I'm closing this ticket, cmp. https://github.com/rakudo/rakudo/issues/1797:
> #124552 can be closed. Test was working but skipped. Unskipped in roast now.
Test PR for rakudo tree: https://github.com/rakudo/rakudo/pull/1806
If that is accepted, someone with a jvm build should check whether it needs to
be fudged for rakudo-j
and if it is fine there as well, this ticket can be closed.
Golf and change of behavior:
$ perl6 -e 'use NativeCall; class T is repr is export { sub
new(size_t $n) returns T is symbol is native { * }; }; T.new();'
$ perl6 -e 'use NativeCall; class T {...}; class T is repr is export
{ sub new(size_t $n) returns T is symbol is native { * }; }; T.new();'
=
This is a very old ticket and we no longer have postcircumfix:<( )>
for objects.
If you redo the example to use CALL-ME it still does not call the
original class method, however, CALL-ME is defined as a submethod,
not a method, and testing the difference:
$ perl6 -e 'class S { method x { "Sx".sa
Well, the current test file succeeds in both cases. This ticket is so old
it may more be the tests have changed than anything else, but there was
also lot of charset work so maybe it is actually fixed.
If you can still figure out how to break something with LC_ALL,
please re-file a github issue
spam rejected
spam rejected
This is also an issue in nqp.
$ nqp -e 'grammar f { regex TOP { ^ $ }; regex foo($i) { .. } };
nqp::say(f.parse("aaa"));'
Too few positionals passed; expected 2 arguments but got 1
Fixing it in nqp first is probably the best first step. To
that end I investigated some and it looks like this w