Re: Prothon and a CPAN-like archive

gyrof wrote:
> Hi All,
> This may have already been discussed, but has consideration been
> given to developing the infrastructure and rigor necessary to
> create a Perl-like CPAN archive for Prothon modules (CPrAN)? I
> realize that this is probably premature, as the language itself is
> undergoing development, but if thought can be given upfront, it
> may ease things down the road.
> I am sorry that Python does not match Perl with regard to the ease
> of finding/archiving/installing modules, and I believe that the
> availability of CPAN for Perl users keeps many of them firmly
> 'addicted' to Perl. Prothon should not miss the boat on this issue.

I would like to make pitch for tying this in with Prosist.  Prosist has the
ability to load, update, and unload object trees in a clean way.  It also
has an identity concept that goes across interpreters.  This identity would
need to be made more global to match URL's etc.

Prosist also offers the marshaling of the objects for transporting across
the network.  It needs security improvements but I'd rather add them to
Prosist than to start over again on a new facility.  Marshalling of code
isn't finished in Prosist but I know what needs to be done.

Prosist could also be improved to add arbitrary versioning.  This would be
useful in any persistence application.

I guess my overall argument is that storing and distributing modules can be
viewed as just an application of the general problem that can be solved with
a good persistence engine.  Let's kill a bunch of birds with one stone.