From @moritz

Rakudo r24153​:

# no error as expected
$ ./perl6 -e 'eval "my &foo :​:= { "foo" }; foo;"'

# get error message from the failed eval (since :​:= seems NYI)​:
$ ./perl6 -e 'eval "my &foo :​:= { "foo" }; foo;"; say $!'
invoke() not implemented in class 'Undef'

# trying to print the return value​:
./perl6 -e 'say eval "my &foo :​:= { "foo" }; foo;"'
Null PMC access in get_string()
current instr.​: 'print' pc 14225 (src/gen_builtins.pir​:8585)
called from Sub 'say' pc 14247 (src/gen_builtins.pir​:8595)
called from Sub '_block11' pc 74 (EVAL_13​:24)
called from Sub 'parrot;PCT;HLLCompiler;eval' pc 892
(src/PCT/HLLCompiler.pir​:508)
called from Sub 'parrot;PCT;HLLCompiler;command_line' pc 1441
(src/PCT/HLLCompiler.pir​:771)
called from Sub 'parrot;Perl6;Compiler;main' pc 16584 (perl6.pir​:168)
*** glibc detected *** corrupted double-linked list​: 0x097bfb58 ***
Aborted

That's certainly wrong ;-)

Moritz

From simon@simon-cozens.org

This causes a null access in PMC, presumably because the eval is
returning the wrong kind of nothing.

From @pmichaud

On Tue Dec 30 17​:46​:31 2008, simon@​simon-cozens.org wrote​:

This causes a null access in PMC, presumably because the eval is
returning the wrong kind of nothing.

This should now be fixed in r34682, assigning to moritz++ so that we can
verify there's a test for it.

Thanks!

Pm

The RT System itself - Status changed from 'new' to 'open'

From @moritz

On Tue Dec 30 18​:01​:29 2008, pmichaud wrote​:

On Tue Dec 30 17​:46​:31 2008, simon@​simon-cozens.org wrote​:

This causes a null access in PMC, presumably because the eval is
returning the wrong kind of nothing.

This should now be fixed in r34682, assigning to moritz++ so that we can
verify there's a test for it.

Added a test to t/spec/S29-scalar/undef.t

(it actually tests that uc(eval('')) lived, which used to die from the
same error).

Cheers,
Moritz

@moritz - Status changed from 'open' to 'resolved'