On Thu, Aug 7, 2008 at 12:51 PM, Jonathan Worthington <[EMAIL PROTECTED]> wrote: > jerry gay wrote: >> >> seems to me this is trying to test whether there's a defined value in >> $!. > > Testing truth falls back to definedness, yes. > >> however, you've jumped straight into blaming parrot's Exception >> PMC, without considering the expected perl 6 behavior. >> > > Well, I was more blaming the *change* in Parrot's Exception PMC for > something failing that did work before, and was interested in the rationale > behind the change (which I now have and agree with). I wasn't claiming that > the current way we did things in Rakudo was correct, and if I thought this > was as simple as a bug in the Exception PMC rather than something needing > discussion I'd have written a patch, not a post on the mailing list. :-) > >> does Exception inherit from Object, > > At this point, no, and I tried making that happen earlier on today before > asking about this, but it doesn't solve the problem.. > >> and if so, doesn't Object's .true method by default return .defined? > > Yes, but we're calling the get_bool vtable method. Which is why making it > "inherit" from Object doesn't help. :-( > >> if so, then we should be wrapping Exception PMCs in perl 6 objects, and >> the inherited .true method will make this test work. >> > > Perhaps. I'm not sure exactly where the answer lies, but we need to be able > to somehow handle exceptions coming from other languages too. Trying to wrap > any exceptions that aren't Perl 6 ones into a more Perl 6-ish kinda thing > may be a solution, though I'm not sure how robust it is.
Add this to the pile of inter-HLL mapping issues in the other thread from a week or so ago. > Thanks, > > Jonathan > > -- Will "Coke" Coleda
