A short introduction to V1.4.

V1.4 means significant speed improvements, as can be seen in this table:
(V1.4 was run with multibuffer support enabled)

Test file:	Size

Binary		1722/1840
Encoded		2461/2461


Timing	Y1.1	Y1.2	Big buf	Multi	V1.4
Encode	4.20	2.24	1.50	1.41	1.05
Decode	3.53	2.28	1.45	1.34	1.34

(Times in minutes)


The following switches are recognized:
/ID, /HE, /NM, /WI, /DE, /EN
Check /HE for a short help on this.

The default behaviour of COD is like this:
If /EN or /DE is specified, this is the work done.
If neither is specified, the input file specification is parsed.
If it ends in .ENC, an attempt to decode the file is done, otherwise
an encode is done. If no extension is given, a decode with the
file type .ENC is first tried, and then an encode with .TSK is tried.
On an encode, the output file is defaulted to the same as the
input file, but with the extension .ENC.
On a decode, the output file is defaulted to the same as the
file name specified _in_ the input file.

The format is bascially like this:
A coded file can hold any stuff. On decode, COD looks for a line
with the text (START) first on it. After this, it expects control
lines. All control lines begin with a left parenthesis, and end
with a right parenthesis. Anything after the right paren, up to
the end of the line is ignored.
The control line saying (DATA) means that data lines will follow.
The follow the same pattern, except that they begin with a
less-than character, and ends with a greater-than character.
If a control line appears in the data part, it is checked
against (END xxx), marking the end of data, including
a checksum.

Data itself is encoded in a variant of radix50, using the characters
A-Z 0-9 +-?.:
The character = is used to mark a repeat block. The next two characters
are the hex encoding of the repeat count, and after this comes the
word that shall be repeated (n + 1 times).