Re^9: goto superclass method

goto seems to do the expected thing when you give it a coderef.

$ perl -le'my $foo = sub { print "bar" }; sub x { goto $foo } print "$
+foo"; x'
CODE(0x813bc08)
bar
[download]

I don't think there's any doubt about what's happening here.

Oh, also, none of my previous examples change at all in behaviour if you use the goto &{ ... } notation instead.

Makeshifts last the longest.

Comment on Re^9: goto superclass method Download Code

Replies are listed 'Best First'.
•Re^10: goto superclass method by merlyn (Sage) on Dec 23, 2004 at 00:30 UTC
Well, then, we're back to my other explanation. You're not calling can("SUPER::bar"). You're calling SUPER::can("bar"), so you're finding yourself. Don't do that. Stop saying SUPER::can. It means practically nothing other than "can", unless you have a "can" method in the current class. -- Randal L. Schwartz, Perl hacker Be sure to read my standard disclaimer if this is a reply.	[reply]
Re^11: goto superclass method by Aristotle (Chancellor) on Dec 23, 2004 at 00:40 UTC
Something must have gotten lost along the way, because the fact that `$self->SUPER::can("bar")` is (usually) the same as `$self->can("bar")` is exactly what I was saying in the very first post… `:-)` (It can only be different if there is a `can` in one of the superclasses though, regardless of whether there is one in the current class.) Basically, I pointed out one of my two mistakes, then explained why it was a mistake, and came up with a broken solution because I missed something I thought I hadn't (that you can ask `UNIVERSALL::can` about `'SUPER::method'`). Then you came along and pointed out my other mistake as well, but said that it somehow invalidated my correct explanation for the mistake I had already spotted. Makeshifts last the longest.	[reply]