[BioRuby] Ruby & D shared libraries

[Pjotr Prins]

Hi Artem,

On Tue, Apr 10, 2012 at 10:54:54PM +0200, Pjotr Prins wrote:
> But really, let's not optimize prematurely. A smart API does not need
> to pump much data...

Don't take this as criticism. Your work is very valuable - in fact,
I'll use it in the next weeks. I just meant to say that during
implementation we'll probably find the transfer of data between the
languages is probably not the true bottle neck (based on my experience
in BioLib). For example with a GFF3 parser, the file would be read and
parsed on the D site.  An iterator would pick up results. Copying the
buffered results in RAM will not be that expensive, relatively
speaking.

There is only one way to find out :)

Pj.