Re: Perl 6 rules and complexity

It's not clear exactly what it will be, but it won't be completely recursive descent. Hopefully it will be as little recursive descent as possible.

Looking at PGE is probably the best way to gauge the algorithmic behavior of the grammar.

Comment on Re: Perl 6 rules and complexity

Replies are listed 'Best First'.
Re^2: Perl 6 rules and complexity by fergal (Chaplain) on Feb 17, 2006 at 01:25 UTC
Judging by Apocalypse 5 it won't be possible to execute arbitrary code in rule actions. At least I can't find any examples of anything besides manipulating "hypothetical" variables, so maybe it will be memoizable. Unfortunately after a quick look, I don't fancy digging through PIR, especially as that seems to be in very early stages and probably bearing no resemblance to it's final state.	[reply]
Re^3: Perl 6 rules and complexity by TimToady (Parson) on Feb 17, 2006 at 02:10 UTC
Any arbitrary code is allowed in rule actions, just as you can put any arbitrary code into a yacc grammar's reductions. Perl likes to optimize for power over purity unless you specify otherwise somehow.	[reply]