Dan Sugalski wrote:
Sam made reference to the CLS last week, and it's a point well-taken. We need a list of standard activities that parrot provides so that people writing compilers can use the provided facilities and people implementing data types can make sure they do the right things to work with the standards.
Before I go rattling this off and starting a PDD that'll likely languish a bit, does anyone want to step up and volunteer to put together a skeleton PDD with the current facilities, at which point we can start moving forward to fill in any holes and getting things detailed enough that it can be reasonably used?
To start it'd include the vtable and MMD functions, the ops that call 'em, and the methods we look for by default.
With that in mind, it is my feeling that such a document is premature. Unless somebody is much further along than I am aware of, we don't know where all the pitfalls are just yet; let alone knowing which of the pitfalls we can pave over, and which we simply need to post a big bold lighthouse at.
I can certainly wait. I'm not looking for a final document now, though, more a detailing of what we currently do. There are a lot of core facilites that we really want people writing compilers to use, but until they're detailed it's mostly luck if they are. (Well, and laziness -- we can be reasonably sure compilers will emit a mul op to do multiplication, but it's good to be explicit)
--
Dan
--------------------------------------it's like this-------------------
Dan Sugalski even samurai
[EMAIL PROTECTED] have teddy bears and even
teddy bears get drunk
