Floats in the protocol

[Graeme Gill]

Simon Thum wrote:
> I'm trying to get some consensus about floats on the wire. I cc'ed some
> who spoke about the issue on the list in a non-exhaustive search.

> Comments/Suggestions please?

An approach I've used when dealing with IEEE floats in binary
protocols is to code an IEEE <-> machine native routine.
Throw some platform sensitive #ifdef's in there and
and the code can be skipped where IEEE is native.

Graeme Gill.