@database "example.guide"
@master "Split!.doc"

@node Main "Split! 1.0 User Manual -- December 11, 1993"

                        Split! 1.0 -- by Dan Fraser

                                User Manual

                             December 11, 1993

Table of Contents:

  @{" Introduction          " link INTRODUCTION} 
  @{" Installation          " link INSTALLATION}
  @{" Usage                 " link USAGE}
  @{" Benchmarks            " link BENCHMARKS}
  @{" Limitations           " link LIMITATIONS}
  @{" Distribution          " link DISTRIBUTION}
  @{" Contacting the Author " link Author}

@endnode
@node INTRODUCTION

INTRODUCTION

	Frequently, I had to move large files between platforms.  When I
        had a 10 megabyte file on my Amiga hard disk that I wanted to take
        to school (an IBM network) for printing, I found that when I LhAed
        the file, it still wouldn't fit on a floppy disk.  After trying
        other file splitters, I found that they were just too slow for my
        needs.  I wrote my own.

@endnode
@node INSTALLATION

INSTALLATION

	Split! is a CLI-based command, and therefore should be placed
        somewhere in your command path.  I reccomend that you move the
        executable to your C: directory before use.

@endnode
@node USAGE

USAGE
	
	Split! must be run from the CLI, and takes only two parameters:  

	Split! <infile> <outfile>

	<infile> is the large file you want to split up into smaller files.

	<outfile> is the root of the destination filename.  For each file
		  created, a numerical extention will be added.

	Once run, Split! will ask you how big (in Kilobytes) to make each
        <outfile>.  To transport it on standard MS-DOS 720k disks, I would
        reccomend a split size of 700.  For 1.44Mb MS-DOS disks, enter 1400.

	Example:

	Split BigFile LittleFile
        SplitSize (Kb): 700

        The files would be called:  LittleFile.1, LittleFile.2, etc.

        
        On the MS-DOS system, you just copy all the files off the disks
        into a directory on the hard disk.  The MS-DOS 'COPY' command
        does a good job of putting them back to gether.

        COPY LittleFile.1/b+LittleFile.2/b+LittleFile.3/b BigFile

        This would join all the smaller files into a larger one called
        'BigFile.'  The '/b' switch on each file is important.  It tells
        COPY that it is joining binary files, and thus will not stop at
        the first CTRL-Z it sees.

@endnode
@node BENCHMARKS

BENCHMARKS

        These tests where done with the file NodeList.316, the FidoNet
        nodelist.  The source file was 2212118 bytes long.  The splitsize
        in both cases was 700k.

        Split! 1.0       Chopper 1.1
        ==========       ===========

        10.23 sec.       140.21 sec.

@endnode
@node LIMITATIONS

LIMITATIONS

        Alas, the programs speed is not a result of my programming 
        prowess.  It is simply because of my big-assed buffer!  By
        processing 32k at a time, your systems hard disk has to seek
        less.  Unfortuantely, this is demanding on _very_ low memory
        systems.  I would have included my PCQ Pascal source code, 
        but I'm too embarrased.

@endnode
@node DISTRIBUTION

DISTRIBUTION

        Although this program is freely distributable, it remains _my_
        program, Copyright (c) Dan Fraser, 1993.

        I shall impose a few limitations on distribution.  It must not be
        sold for profit.  It must not be placed in public domain
        collections for which more than $4 a disk is charged, and it may
        not be used in commercial product distributions without my written
        permission.

@endnode
@node Author

CONTACTING THE AUTHOR

        Dan Fraser
        FidoNet: 1:221/125.1
        
        Snail Mail:

        88 Windsor Avenue 
        London, Ontario
        CANADA
        N6C 1Z9

@endnode