On Apr 6, 2006, at 5:08 PM, Patrick R. Michaud wrote:
On Thu, Apr 06, 2006 at 02:04:06PM -0700, Bernhard Schmalhofer via
RT wrote:
Hi,
as far as I see, the Perl* PMCs are no longer used in the Parrot
core.
Thanks, Bernhard.
There is still some usage in unmaintained language implementations:
BASIC/compiler unmaintained ?
BASIC/interpreter unmaintained ?
forth unmaintained ?
miniperl unmaintained ?
parakeet unmaintained ?
regex unmaintained ?
'punie' seems to be the only maintained language implementation using
Perl* PMCs. Also some tests and examples are using the Perl* PMCs.
There has been some discussion of this. First, Chip has declared
that Parrot won't change significantly until 1.0, and in some
respects I think this may include the Perl* PMCs.
For the Perl 6 compiler, we will likely create a separate set
of "Perl6*" PMCs. I'm hoping that we can build a common set of
Perl6* PMCs that would meet the needs of the p6 compiler, Pugs,
and whoever else may need PMC objects with Perl 6 semantics.
Pm
The existing Perl PMCs are not part of parrot's design at the moment,
SFAIK. Changing them to dynops is, in my opinion just an
implementation detail. The only reason they are used at all in the
languages listed above (aside, perhaps, from punie) is that parrot's
core PMCs came later - they used the most functional PMCs that
existed at the time. If there's no compelling reason to keep punie
using the Perl5 pmcs, I'll happily patch it to use core PMCs instead.
Given the discussion about using new PMCs for Perl6 behavior, I still
endorse making these dynamic, removing their special status. If we
can find a language to claim them, lets go one more step and move
them under languages/. Heck, we can do what was done with the python
PMCs - just move the pmcs now to languages/perl5 sans a currently
supported language with them, where someone might claim them at some
point.
My 2₫.
