Adding modifiers to NCI

[chromatic]

On Thursday 25 December 2008 23:11:40 Allison Randal wrote:

> (First design principles: similar things should be similar, and different
> things should be different.)

Having written or maintained more than one thunking layer between different 
calling conventions (Perl 5 to C, Parrot to C, Perl 5 to Parrot, Parrot to 
Perl 5), let me assure you that calling conventions between languages differ 
more than they are similar.

Forcing all creators of NCI bindings to map well-understood C concepts to 
completely different semantics in an attempt to unify parsing -- especially 
when there already exists a usable, forty-year old mechanism for specifying 
the signatures of C functions -- is a foolish consistency.

> Shifting this syntax over to PCC-style is a trivial conversion.

If that is the case, you may convince me.

-- c