Q:PIR to nqp in HLL::Grammar

[kboga]

Passes all of nqp's tests & rakudo's current passing spectests.

[kboga]

I see you guys are porting over all pir & Q:PIR to nqp so maybe parts of this pull request could be used now?

notes:

• More recent changes should be hold into account e.g. additions to https://github.com/perl6/nqp/commits/master/src/HLL/Grammar.pm
• Maybe avoiding using parrot's VTABLE calls to ease porting wouldn't be bad as well.
• Could be optimized by using natives which didn't exist back then.
• Some pir::op's could now be translated to more recently added nqp::op's.
• Cursor's encapsulation shouldn't be broken.

[FROGGS]

pmichaud: to whoever is working on nqp #45
pmichaud: + my $opassoc := ~%opO<assoc>;
pmichaud: needs to be my str $opassoc

pmichaud as in, it might even be worth writing things like
pmichaud my %opO := %opOPER<O>
pmichaud to instead be
pmichaud my %opO := nqp::atkey(%opOPER, 'O')
pmichaud depending on how the translation actually takes place.
pmichaud (and to avoid any non-existence checks)

[kboga]

I'm sorry but I cannot work on this right now, just wanted to throw out that this pull request exists and that some bits could probably be cherrypicked from it. (although this being an 8 months old pull request).
And yes, the code needs all possible optimizations it could get.

[FROGGS]

Thats not what I meant. I just wanted to put that information there for ppl where that info isnt obvious, like me.

kboga notifications@github.com hat geschrieben:

I'm sorry but I cannot work on this right now, just wanted to throw out that this pull request exists and that some bits could probably be cherrypicked from it. (although it being an 8 months old).
And yes, the code needs all possible optimizations it could get.

---

Reply to this email directly or view it on GitHub:
#45 (comment)

[jnthn]

All of HLL::Grammar is now in NQP. I took a lot from this pull request, and also updated/optimized it in various places. Thanks! :-)