NAME

       EBook::Tools::LZSS - Lempel-Ziv-Storer-Szymanski compression and decompression

SYNOPSIS

        use EBook::Tools::LZSS;
        my $lzss = EBook::Tools::LZSS->new(lengthbits => 3,
                                           offsetbits => 14,
                                           windowinit => 'the man');
        my $textref = $lzss->uncompress(\$data);

CONSTRUCTOR AND INITIALIZATION

   "new(%args)"
       Instantiates a new EBook::Tools::LZSS object.

       Arguments

       All arguments are optional, but must be identical between compression and decompression for the result to
       be valid.

       •   "lengthbits"

           The  number  of  bits  used to encode the length of a LZSS reference.  If not specified defaults to 4
           bits.

           The eBookwise .IMP format typically compresses with 3 length bits.

           Note that the actual length of the LZSS reference in bytes is greater than the value  stored  in  the
           length  bits.  The actual number of bytes returned is the decoded length bits value plus "maxuncoded"
           plus 1,

       •   "offsetbits"

           The number of bits used to encode the offset to a LZSS reference.  This also determines the  size  of
           the sliding window of reference data.  If not specified, it defaults to 12 bits (4096-byte window).

           The eBookwise .IMP format typically compresses with 14 offset bits (16384-byte window).

       •   "windowinit"

           A  string used to initalize the sliding window.  If specified, this string MUST be the same length as
           the window size, or the subroutine will croak.  If not specified, the window will be initialized with
           spaces.

       •   "windowstart"

           The first byte position in the window that will be overwritten by decoded text.   If  not  specified,
           defaults to 0.

       •   "maxuncoded"

           The  maximum number of uncoded bytes (?).  This currently isn't used for that purpose, but determines
           the actual length of a LZSS reference.

       •   "screwybits"

           If set to true and the number of offset bits is greater than 8, then the offset  bits  will  be  read
           first  in  a  chunk of 8 for the least significant bits, and then the remaining bits will be read and
           use as the most significant bits.   This  seems  to  be  necessary  for  compatibility  with  Michael
           Dipperstein's LZSS C compression library but does not hold true for IMP e-book LZSS decompression.

       •   "verbose"

           If set to true, compression and uncompression will provide additional status feedback on STDOUT.

METHODS

   "uncompress(\$dataref)"
       Takes  a  reference  to a compressed data string, uncompresses the data string and returns a reference to
       the uncompressed string.

BUGS AND LIMITATIONS

       •   Compression not yet implemented.

       •   The LZSS algorithm isn't documented in the POD.

       •   Once compression is implemented and the algorithm is documented, this module is a good candidate  for
           being forked off into its own project.  Relying on it to keep its current name may be unwise.

AUTHOR

       Zed Pobre <zed@debian.org>

       The  design  of  this  module  was  based on the C LZSS library by Michael Dipperstein, version 0.5.2, at
       http://michael.dipperstein.com/lzss/

LICENSE AND COPYRIGHT

       Copyright 2008 Zed Pobre

       Licensed to the public under the terms of the GNU GPL, version 2.

perl v5.24.1                                       2017-05-22                            EBook::Tools::LZSS(3pm)