This is a text file to test the Huffman Encoder and Decoder.

It was written in a text editor using ascii text and then converted
to binary and hexadecimal text formats (with leading 0x) for each of
8-, 16- and 32-bit data so that it can be used as test data for the
Huffman routines.

Each sentence, in each paragraph, is hard returned and punctuation is
used to add variety to the symbols and the probability of occurrence
for the symbols and hence the codewords that are produced.

Random occurrences of words such as "queen" and "zebra" are included
so that those letters of the alphabet that don't occur very often are
incorporated into the test routine.

White space and capital letters are also a key factor with the most
frequently occurring symbol expected [but not proven as yet!] being
a 'space' character or 0x20 as it is in hexadecimal format.

Other characters such as *, $, %, , #, # and the "curly braces" {}
could have been included have not been??!!!!!

All that remains now is to use the file to test the Huffman functions...

Contact Information
_______ ___________

For further technical support please contact the Software Support 
group at ARM:

  Software Support
  ARM Limited
  Fulbourn Road
  Cherry Hinton
  Cambridge
  CB1 4JN
  UK

  sw-support@arm.com
