementations as a marker for the Data Descriptor
          record.  Conflict in this alternate assignment can be
          avoided by ensuring the position of the signature
          within the ZIP file to determine the use for which it
          is intended. 

  L.  General notes:

      1)  All fields unless otherwise noted are unsigned and stored
          in Intel low-byte:high-byte, low-word:high-word order.

      2)  String fields are not null terminated, since the
          length is given explicitly.

      3)  The entries in the central directory may not necessarily
          be in the same order that files appear in the .ZIP file.

      4)  If one of the fields in the end of central directory
          record is too small to hold required data, the field
          should be set to -1 (0xFFFF or 0xFFFFFFFF) and the
          ZIP64 format record should be created.

      5)  The end of central directory record and the
          Zip64 end of central directory locator record must
          reside on the same disk when splitting or spanning
          an archive.

VI. Explanation of compression methods
--------------------------------------

UnShrinking - Method 1
----------------------

Shrinking is a Dynamic Ziv-Lempel-Welch compression algorithm
with partial clearing.  The initial code size is 9 bits, and
the maximum code size is 13 bits.  Shrinking differs from
conventional Dynamic Ziv-Lempel-Welch implementations in several
respects:

1)  The code size is controlled by the compressor, and is not
    automatically increased when codes larger than the current
    code size are created (but not necessarily used).  When
    the decompressor encounters the code sequence 256
    (decimal) followed by 1, it should increase the code size
    read from the input stream to the next bit size.  No
    blocking of the codes is performed, so the next code at
    the increased size should be read from the input stream
    immediately after where the previous code at the smaller
    bit size was read.  Again, the decompressor should not
    increase the code size used until the sequence 256,1 is
    encountered.

2)  When the table becomes full, total clearing is not
    performed.  Rather, when the compressor emits the code
    sequence 256,2 (decimal), the decompressor should clear
    all leaf nodes from the Ziv-Lempel tree, and continue to
    use the current code size.  The nodes that are cleared
    from the Ziv-Lempel tree are then re-used, with the lowest
    code value re-used first, and the highest code value
    re-used last.  The compressor can emit the sequence 256,2
    at any time.

Expanding - Methods 2-5
-----------------------

The Reducing algorithm is actually a combination of two
distinct algorithms.  The first algorithm compresses repeated
byte sequences, and the second algorithm takes the compressed
stream from the first algorithm and applies a probabilistic
compression method.

The probabilistic compression stores an array of 'follower
sets' S(j), for j=0 to 255, corresponding to each possible
ASCII character.  Each set contains between 0 and 32
characters, to be denoted as S(j)[0],...,S(j)[m], where m<32.
The sets are stored at the beginning of the data area for a
Reduced file, in reverse order, with S(255) first, and S(0)
last.

The sets are encoded as { N(j), S(j)[0],...,S(j)[N(j)-1] },
where N(j) is the size of set S(j).  N(j) can be 0, in which
case the follower set for S(j) is empty.  Each N(j) value is
encoded in 6 bits, followed by N(j) eight bit character values
corresponding to S(j)[0] to S(j)[N(j)-1] respectively.  If
N(j) is 0, then no values for S(j) are stored, and the value
for N(j-1) immediately follows.

Immediately after the follower sets, is the compressed data
stream.  The compressed data stream can be interpreted for the
probabilistic decompression as follows:

let Last-Character <- 0.
loop until done
    if the follower set S(Last-Character) is empty then
        read 8 bits from the input stream, and copy this
        value to the output stream.
    otherwise if the follower set S(Last-Character) is non-empty then
        read 1 bit from the input stream.
        if this bit is not zero then
            read 8 bits from the input stream, and copy this
            value to the output stream.
        otherwise if this bit is zero then
            read B(N(Last-Character)) bits from the input
            stream, and assign this value to I.
            Copy the value of S(Last-Character)[I] to the
            output stream.

    assign the last value placed on the output stream to
    Last-Character.
end loop

B(N(j)) is defined as the minimal number of bits required to
encode the value N(j)-1.

The decompressed stream from above can then be expanded to
re-create the original file as follows:

let State <- 0.

loop until done
    read 8 bits from the input stream into C.
    case State of
        0:  if C is not equal to DLE (144 decimal) then
                copy C to the output stream.
            otherwise if C is equal to DLE then
                let State <- 1.

        1:  if C is non-zero then
                let V <- C.
                let Len <- L(V)
                let State <- F(Len).
            otherwise if C is zero then
                copy the value 144 (decimal) to the output stream.
                let State <- 0

        2:  let Len <- Len + C
            let State <- 3.

        3:  move backwards D(V,C) bytes in the output stream
            (if this position is before the start of the output
            stream, then assume that all the data before the
            start of the output stream is filled with zeros).
            copy Len+3 bytes from this position to the output stream.
            let State <- 0.
    end case
end loop

The functions F,L, and D are dependent on the 'compression
factor', 1 through 4, and are defined as follows:

For compression factor 1:
    L(X) equals the lower 7 bits of X.
    F(X) equals 2 if X equals 127 otherwise F(X) equals 3.
    D(X,Y) equals the (upper 1 bit of X) * 256 + Y + 1.
For compression factor 2:
    L(X) equals the lower 6 bits of X.
    F(X) equals 2 if X equals 63 otherwise F(X) equals 3.
    D(X,Y) equals the (upper 2 bits of X) * 256 + Y + 1.
For compression factor 3:
    L(X) equals the lower 5 bits of X.
    F(X) equals 2 if X equals 31 otherwise F(X) equals 3.
    D(X,Y) equals the (upper 3 bits of X) * 256 + Y + 1.
For compression factor 4:
    L(X) equals the lower 4 bits of X.
    F(X) equals 2 if X equals 15 otherwise F(X) equals 3.
    D(X,Y) equals the (upper 4 bits of X) * 256 + Y + 1.

Imploding - Method 6
--------------------

The Imploding algorithm is actually a combination of two distinct
algorithms.  The first algorithm compresses repeated byte
sequences using a sliding dictionary.  The second algorithm is
used to compress the encoding of the sliding dictionary output,
using multiple Shannon-Fano trees.

The Imploding algorithm can use a 4K or 8K sliding dictionary
size. The dictionary size used can be determined by bit 1 in the
general purpose flag word; a 0 bit indicates a 4K dictionary
while a 1 bit indicates an 8K dictionary.

The Shannon-Fano trees are stored at the start of the compressed
file. The number of trees stored is defined by bit 2 in the
general purpose flag word; a 0 bit indicates two trees stored, a
1 bit indicates three trees are stored.  If 3 trees are stored,
the first Shannon-Fano tree represents the encoding of the
Literal characters, the second tree represents the encoding of
the Length information, the third represents the encoding of the
Distance information.  When 2 Shannon-Fano trees are stored, the
Length tree is stored first, followed by the Distance tree.

The Literal Shannon-Fano tree, if present is used to represent
the entire ASCII character set, and contains 256 values.  This
tree is used to compress any data not compressed by the sliding
dictionary algorithm.  When this tree is present, the Minimum
Match Length for the sliding dictionary is 3.  If this tree is
not present, the Minimum Match Length is 2.

The Length Shannon-Fano tree is used to compress the Length part
of the (length,distance) pairs from the sliding dictionary
output.  The Length tree contains 64 values, ranging from the
Minimum Match Length, to 63 plus the Minimum Match Length.

The Distance Shannon-Fano tree is used to compress the Distance
part of the (length,distance) pairs from the sliding dictionary
output. The Distance tree contains 64 values, ranging from 0 to
63, representing the upper 6 bits of the distance value.  The
distance values themselves will be between 0 and the sliding
dictionary size, either 4K or 8K.

The Shannon-Fano trees themselves are stored in a compressed
format. The first byte of the tree data represents the number of
bytes of data representing the (compressed) Shannon-Fano tree
minus 1.  The remaining bytes represent the Shannon-Fano tree
data encoded as:

    High 4 bits: Number of values at this bit length + 1. (1 - 16)
    Low  4 bits: Bit Length needed to represent value + 1. (1 - 16)

The Shannon-Fano codes can be constructed from the bit lengths
using the following algorithm:

1)  Sort the Bit Lengths in ascending order, while retaining the
    order of the original lengths stored in the file.

2)  Generate the Shannon-Fano trees:

    Code <- 0
    CodeIncrement <- 0
    LastBitLength <- 0
    i <- number of Shannon-Fano codes - 1   (either 255 or 63)

    loop while i >= 0
        Code = Code + CodeIncrement
        if BitLength(i) <> LastBitLength then
            LastBitLength=BitLength(i)
            CodeIncrement = 1 shifted left (16 - LastBitLength)
        ShannonCode(i) = Code
        i <- i - 1
    end loop

3)  Reverse the order of all the bits in the above ShannonCode()
    vector, so that the most significant bit becomes the least
    significant bit.  For example, the value 0x1234 (hex) would
    become 0x2C48 (hex).

4)  Restore the order of Shannon-Fano codes as originally stored
    within the file.

Example:

    This example will show the encoding of a Shannon-Fano tree
    of size 8.  Notice that the actual Shannon-Fano trees used
    for Imploding are either 64 or 256 entries in size.

Example:   0x02, 0x42, 0x01, 0x13

    The first byte indicates 3 values in this table.  Decoding the
    bytes:
            0x42 = 5 codes of 3 bits long
            0x01 = 1 code  of 2 bits long
            0x13 = 2 codes of 4 bits long

    This would generate the original bit length array of:
    (3, 3, 3, 3, 3, 2, 4, 4)

    There are 8 codes in this table for the values 0 thru 7.  Using
    the algorithm to obtain the Shannon-Fano codes produces:

                                  Reversed     Order     Original
Val  Sorted   Constructed Code      Value     Restored    Length
---  ------   -----------------   --------    --------    ------
0:     2      1100000000000000        11       101          3
1:     3      1010000000000000       101       001          3
2:     3      1000000000000000       001       110          3
3:     3      0110000000000000       110       010          3
4:     3      0100000000000000       010       100          3
5:     3      0010000000000000       100        11          2
6:     4      0001000000000000      1000      1000          4
7:     4      0000000000000000      0000      0000          4

The values in the Val, Order Restored and Original Length columns
now represent the Shannon-Fano encoding tree that can be used for
decoding the Shannon-Fano encoded data.  How to parse the
variable length Shannon-Fano values from the data stream is beyond
the scope of this document.  (See the references listed at the end of
this document for more information.)  However, traditional decoding
schemes used for Huffman variable length decoding, such as the
Greenlaw algorithm, can be successfully applied.

The compressed data stream begins immediately after the
compressed Shannon-Fano data.  The compressed data stream can be
interpreted as follows:

loop until done
    read 1 bit from input stream.

    if this bit is non-zero then       (encoded data is literal data)
        if Literal Shannon-Fano tree is present
            read and decode character using Literal Shannon-Fano tree.
        otherwise
            read 8 bits from input stream.
        copy character to the output stream.
    otherwise              (encoded data is sliding dictionary match)
        if 8K dictionary size
            read 7 bits for offset Distance (lower 7 bits of offset).
        otherwise
            read 6 bits for offset Distance (lower 6 bits of offset).

        using the Distance Shannon-Fano tree, read and decode the
          upper 6 bits of the Distance value.

        using the Length Shannon-Fano tree, read and decode
          the Length value.

        Length <- Length + Minimum Match Length

        if Length = 63 + Minimum Match Length
            read 8 bits from the input stream,
            add this value to Length.

        move backwards Distance+1 bytes in the output stream, and
        copy Length characters from this position to the output
        stream.  (if this position is before the start of the output
        stream, then assume that all the data before the start of
        the output stream is filled with zeros).
end loop

Tokenizing - Method 7
---------------------

This method is not used by PKZIP.

Deflating - Method 8
--------------------

The Deflate algorithm is similar to the Implode algorithm using
a sliding dictionary of up to 32K with secondary compression
from Huffman/Shannon-Fano codes.

The compressed data is stored in blocks with a header describing
the block and the Huffman codes used in the data block.  The header
format is as follows:

   Bit 0: Last Block bit     This bit is set to 1 if this is the last
                             compressed block in the data.
   Bits 1-2: Block type
      00 (0) - Block is stored - All stored data is byte aligned.
               Skip bits until next byte, then next word = block
               length, followed by the ones compliment of the block
               length word. Remaining data in block is the stored
               data.

      01 (1) - Use fixed Huffman codes for literal and distance codes.
               Lit Code    Bits             Dist Code   Bits
               ---------   ----             ---------   ----
                 0 - 143    8                 0 - 31      5
               144 - 255    9
               256 - 279    7
               280 - 287    8

               Literal codes 286-287 and distance codes 30-31 are
               never used but participate in the huffman construction.

      10 (2) - Dynamic Huffman codes.  (See expanding Huffman codes)

      11 (3) - Reserved - Flag a "Error in compressed data" if seen.

Expanding Huffman Codes
-----------------------
If the data block is stored with dynamic Huffman codes, the Huffman
codes are sent in the following compressed format:

   5 Bits: # of Literal codes sent - 256 (256 - 286)
           All other codes are never sent.
   5 Bits: # of Dist codes - 1           (1 - 32)
   4 Bits: # of Bit Length codes - 3     (3 - 19)

The Huffman codes are sent as bit lengths and the codes are built as
described in the implode algorithm.  The bit lengths themselves are
compressed with Huffman codes.  There are 19 bit length codes:

   0 - 15: Represent bit lengths of 0 - 15
       16: Copy the previous bit length 3 - 6 times.
           The next 2 bits indicate repeat length (0 = 3, ... ,3 = 6)
              Example:  Codes 8, 16 (+2 bits 11), 16 (+2 bits 10) will
                        expand to 12 bit lengths of 8 (1 + 6 + 5)
       17: Repeat a bit length of 0 for 3 - 10 times. (3 bits of length)
       18: Repeat a bit length of 0 for 11 - 138 times (7 bits of length)

The lengths of the bit length codes are sent packed 3 bits per value
(0 - 7) in the following order:

   16, 17, 18, 0, 8, 7, 9, 6, 10, 5, 11, 4, 12, 3, 13, 2, 14, 1, 15

The Huffman codes should be built as described in the Implode algorithm
except codes are assigned starting at the shortest bit length, i.e. the
shortest code should be all 0's rather than all 1's.  Also, codes with
a bit length of zero do not participate in the tree construction.  The
codes are then used to decode the bit lengths for the literal and
distance tables.

The bit lengths for the literal tables are sent first with the number
of entries sent described by the 5 bits sent earlier.  There are up
to 286 literal characters; the first 256 represent the respective 8
bit character, code 256 represents the End-Of-Block code, the remaining
29 codes represent copy lengths of 3 thru 258.  There are up to 30
distance codes representing distances from 1 thru 32k as described
below.

                             Length Codes
                             ------------
      Extra             Extra              Extra              Extra
 Code Bits Length  Code Bits Lengths  Code Bits Lengths  Code Bits Length(s)
 ---- ---- ------  ---- ---- -------  ---- ---- -------  ---- ---- ---------
  257   0     3     265   1   11,12    273   3   35-42    281   5  131-162
  258   0     4     266   1   13,14    274   3   43-50    282   5  163-194
  259   0     5     267   1   15,16    275   3   51-58    283   5  195-226
  260   0     6     268   1   17,18    276   3   59-66    284   5  227-257
  261   0     7     269   2   19-22    277   4   67-82    285   0    258
  262   0     8     270   2   23-26    278   4   83-98
  263   0     9     271   2   27-30    279   4   99-114
  264   0    10     272   2   31-34    280   4  115-130

                            Distance Codes
                            --------------
      Extra           Extra             Extra               Extra
 Code Bits Dist  Code Bits  Dist   Code Bits Distance  Code Bits Distance
 ---- ---- ----  ---- ---- ------  ---- ---- --------  ---- ---- --------
   0   0    1      8   3   17-24    16    7  257-384    24   11  4097-6144
   1   0    2      9   3   25-32    17    7  385-512    25   11  6145-8192
   2   0    3     10   4   33-48    18    8  513-768    26   12  8193-12288
   3   0    4     11   4   49-64    19    8  769-1024   27   12 12289-16384
   4   1   5,6    12   5   65-96    20    9 1025-1536   28   13 16385-24576
   5   1   7,8    13   5   97-128   21    9 1537-2048   29   13 24577-32768
   6   2   9-12   14   6  129-192   22   10 2049-3072
   7   2  13-16   15   6  193-256   23   10 3073-4096

The compressed data stream begins immediately after the
compressed header data.  The compressed data stream can be
interpreted as follows:

do
   read header from input stream.

   if stored block
      skip bits until byte aligned
      read count and 1's compliment of count
      copy count bytes data block
   otherwise
      loop until end of block code sent
         decode literal character from input stream
         if literal < 256
            copy character to the output stream
         otherwise
            if literal = end of block
               break from loop
            otherwise
               decode distance from input stream

               move backwards distance bytes in the output stream, and
               copy length characters from this position to the output
               stream.
      end loop
while not last block

if data descriptor exists
   skip bits until byte aligned
   read crc and sizes
endif

Enhanced Deflating - Method 9
-----------------------------

The Enhanced Deflating algorithm is similar to Deflate but
uses a sliding dictionary of up to 64K. Deflate64(tm) is supported
by the Deflate extractor.

BZIP2 - Method 12
-----------------

BZIP2 is an open-source data compression algorithm developed by
Julian Seward.  Information and source code for this algorithm
can be found on the internet.

LZMA - Method 14 (EFS)
----------------------

LZMA is a block-oriented, general purpose data compression algorithm 
developed and maintained by Igor Pavlov.  It is a derivative of LZ77
that utilizes Markov chains and a range coder.  Information and
source code for this algorithm can be found on the internet.  Consult
with the author of this algorithm for information on terms or
restrictions on use.

Support for LZMA within the ZIP format is defined as follows:  

The Compression method field within the ZIP Local and Central
Header records will be set to the value 14 to indicate data was
compressed using LZMA.

The Version needed to extract field within the ZIP Local and
Central Header records will be set to 6.3 to indicate the
minimum ZIP format version supporting this feature.

File data compressed using the LZMA algorithm must be placed
immediately following the Local Header for the file.  If a
standard ZIP encryption header is required, it will follow
the Local Header and will precede the LZMA compressed file
data segment.  The location of LZMA compressed data segment
within the ZIP format will be as shown:

    [local header file 1]
    [encryption header file 1]
    [LZMA compressed data segment for file 1]
    [data descriptor 1]
    [local header file 2]

The encryption header and data descriptor records may
be conditionally present.  The LZMA Compressed Data Segment
will consist of an LZMA Properties Header followed by the
LZMA Compressed Data as shown:

    [LZMA properties header for file 1]
    [LZMA compressed data for file 1]

The LZMA Compressed Data will be stored as provided by the
LZMA compression library.  Compressed size, uncompressed
size and other file characteristics about the file being
compressed must be stored in standard ZIP storage format.

The LZMA Properties Header will store specific data required to
decompress the LZMA compressed Data.  This data is set by the
LZMA compression engine using the function WriteCoderProperties()
as documented within the LZMA SDK.
 
Storage fields for the property information within the LZMA
Properties Header are as follows:

     LZMA Version Information 2 bytes
     LZMA Properties Size 2 bytes
     LZMA Properties Data variable, defined by "LZMA Properties Size"

LZMA Version Information - this field identifies which version of
     the LZMA SDK was used to compress a file.  The first byte will
     store the major version number of the LZMA SDK and the second
     byte will store the minor number. 

LZMA Properties Size - this field defines the size of the remaining
     property data.  Typically this size should be determined by the
     version of the SDK.  This size field is included as a convenience
     and to help avoid any ambiguity should it arise in the future due
     to changes in this compression algorithm.

LZMA Property Data - this variable sized field records the required
     values for the decompressor as defined by the LZMA SDK.  The
     data stored in this field should be obtained using the
     WriteCoderProperties() in the version of the SDK defined by
     the "LZMA Version Information" field. 

The layout of the "LZMA Properties Data" field is a function of the
LZMA compression algorithm.  It is possible that this layout may be
changed by the author over time.  The data layout in version 4.32
of the LZMA SDK defines a 5 byte array that uses 4 bytes to store
the dictionary size in little-endian order. This is preceded by a
single packed byte as the first element of the array that contains
the following fields:

     PosStateBits
     LiteralPosStateBits
     LiteralContextBits

Refer to the LZMA documentation for a more detailed explanation of
these fields. 

Data compressed with method 14, LZMA, may include an end-of-stream
(EOS) marker ending the compressed data stream.  This marker is not
required, but its use is highly recommended to facilitate processing
and implementers should include the EOS marker whenever possible.
When the EOS marker is used, general purpose bit 1 must be set.  If
general purpose bit 1 is not set, the EOS marker is not present.

WavPack - Method 97
-------------------

Information describing the use of compression method 97 is
provided by WinZIP International, LLC.  This method relies on the
open source WavPack audio compression utility developed by David Bryant. 
Information on WavPack is available at www.wavpack.com.  Please consult
with the author of this algorithm for information on terms and
restrictions on use.

WavPack data for a file begins immediately after the end of the
local header data.  This data is the output from WavPack compression
routines.  Within the ZIP file, the use of WavPack compression is
indicated by setting the compression method field to a value of 97
in both the local header and the central directory header.  The Version
needed to extract and version made by fields use the same values as are
used for data compressed using the Deflate algorithm.

An implementation note for storing digital sample data when using
WavPack compression within ZIP files is that all of the bytes of
the sample data should be compressed.  This includes any unused
bits up to the byte boundary.  An example is a 2 byte sample that
uses only 12 bits for the sample data with 4 unused bits.  If only
12 bits are passed as the sample size to the WavPack routines, the 4
unused bits will be set to 0 on extraction regardless of their original
state.  To avoid this, the full 16 bits of the sample data size
should be provided.

PPMd - Method 98
----------------

PPMd is a data compression algorithm developed by Dmitry Shkarin
which includes a carryless rangecoder developed by Dmitry Subbotin.
This algorithm is based on predictive phrase matching on multiple
order contexts.  Information and source code for this algorithm
can be found on the internet. Consult with the author of this
algorithm for information on terms or restrictions on use.

Support for PPMd within the ZIP format currently is provided only
for version I, revision 1 of the algorithm.  Storage requirements
for using this algorithm are as follows:

Parameters needed to control the algorithm are stored in the two
bytes immediately preceding the compressed data.  These bytes are
used to store the following fields:

Model order - sets the maximum model order, default is 8, possible
              values are from 2 to 16 inclusive

Sub-allocator size - sets the size of sub-allocator in MB, default is 50,
            possible values are from 1MB to 256MB inclusive

Model restoration method - sets the method used to restart context
            model at memory insufficiency, values are:

            0 - restarts model from scratch - default
            1 - cut off model - decreases performance by as much as 2x
            2 - freeze context tree - not recommended

An example for packing these fields into the 2 byte storage field is
illustrated below.  These values are stored in Intel low-byte/high-byte
order.

wPPMd = (Model order - 1) +
        ((Sub-allocator size - 1) << 4) +
        (Model restoration method << 12)


VII. Traditional PKWARE Encryption
----------------------------------

The following information discusses the decryption steps
required to support traditional PKWARE encryption.  This
form of encryption is considered weak by today's standards
and its use is recommended only for situations with
low security needs or for compatibility with older .ZIP
applications.

Decryption
----------

PKWARE is grateful to Mr. Roger Schlafly for his expert contribution
towards the development of PKWARE's traditional encryption.

PKZIP encrypts the compressed data stream.  Encrypted files must
be decrypted before they can be extracted.

Each encrypted file has an extra 12 bytes stored at the start of
the data area defining the encryption header for that file.  The
encryption header is originally set to random values, and then
itself encrypted, using three, 32-bit keys.  The key values are
initialized using the supplied encryption password.  After each byte
is encrypted, the keys are then updated using pseudo-random number
generation techniques in combination with the same CRC-32 algorithm
used in PKZIP and described elsewhere in this document.

The following is the basic steps required to decrypt a file:

1) Initialize the three 32-bit keys with the password.
2) Read and decrypt the 12-byte encryption header, further
   initializing the encryption keys.
3) Read and decrypt the compressed data stream using the
   encryption keys.

Step 1 - Initializing the encryption keys
-----------------------------------------

Key(0) <- 305419896
Key(1) <- 591751049
Key(2) <- 878082192

loop for i <- 0 to length(password)-1
    update_keys(password(i))
end loop

Where update_keys() is defined as:

update_keys(char):
  Key(0) <- crc32(key(0),char)
  Key(1) <- Key(1) + (Key(0) & 000000ffH)
  Key(1) <- Key(1) * 134775813 + 1
  Key(2) <- crc32(key(2),key(1) >> 24)
end update_keys

Where crc32(old_crc,char) is a routine that given a CRC value and a
character, returns an updated CRC value after applying the CRC-32
algorithm described elsewhere in this document.

Step 2 - Decrypting the encryption header
-----------------------------------------

The purpose of this step is to further initialize the encryption
keys, based on random data, to render a plaintext attack on the
data ineffective.

Read the 12-byte encryption header into Buffer, in locations
Buffer(0) thru Buffer(11).

loop for i <- 0 to 11
    C <- buffer(i) ^ decrypt_byte()
    update_keys(C)
    buffer(i) <- C
end loop

Where decrypt_byte() is defined as:

unsigned char decrypt_byte()
    local unsigned short temp
    temp <- Key(2) | 2
    decrypt_byte <- (temp * (temp ^ 1)) >> 8
end decrypt_byte

After the header is decrypted,  the last 1 or 2 bytes in Buffer
should be the high-order word/byte of the CRC for the file being
decrypted, stored in Intel low-byte/high-byte order.  Versions of
PKZIP prior to 2.0 used a 2 byte CRC check; a 1 byte CRC check is
used on versions after 2.0.  This can be used to test if the password
supplied is correct or not.

Step 3 - Decrypting the compressed data stream
----------------------------------------------

The compressed data stream can be decrypted as follows:

loop until done
    read a character into C
    Temp <- C ^ decrypt_byte()
    update_keys(temp)
    output Temp
end loop


VIII. Strong Encryption Specification
-------------------------------------

The Strong Encryption technology defined in this specification is
covered under a pending patent application. The use or implementation
in a product of certain technological aspects set forth in the current
APPNOTE, including those with regard to strong encryption, patching,
or extended tape operations requires a license from PKWARE. Portions
of this Strong Encryption technology are available for use at no charge.
Contact PKWARE for licensing terms and conditions. Refer to section II
of this APPNOTE (Contacting PKWARE) for information on how to
contact PKWARE.

Version 5.x of this specification introduced support for strong
encryption algorithms.  These algorithms can be used with either
a password or an X.509v3 digital certificate to encrypt each file.
This format specification supports either password or certificate
based encryption to meet the security needs of today, to enable
interoperability between users within both PKI and non-PKI
environments, and to ensure interoperability between different
computing platforms that are running a ZIP program. 

Password based encryption is the most common form of encryption
people are familiar with.  However, inherent weaknesses with
passwords (e.g. susceptibility to dictionary/brute force attack)
as well as password management and support issues make certificate
based encryption a more secure and scalable option.  Industry
efforts and support are defining and moving towards more advanced
security solutions built around X.509v3 digital certificates and
Public Key Infrastructures(PKI) because of the greater scalability,
administrative options, and more robust security over traditional
password based encryption.

Most standard encryption algorithms are supported with this
specification. Reference implementations for many of these
algorithms are available from either commercial or open source
distributors.  Readily available cryptographic toolkits make
implementation of the encryption features straight-forward. 
This document is not intended to provide a treatise on data
encryption principles or theory.  Its purpose is to document the
data structures required for implementing interoperable data
encryption within the .ZIP format.  It is strongly recommended that
you have a good understanding of data encryption before reading
further.

The algorithms introduced in Version 5.0 of this specification
include:

    RC2 40 bit, 64 bit, and 128 bit
    RC4 40 bit, 64 bit, and 128 bit
    DES
    3DES 112 bit and 168 bit
 
Version 5.1 adds support for the following:

    AES 128 bit, 192 bit, and 256 bit


Version 6.1 introduces encryption data changes to support
interoperability with Smartcard and USB Token certificate storage
methods which do not support the OAEP strengthening standard.

Version 6.2 introduces support for encrypting metadata by compressing
and encrypting the central directory data structure to reduce information
leakage.   Information leakage can occur in legacy ZIP applications
through exposure of information about a file even though that file is
stored encrypted.  The information exposed consists of file
characteristics stored within the records and fields defined by this
specification.  This includes data such as a files name, its original
size, timestamp and CRC32 value.

Version 6.3 introduces support for encrypting data using the Blowfish
and Twofish algorithms.  These are symmetric block ciphers developed
by Bruce Schneier.  Blowfish supports using a variable length key from
32 to 448 bits.  Block size is 64 bits.  Implementations should use 16
rounds and the only mode supported within ZIP files is CBC. Twofish
supports key sizes 128, 192 and 256 bits.  Block size is 128 bits. 
Implementations should use 16 rounds and the only mode supported within
ZIP files is CBC.  Information and source code for both Blowfish and
Twofish algorithms can be found on the internet.  Consult with the author
of these algorithms for information on terms or restrictions on use.

Central Directory Encryption provides greater protection against
information leakage by encrypting the Central Directory structure and
by masking key values that are replicated in the unencrypted Local
Header.   ZIP compatible programs that cannot interpret an encrypted
Central Directory structure cannot rely on the data in the corresponding
Local Header for decompression information. 

Extra Field records that may contain information about a file that should
not be exposed should not be stored in the Local Header and should only
be written to the Central Directory where they can be encrypted.  This
design currently does not support streaming.  Information in the End of
Central Directory record, the Zip64 End of Central Directory Locator,
and the Zip64 End of Central Directory records are not encrypted.  Access
to view data on files within a ZIP file with an encrypted Central Directory
requires the appropriate password or private key for decryption prior to
viewing any files, or any information about the files, in the archive. 

Older ZIP compatible programs not familiar with the Central Directory
Encryption feature will no longer be able to recognize the Central
Directory and may assume the ZIP file is corrupt.  Programs that
attempt streaming access using Local Headers will see invalid
information for each file.  Central Directory Encryption need not be
used for every ZIP file.  Its use is recommended for greater security. 
ZIP files not using Central Directory Encryption should operate as
in the past.

This strong encryption feature specification is intended to provide for
scalable, cross-platform encryption needs ranging from simple password
encryption to authenticated public/private key encryption. 

Encryption provides data confidentiality and privacy.  It is
recommended that you combine X.509 digital signing with encryption
to add authentication and non-repudiation.


Single Password Symmetric Encryption Method:
-------------------------------------------

The Single Password Symmetric Encryption Method using strong
encryption algorithms operates similarly to the traditional
PKWARE encryption defined in this format.  Additional data
structures are added to support the processing needs of the
strong algorithms.

The Strong Encryption data structures are:

1. General Purpose Bits - Bits 0 and 6 of the General Purpose bit
flag in both local and central header records.  Both bits set
indicates strong encryption.  Bit 13, when set indicates the Central
Directory is encrypted and that selected fields in the Local Header
are masked to hide their actual value.


2. Extra Field 0x0017 in central header only.

     Fields to consider in this record are:

     Format - the data format identifier for this record.  The only
     value allowed at this time is the integer value 2.

     AlgId - integer identifier of the encryption algorithm from the
     following range

         0x6601 - DES
         0x6602 - RC2 (version needed to extract < 5.2)
         0x6603 - 3DES 168
         0x6609 - 3DES 112
         0x660E - AES 128
         0x660F - AES 192
         0x6610 - AES 256
         0x6702 - RC2 (version needed to extract >= 5.2)
         0x6720 - Blowfish
         0x6721 - Twofish
         0x6801 - RC4
         0xFFFF - Unknown algorithm

     Bitlen - Explicit bit length of key

         32 - 448 bits
  
     Flags - Processing flags needed for decryption

         0x0001 - Password is required to decrypt
         0x0002 - Certificates only
         0x0003 - Password or certificate required to decrypt

         Values > 0x0003 reserved for certificate processing


3. Decryption header record preceding compressed file data.

         -Decryption Header:

          Value     Size     Description
          -----     ----     -----------
          IVSize    2 bytes  Size of initialization vector (IV)
          IVData    IVSize   Initialization vector for this file
          Size      4 bytes  Size of remaining decryption header data
          Format    2 bytes  Format definition for this record
          AlgID     2 bytes  Encryption algorithm identifier
          Bitlen    2 bytes  Bit length of encryption key
          Flags     2 bytes  Processing flags
          ErdSize   2 bytes  Size of Encrypted Random Data
          ErdData   ErdSize  Encrypted Random Data
          Reserved1 4 bytes  Reserved certificate processing data
          Reserved2 (var)    Reserved for certificate processing data
          VSize     2 bytes  Size of password validation data
          VData     VSize-4  Password validation data
          VCRC32    4 bytes  Standard ZIP CRC32 of password validation data

     IVData - The size of the IV should match the algorithm block size.
              The IVData can be completely random data.  If the size of
              the randomly generated data does not match the block size
              it should be complemented with zero's or truncated as
              necessary.  If IVSize is 0,then IV = CRC32 + Uncompressed
              File Size (as a 64 bit little-endian, unsigned integer value).

     Format - the data format identifier for this record.  The only
     value allowed at this time is the integer value 3.

     AlgId - integer identifier of the encryption algorithm from the
     following range

         0x6601 - DES
         0x6602 - RC2 (version needed to extract < 5.2)
         0x6603 - 3DES 168
         0x6609 - 3DES 112
         0x660E - AES 128
         0x660F - AES 192
         0x6610 - AES 256
         0x6702 - RC2 (version needed to extract >= 5.2)
         0x6720 - Blowfish
         0x6721 - Twofish
         0x6801 - RC4
         0xFFFF - Unknown algorithm

     Bitlen - Explicit bit length of key

         32 - 448 bits
  
     Flags - Processing flags needed for decryption

         0x0001 - Password is required to decrypt
         0x0002 - Certificates only
         0x0003 - Password or certificate required to decrypt

         Values > 0x0003 reserved for certificate processing

     ErdData - Encrypted random data is used to store random data that
               is used to generate a file session key for encrypting
               each file.  SHA1 is used to calculate hash data used to
               derive keys.  File session keys are derived from a master
               session key generated from the user-supplied password.
               If the Flags field in the decryption header contains
               the value 0x4000, then the ErdData field must be
               decrypted using 3DES. If the value 0x4000 is not set,
               then the ErdData field must be decrypted using AlgId.


     Reserved1 - Reserved for certificate processing, if value is
               zero, then Reserved2 data is absent.  See the explanation
               under the Certificate Processing Method for details on
               this data structure.

     Reserved2 - If present, the size of the Reserved2 data structure
               is located by skipping the first 4 bytes of this field
               and using the next 2 bytes as the remaining size.  See
               the explanation under the Certificate Processing Method
               for details on this data structure.

     VSize - This size value will always include the 4 bytes of the
             VCRC32 data and will be greater than 4 bytes.

     VData - Random data for password validation.  This data is VSize
             in length and VSize must be a multiple of the encryption
             block size.  VCRC32 is a checksum value of VData. 
             VData and VCRC32 are stored encrypted and start the
             stream of encrypted data for a file.


4. Useful Tips

Strong Encryption is always applied to a file after compression. The
block oriented algorithms all operate in Cypher Block Chaining (CBC)
mode.  The block size used for AES encryption is 16.  All other block
algorithms use a block size of 8.  Two ID's are defined for RC2 to
account for a discrepancy found in the implementation of the RC2
algorithm in the cryptographic library on Windows XP SP1 and all
earlier versions of Windows.  It is recommended that zero length files
not be encrypted, however programs should be prepared to extract them
if they are found within a ZIP file.

A pseudo-code representation of the encryption process is as follows:

Password = GetUserPassword()
MasterSessionKey = DeriveKey(SHA1(Password))
RD = CryptographicStrengthRandomData()
For Each File
   IV = CryptographicStrengthRandomData()
   VData = CryptographicStrengthRandomData()
   VCRC32 = CRC32(VData)
   FileSessionKey = DeriveKey(SHA1(IV + RD)
   ErdData = Encrypt(RD,MasterSessionKey,IV)
   Encrypt(VData + VCRC32 + FileData, FileSessionKey,IV)
Done

The function names and parameter requirements will depend on
the choice of the cryptographic toolkit selected.  Almost any
toolkit supporting the reference implementations for each
algorithm can be used.  The RSA BSAFE(r), OpenSSL, and Microsoft
CryptoAPI libraries are all known to work well. 


Single Password - Central Directory Encryption:
-----------------------------------------------

Central Directory Encryption is achieved within the .ZIP format by
encrypting the Central Directory structure.  This encapsulates the metadata
most often used for processing .ZIP files.  Additional metadata is stored for
redundancy in the Local Header for each file.  The process of concealing
metadata by encrypting the Central Directory does not protect the data within
the Local Header.  To avoid information leakage from the exposed metadata
in the Local Header, the fields containing information about a file are masked. 

Local Header:

Masking replaces the true content of the fields for a file in the Local
Header with false information.  When masked, the Local Header is not
suitable for streaming access and the options for data recovery of damaged
archives is reduced.  Extra Data fields that may contain confidential
data should not be stored within the Local Header.  The value set into
the Version needed to extract field should be the correct value needed to
extract the file without regard to Central Directory Encryption. The fields
within the Local Header targeted for masking when the Central Directory is
encrypted are:

        Field Name                     Mask Value
        ------------------             ---------------------------
        compression method              0
        last mod file time              0
        last mod file date              0
        crc-32                          0
        compressed size                 0
        uncompressed size               0
        file name (variable size)       Base 16 value from the
                                        range 1 - 0xFFFFFFFFFFFFFFFF
                                        represented as a string whose
                                        size will be set into the
                                        file name length field

The Base 16 value assigned as a masked file name is simply a sequentially
incremented value for each file starting with 1 for the first file. 
Modifications to a ZIP file may cause different values to be stored for
each file.  For compatibility, the file name field in the Local Header
should never be left blank.  As of Version 6.2 of this specification,
the Compression Method and Compressed Size fields are not yet masked.
Fields having a value of 0xFFFF or 0xFFFFFFFF for the ZIP64 format
should not be masked. 

Encrypting the Central Directory:

Encryption of the Central Directory does not include encryption of the
Central Directory Signature data, the Zip64 End of Central Directory
record, the Zip64 End of Central Directory Locator, or the End
of Central Directory record.  The ZIP file comment data is never
encrypted.

Before encrypting the Central Directory, it may optionally be compressed.
Compression is not required, but for storage efficiency it is assumed
this structure will be compressed before encrypting.  Similarly, this
specification supports compressing the Central Directory without
requiring that it also be encrypted.  Early implementations of this
feature will assume the encryption method applied to files matches the
encryption applied to the Central Directory.

Encryption of the Central Directory is done in a manner similar to
that of file encryption.  The encrypted data is preceded by a
decryption header.  The decryption header is known as the Archive
Decryption Header.  The fields of this record are identical to
the decryption header preceding each encrypted file.  The location
of the Archive Decryption Header is determined by the value in the
Start of the Central Directory field in the Zip64 End of Central
Directory record.  When the Central Directory is encrypted, the
Zip64 End of Central Directory record will always be present.

The layout of the Zip64 End of Central Directory record for all
versions starting with 6.2 of this specification will follow the
Version 2 format.  The Version 2 format is as follows:

The leading fixed size fields within the Version 1 format for this
record remain unchanged.  The record signature for both Version 1
and Version 2 will be 0x06064b50.  Immediately following the last
byte of the field known as the Offset of Start of Central
Directory With Respect to the Starting Disk Number will begin the
new fields defining Version 2 of this record. 

New fields for Version 2:

Note: all fields stored in Intel low-byte/high-byte order.

          Value                 Size       Description
          -----                 ----       -----------
          Compression Method    2 bytes    Method used to compress the
                                           Central Directory
          Compressed Size       8 bytes    Size of the compressed data
          Original   Size       8 bytes    Original uncompressed size
          AlgId                 2 bytes    Encryption algorithm ID
          BitLen                2 bytes    Encryption key length
          Flags                 2 bytes    Encryption flags
          HashID                2 bytes    Hash algorithm identifier
          Hash Length           2 bytes    Length of hash data
          Hash Data             (variable) Hash data

The Compression Method accepts the same range of values as the
corresponding field in the Central Header.

The Compressed Size and Original Size values will not include the
data of the Central Directory Signature which is compressed or
encrypted.

The AlgId, BitLen, and Flags fields accept the same range of values
the corresponding fields within the 0x0017 record.

Hash ID identifies the algorithm used to hash the Central Directory
data.  This data does not have to be hashed, in which case the
values for both the HashID and Hash Length will be 0.  Possible
values for HashID are:

      Value         Algorithm
     ------         ---------
     0x0000          none
     0x0001          CRC32
     0x8003          MD5
     0x8004          SHA1
     0x8007          RIPEMD160
     0x800C          SHA256
     0x800D          SHA384
     0x800E          SHA512

When the Central Directory data is signed, the same hash algorithm
used to hash the Central Directory for signing should be used.
This is recommended for processing efficiency, however, it is
permissible for any of the above algorithms to be used independent
of the signing process.

The Hash Data will contain the hash data for the Central Directory.
The length of this data will vary depending on the algorithm used.

The Version Needed to Extract should be set to 62.

The value for the Total Number of Entries on the Current Disk will
be 0.  These records will no longer support random access when
encrypting the Central Directory.

When the Central Directory is compressed and/or encrypted, the
End of Central Directory record will store the value 0xFFFFFFFF
as the value for the Total Number of Entries in the Central
Directory.  The value stored in the Total Number of Entries in
the Central Directory on this Disk field will be 0.  The actual
values will be stored in the equivalent fields of the Zip64
End of Central Directory record.

Decrypting and decompressing the Central Directory is accomplished
in the same manner as decrypting and decompressing a file.

Certificate Processing Method:
-----------------------------

The Certificate Processing Method of for ZIP file encryption
defines the following additional data fields:

1. Certificate Flag Values

Additional processing flags that can be present in the Flags field of both
the 0x0017 field of the central directory Extra Field and the Decryption
header record preceding compressed file data are:

         0x0007 - reserved for future use
         0x000F - reserved for future use
         0x0100 - Indicates non-OAEP key wrapping was used.  If this
                  this field is set, the version needed to extract must
                  be at least 61.  This means OAEP key wrapping is not
                  used when generating a Master Session Key using
                  ErdData.
         0x4000 - ErdData must be decrypted using 3DES-168, otherwise use the
                  same algorithm used for encrypting the file contents.
         0x8000 - reserved for future use


2. CertData - Extra Field 0x0017 record certificate data structure

The data structure used to store certificate data within the section
of the Extra Field defined by the CertData field of the 0x0017
record are as shown:

          Value     Size     Description
          -----     ----     -----------
          RCount    4 bytes  Number of recipients. 
          HashAlg   2 bytes  Hash algorithm identifier
          HSize     2 bytes  Hash size
          SRList    (var)    Simple list of recipients hashed public keys

         
     RCount    This defines the number intended recipients whose
               public keys were used for encryption.  This identifies
               the number of elements in the SRList.

     HashAlg   This defines the hash algorithm used to calculate
               the public key hash of each public key used
               for encryption. This field currently supports
               only the following value for SHA-1

               0x8004 - SHA1

     HSize     This defines the size of a hashed public key.

     SRList    This is a variable length list of the hashed
               public keys for each intended recipient.  Each
               element in this list is HSize.  The total size of
               SRList is determined using RCount * HSize.


3. Reserved1 - Certificate Decryption Header Reserved1 Data:

          Value     Size     Description
          -----     ----     -----------
          RCount    4 bytes  Number of recipients. 
         
     RCount    This defines the number intended recipients whose
               public keys were used for encryption.  This defines
               the number of elements in the REList field defined below.


4. Reserved2 - Certificate Decryption Header Reserved2 Data Structures:


          Value     Size     Description
          -----     ----     -----------
          HashAlg   2 bytes  Hash algorithm identifier
          HSize     2 bytes  Hash size
          REList    (var)    List of recipient data elements


     HashAlg   This defines the hash algorithm used to calculate
               the public key hash of each public key used
               for encryption. This field currently supports
               only the following value for SHA-1

               0x8004 - SHA1

     HSize     This defines the size of a hashed public key
               defined in REHData.

     REList    This is a variable length of list of recipient data. 
               Each element in this list consists of a Recipient
               Element data structure as follows:


    Recipient Element (REList) Data Structure:

          Value     Size     Description
          -----     ----     -----------
          RESize    2 bytes  Size of REHData + REKData
          REHData   HSize    Hash of recipients public key
          REKData   (var)    Simple key blob


     RESize    This defines the size of an individual REList
               element.  This value is the combined size of the
               REHData field + REKData field.  REHData is defined by
               HSize.  REKData is variable and can be calculated
               for each REList element using RESize and HSize.

     REHData   Hashed public key for this recipient.

     REKData   Simple Key Blob.  The format of this data structure
               is identical to that defined in the Microsoft
               CryptoAPI and generated using the CryptExportKey()
               function.  The version of the Simple Key Blob
               supported at this time is 0x02 as defined by
               Microsoft.

Certificate Processing - Central Directory Encryption:
------------------------------------------------------

Central Directory Encryption using Digital Certificates will
operate in a manner similar to that of Single Password Central
Directory Encryption.  This record will only be present when there
is data to place into it.  Currently, data is placed into this
record when digital certificates are used for either encrypting
or signing the files within a ZIP file.  When only password
encryption is used with no certificate encryption or digital
signing, this record is not currently needed. When present, this
record will appear before the start of the actual Central Directory
data structure and will be located immediately after the Archive
Decryption Header if the Central Directory is encrypted.

The Archive Extra Data record will be used to store the following
information.  Additional data may be added in future versions.

Extra Data Fields:

0x0014 - PKCS#7 Store for X.509 Certificates
0x0016 - X.509 Certificate ID and Signature for central directory
0x0019 - PKCS#7 Encryption Recipient Certificate List

The 0x0014 and 0x0016 Extra Data records that otherwise would be
located in the first record of the Central Directory for digital
certificate processing. When encrypting or compressing the Central
Directory, the 0x0014 and 0x0016 records must be located in the
Archive Extra Data record and they should not remain in the first
Central Directory record.  The Archive Extra Data record will also
be used to store the 0x0019 data.

When present, the size of the Archive Extra Data record will be
included in the size of the Central Directory.  The data of the
Archive Extra Data record will also be compressed and encrypted
along with the Central Directory data structure.

Certificate Processing Differences:

The Certificate Processing Method of encryption differs from the
Single Password Symmetric Encryption Method as follows.  Instead
of using a user-defined password to generate a master session key,
cryptographically random data is used.  The key material is then
wrapped using standard key-wrapping techniques.  This key material
is wrapped using the public key of each recipient that will need
to decrypt the file using their corresponding private key.

This specification currently assumes digital certificates will follow
the X.509 V3 format for 1024 bit and higher RSA format digital
certificates.  Implementation of this Certificate Processing Method
requires supporting logic for key access and management.  This logic
is outside the scope of this specification.

OAEP Processing with Certificate-based Encryption:

OAEP stands for Optimal Asymmetric Encryption Padding.  It is a
strengthening technique used for small encoded items such as decryption
keys.  This is commonly applied in cryptographic key-wrapping techniques
and is supported by PKCS #1.  Versions 5.0 and 6.0 of this specification
were designed to support OAEP key-wrapping for certificate-based
decryption keys for additional security. 

Support for private keys stored on Smartcards or Tokens introduced
a conflict with this OAEP logic.  Most card and token products do
not support the additional strengthening applied to OAEP key-wrapped
data.  In order to resolve this conflict, versions 6.1 and above of this
specification will no longer support OAEP when encrypting using
digital certificates.

Versions of PKZIP available during initial development of the
certificate processing method set a value of 61 into the
version needed to extract field for a file.  This indicates that
non-OAEP key wrapping is used.  This affects certificate encryption
only, and password encryption functions should not be affected by
this value.  This means values of 61 may be found on files encrypted
with certificates only, or on files encrypted with both password
encryption and certificate encryption.  Files encrypted with both
methods can safely be decrypted using the password methods documented.

IX. Change Process
------------------

In order for the .ZIP file format to remain a viable definition, this
specification should be considered as open for periodic review and
revision.  Although this format was originally designed with a
certain level of extensibility, not all changes in technology
(present or future) were or will be necessarily considered in its
design.  If your application requires new definitions to the
extensible sections in this format, or if you would like to
submit new data structures, please forward your request to
zipformat@pkware.com.  All submissions will be reviewed by the
ZIP File Specification Committee for possible inclusion into
future versions of this specification.  Periodic revisions
to this specification will be published to ensure interoperability.
We encourage comments and feedback that may help improve clarity
or content.

X. Incorporating PKWARE Proprietary Technology into Your Product
----------------------------------------------------------------

PKWARE is committed to the interoperability and advancement of the
.ZIP format.  PKWARE offers a free license for certain technological
aspects described above under certain restrictions and conditions.
However, the use or implementation in a product of certain technological
aspects set forth in the current APPNOTE, including those with regard to
strong encryption, patching, or extended tape operations requires a
license from PKWARE.  Please contact PKWARE with regard to acquiring
a license.

XI. Acknowledgements
---------------------

In addition to the above mentioned contributors to PKZIP and PKUNZIP,
I would like to extend special thanks to Robert Mahoney for suggesting
the extension .ZIP for this software.

XII. References
---------------

    Fiala, Edward R., and Greene, Daniel H., "Data compression with
       finite windows",  Communications of the ACM, Volume 32, Number 4,
       April 1989, pages 490-505.

    Held, Gilbert, "Data Compression, Techniques and Applications,
       Hardware and Software Considerations", John Wiley & Sons, 1987.

    Huffman, D.A., "A method for the construction of minimum-redundancy
       codes", Proceedings of the IRE, Volume 40, Number 9, September 1952,
       pages 1098-1101.

    Nelson, Mark, "LZW Data Compression", Dr. Dobbs Journal, Volume 14,
       Number 10, October 1989, pages 29-37.

    Nelson, Mark, "The Data Compression Book",  M&T Books, 1991.

    Storer, James A., "Data Compression, Methods and Theory",
       Computer Science Press, 1988

    Welch, Terry, "A Technique for High-Performance Data Compression",
       IEEE Computer, Volume 17, Number 6, June 1984, pages 8-19.

    Ziv, J. and Lempel, A., "A universal algorithm for sequential data
       compression", Communications of the ACM, Volume 30, Number 6,
       June 1987, pages 520-540.

    Ziv, J. and Lempel, A., "Compression of individual sequences via
       variable-rate coding", IEEE Transactions on Information Theory,
       Volume 24, Number 5, September 1978, pages 530-536.


APPENDIX A - AS/400 Extra Field (0x0065) Attribute Definitions
--------------------------------------------------------------

Field Definition Structure:

   a. field length including length             2 bytes
   b. field code                                2 bytes
   c. data                                      x bytes

Field Code  Description
   4001     Source type i.e. CLP etc
   4002     The text description of the library
   4003     The text description of the file
   4004     The text description of the member
   4005     x'F0' or 0 is PF-DTA,  x'F1' or 1 is PF_SRC
   4007     Database Type Code                  1 byte
   4008     Database file and fields definition
   4009     GZIP file type                      2 bytes
   400B     IFS code page                       2 bytes
   400C     IFS Creation Time                   4 bytes
   400D     IFS Access Time                     4 bytes
   400E     IFS Modification time               4 bytes
   005C     Length of the records in the file   2 bytes
   0068     GZIP two words                      8 bytes

APPENDIX B - z/OS Extra Field (0x0065) Attribute Definitions
------------------------------------------------------------

Field Definition Structure:

   a. field length including length             2 bytes
   b. field code                                2 bytes
   c. data                                      x bytes

Field Code  Description
   0001     File Type                           2 bytes
   0002     NonVSAM Record Format               1 byte
   0003     Reserved       
   0004     NonVSAM Block Size                  2 bytes Big Endian
   0005     Primary Space Allocation            3 bytes Big Endian
   0006     Secondary Space Allocation          3 bytes Big Endian
   0007     Space Allocation Type1 byte flag       
   0008     Modification Date                   Retired with PKZIP 5.0 +
   0009     Expiration Date                     Retired with PKZIP 5.0 +
   000A     PDS Directory Block Allocation      3 bytes Big Endian binary value
   000B     NonVSAM Volume List                 variable       
   000C     UNIT Reference                      Retired with PKZIP 5.0 +
   000D     DF/SMS Management Class             8 bytes EBCDIC Text Value
   000E     DF/SMS Storage Class                8 bytes EBCDIC Text Value
   000F     DF/SMS Data Class                   8 bytes EBCDIC Text Value
   0010     PDS/PDSE Member Info.               30 bytes   
   0011     VSAM sub-filetype                   2 bytes       
   0012     VSAM LRECL                          13 bytes EBCDIC "(num_avg num_max)"
   0013     VSAM Cluster Name                   Retired with PKZIP 5.0 +
   0014     VSAM KSDS Key Information           13 bytes EBCDIC "(num_length num_position)"
   0015     VSAM Average LRECL                  5 bytes EBCDIC num_value padded with blanks
   0016     VSAM Maximum LRECL                  5 bytes EBCDIC num_value padded with blanks
   0017     VSAM KSDS Key Length                5 bytes EBCDIC num_value padded with blanks
   0018     VSAM KSDS Key Position              5 bytes EBCDIC num_value padded with blanks
   0019     VSAM Data Name                      1-44 bytes EBCDIC text string
   001A     VSAM KSDS Index Name                1-44 bytes EBCDIC text string
   001B     VSAM Catalog Name                   1-44 bytes EBCDIC text string
   001C     VSAM Data Space Type                9 bytes EBCDIC text string
   001D     VSAM Data Space Primary             9 bytes EBCDIC num_value left-justified
   001E     VSAM Data Space Secondary           9 bytes EBCDIC num_value left-justified
   001F     VSAM Data Volume List               variable EBCDIC text list of 6-character Volume IDs
   0020     VSAM Data Buffer Space              8 bytes EBCDIC num_value left-justified
   0021     VSAM Data CISIZE                    5 bytes EBCDIC num_value left-justified
   0022     VSAM Erase Flag                     1 byte flag       
   0023     VSAM Free CI %                      3 bytes EBCDIC num_value left-justified
   0024     VSAM Free CA %                      3 bytes EBCDIC num_value left-justified
   0025     VSAM Index Volume List              variable EBCDIC text list of 6-character Volume IDs
   0026     VSAM Ordered Flag                   1 byte flag       
   0027     VSAM REUSE Flag                     1 byte flag       
   0028     VSAM SPANNED Flag                   1 byte flag       
   0029     VSAM Recovery Flag                  1 byte flag       
   002A     VSAM  WRITECHK  Flag                1 byte flag       
   002B     VSAM Cluster/Data SHROPTS           3 bytes EBCDIC "n,y"   
   002C     VSAM Index SHROPTS                  3 bytes EBCDIC "n,y"   
   002D     VSAM Index Space Type               9 bytes EBCDIC text string
   002E     VSAM Index Space Primary            9 bytes EBCDIC num_value left-justified
   002F     VSAM Index Space Secondary          9 bytes EBCDIC num_value left-justified
   0030     VSAM Index CISIZE                   5 bytes EBCDIC num_value left-justified
   0031     VSAM Index IMBED                    1 byte flag       
   0032     VSAM Index Ordered Flag             1 byte flag       
   0033     VSAM REPLICATE Flag                 1 byte flag       
   0034     VSAM Index REUSE Flag               1 byte flag       
   0035     VSAM Index WRITECHK Flag            1 byte flag Retired with PKZIP 5.0 +
   0036     VSAM Owner                          8 bytes EBCDIC text string
   0037     VSAM Index Owner                    8 bytes EBCDIC text string
   0038     Reserved
   0039     Reserved
   003A     Reserved
   003B     Reserved
   003C     Reserved
   003D     Reserved
   003E     Reserved
   003F     Reserved
   0040     Reserved
   0041     Reserved
   0042     Reserved
   0043     Reserved
   0044     Reserved
   0045     Reserved
   0046     Reserved
   0047     Reserved
   0048     Reserved
   0049     Reserved
   004A     Reserved
   004B     Reserved
   004C     Reserved
   004D     Reserved
   004E     Reserved
   004F     Reserved
   0050     Reserved
   0051     Reserved
   0052     Reserved
   0053     Reserved
   0054     Reserved
   0055     Reserved
   0056     Reserved
   0057     Reserved
   0058     PDS/PDSE Member TTR Info.           6 bytes  Big Endian
   0059     PDS 1st LMOD Text TTR               3 bytes  Big Endian
   005A     PDS LMOD EP Rec #                   4 bytes  Big Endian
   005B     Reserved
   005C     Max Length of records               2 bytes  Big Endian
   005D     PDSE Flag                           1 byte flag
   005E     Reserved
   005F     Reserved
   0060     Reserved
   0061     Reserved
   0062     Reserved
   0063     Reserved
   0064     Reserved
   0065     Last Date Referenced                4 bytes  Packed Hex "yyyymmdd"
   0066     Date Created                        4 bytes  Packed Hex "yyyymmdd"
   0068     GZIP two words                      8 bytes
   0071     Extended NOTE Location              12 bytes Big Endian
   0072     Archive device UNIT                 6 bytes  EBCDIC
   0073     Archive 1st Volume                  6 bytes  EBCDIC
   0074     Archive 1st VOL File Seq#           2 bytes  Binary

APPENDIX C - Zip64 Extensible Data Sector Mappings (EFS)
--------------------------------------------------------

          -Z390   Extra Field:

          The following is the general layout of the attributes for the
          ZIP 64 "extra" block for extended tape operations. Portions of
          this extended tape processing technology is covered under a
          pending patent application. The use or implementation in a
          product of certain technological aspects set forth in the
          current APPNOTE, including those with regard to strong encryption,
          patching or extended tape operations, requires a license from
          PKWARE.  Please contact PKWARE with regard to acquiring a license.
 

          Note: some fields stored in Big Endian format.  All text is
      in EBCDIC format unless otherwise specified.

          Value       Size          Description
          -----       ----          -----------
  (Z390)  0x0065      2 bytes       Tag for this "extra" block type
          Size        4 bytes       Size for the following data block
          Tag         4 bytes       EBCDIC "Z390"
          Length71    2 bytes       Big Endian
          Subcode71   2 bytes       Enote type code
          FMEPos      1 byte
          Length72    2 bytes       Big Endian
          Subcode72   2 bytes       Unit type code
          Unit        1 byte        Unit
          Length73    2 bytes       Big Endian
          Subcode73   2 bytes       Volume1 type code
          FirstVol    1 byte        Volume
          Length74    2 bytes       Big Endian
          Subcode74   2 bytes       FirstVol file sequence
          FileSeq     2 bytes       Sequence

APPENDIX D - Language Encoding (EFS)
------------------------------------

The ZIP format has historically supported only the original IBM PC character
encoding set, commonly referred to as IBM Code Page 437.  This limits storing
file name characters to only those within the original MS-DOS range of values
and does not properly support file names in other character encodings, or
languages. To address this limitation, this specification will support the
following change.

If general purpose bit 11 is unset, the file name and comment should conform
to the original ZIP character encoding.  If general purpose bit 11 is set, the
filename and comment must support The Unicode Standard, Version 4.1.0 or
greater using the character encoding form defined by the UTF-8 storage
specification.  The Unicode Standard is published by the The Unicode
Consortium (www.unicode.org).  UTF-8 encoded data stored within ZIP files
is expected to not include a byte order mark (BOM).

Applications may choose to supplement this file name storage through the use
of the 0x0008 Extra Field.  Storage for this optional field is currently
undefined, however it will be used to allow storing extended information
on source or target encoding that may further assist applications with file
name, or file content encoding tasks.  Please contact PKWARE with any
requirements on how this field should be used.

The 0x0008 Extra Field storage may be used with either setting for general
purpose bit 11.  Examples of the intended usage for this field is to store
whether "modified-UTF-8" (JAVA) is used, or UTF-8-MAC.  Similarly, other
commonly used character encoding (code page) designations can be indicated
through this field.  Formalized values for use of the 0x0008 record remain
undefined at this time.  The definition for the layout of the 0x0008 field
will be published when available.  Use of the 0x0008 Extra Field provides
for storing data within a ZIP file in an encoding other than IBM Code
Page 437 or UTF-8.

General purpose bit 11 will not imply any encoding of file content or
password.  Values defining character encoding for file content or
password must be stored within the 0x0008 Extended Language Encoding
Extra Field.

Ed Gordon of the Info-ZIP group has defined a pair of "extra field" records
that can be used to store UTF-8 file name and file comment fields.  These
records can be used for cases when the general purpose bit 11 method
for storing UTF-8 data in the standard file name and comment fields is
not desirable.  A common case for this alternate method is if backward
compatibility with older programs is required.

Definitions for the record structure of these fields are included above
in the section on 3rd party mappings for "extra field" records.  These
records are identified by Header ID's 0x6375 (Info-ZIP Unicode Comment
Extra Field) and 0x7075 (Info-ZIP Unicode Path Extra Field).

The choice of which storage method to use when writing a ZIP file is left
to the implementation.  Developers should expect that a ZIP file may
contain either method and should provide support for reading data in
either format. Use of general purpose bit 11 reduces storage requirements
for file name data by not requiring additional "extra field" data for
each file, but can result in older ZIP programs not being able to extract
files.  Use of the 0x6375 and 0x7075 records will result in a ZIP file
that should always be readable by older ZIP programs, but requires more
storage per file to write file name and/or file comment fields.

 



ZIP文件格式分为三部分:压缩源文件数据区+压缩源文件目录区+压缩源文件目录结束标志
     A、压缩源文件数据区
    在这个数据区中每一个压缩的源文件/目录都是一条记录,记录的格式如下:

     [文件头+ 文件数据 + 数据描述符]
     1、文件头结构

          组成                       长度
       文件头标记                   4 bytes   (0x04034b50)
       解压文件所需 pkware 版本     2 bytes
       全局方式位标记               2 bytes
     压缩方式                     2 bytes
     最后修改文件时间              2 bytes
     最后修改文件日期              2 bytes
     CRC-32校验                   4 bytes
     压缩后尺寸                   4 bytes
     未压缩尺寸                   4 bytes
     文件名长度                   2 bytes

       扩展记录长度                 2 bytes
     文件名                      (不定长度)
     扩展字段                    (不定长度)

     2、文件数据

    

     3、数据描述符

       组成                        长度
     CRC-32校验                     4 bytes
     压缩后尺寸                     4 bytes
     未压缩尺寸                     4 bytes

      这 个数据描述符只在全局方式位标记的第3位设为1时才存在(见后详解),紧接在压缩数据的最后一个字节后。这个数据描述符只用在不能对输出的 ZIP 文件进行检索时使用。例如:在一个不能检索的驱动器(如:磁带机上)上的 ZIP 文件中。如果是磁盘上的ZIP文件一般没有这个数据描述符。

       B、压缩源文件目录区

      在这个数据区中每一条纪录对应在压缩源文件数据区中的一条数据

      组成                             长度

      目录中文件文件头标记              4 bytes  (0x02014b50)

      压缩使用的 pkware 版本          2 bytes

      解压文件所需 pkware 版本         2 bytes

      全局方式位标记                   2 bytes

      压缩方式                        2 bytes

      最后修改文件时间                 2 bytes

      最后修改文件日期                 2 bytes

      CRC-32校验                 4 bytes

      压缩后尺寸                      4 bytes

      未压缩尺寸                      4 bytes

      文件名长度                      2 bytes

      扩展字段长度                    2 bytes

      文件注释长度                    2 bytes

      磁盘开始号                      2 bytes

      内部文件属性                    2 bytes

      外部文件属性                    4 bytes

        局部头部偏移量                  4 bytes

      文件名                       (不定长度)

      扩展字段                     (不定长度)

        文件注释                     (不定长度)

     C、压缩源文件目录结束标志

       组成                           长度

        目录结束标记                     4 bytes  (0x02014b50)

         当前磁盘编号                    2 bytes

        目录区开始磁盘编号               2 bytes

      本磁盘上纪录总数                 2 bytes

      目录区中纪录总数                 2 bytes

      目录区尺寸大小                   4 bytes

      目录区对第一张磁盘的偏移量        4 bytes

      ZIP 文件注释长度                 2 bytes

      ZIP 文件注释                   (不定长度)