python implementation of a new integer encoding algorithm.
Denis McMahon
denismfmcmahon at gmail.com
Thu Feb 19 21:46:25 EST 2015
On Tue, 17 Feb 2015 03:22:47 -0800, janhein.vanderburg wrote:
> In http://optarbvalintenc.blogspot.nl/ I propose a new way to encode
> arbitrarily valued integers
I'm not quite sure I understand the problem that you're trying to solve
with this.
If I want to transmit some arbitrarily huge integer value between two
systems there are several ways to do it:
Numeric text - roughly 24 bits are used to carry 10 bits of value
Alphanumeric text - 8 bits are used to carry every 4 bits of value
A bit sequence of length n bits where 2^n > the value I wish to convey,
providing that there is some mechanism by which the sender can tell the
receiver "the next n bits represent a binary integer." It might be
desirable to pad the bit stream to a byte boundary. Assuming an 8 bit
clean transmission path and a maximum padding of 7 bits, this is
increasingly efficient as the integer being transmitted gets larger.
The thing is, for all practical purposes, any integer that is capable of
being processed as an integer by a computer is probably already being
stored in a binary format in a storage space of n bytes, where n is a
power of 2. Very little additional processing should be needed to
packetize those n bytes and transmit them.
Additionally, you're talking about efficiency and the need to optimise,
but you're using a scripted language. If you need a highly efficient
protocol to transfer binary numbers between two systems with minimum
wasted bits and maximum cpu and memory efficiency, python really isn't
the language in which to solve your problem.
Perhaps it's time to take a step back and redefine the problem a bit more
clearly, because at the moment I'm not sure you're solution will ever
solve anything that needs solving.
--
Denis McMahon, denismfmcmahon at gmail.com
More information about the Python-list
mailing list