From @masak

<masak> r​: for our $​:: ();
<p6eval> rakudo dca0fa​: OUTPUT«===SORRY!===␤ResizablePMCArray​: index
out of bounds!␤»
* masak submits LTA error diakopter++ rakudobug

From @FROGGS

still the same​:

<FROGGS> r​: for our $​:: ();
<camelia> rakudo fe7049​: OUTPUT«===SORRY!===␤ResizablePMCArray​: index
out of bounds!␤»
<FROGGS> r​: for my $​:: ();
<camelia> rakudo fe7049​: OUTPUT«===SORRY!===␤ResizablePMCArray​: index
out of bounds!␤»

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

From @usev6

The internal error message has changed slightly (but is still LTA)​:

$ perl6-m -e 'for our $​:: ()'
===SORRY!===
MVMArray​: Index out of bounds

$ perl6-p -e 'for our $​:: ()'
===SORRY!===
QRPA​: index out of bounds

$ perl6-j -e 'for our $​:: ()'
===SORRY!===
VMArray​: Index out of bounds

STD complains about a missing block​:

$ viv -c -e 'for our $​:: ()'
===SORRY!===
Missing block at (eval) line 1 (EOF)​:
------> for our $​:: ()⏏<EOL>
  expecting initializer
Parse failed

But it looks as if the problem ('Index out of bounds') is not related to the for loop​:

$ perl6-m -e '$​::'
===SORRY!===
MVMArray​: Index out of bounds

With other sigils the same error occurs​:

$ perl6-m -e '@​​::'
===SORRY!===
MVMArray​: Index out of bounds

$ perl6-m -e '&​::'
===SORRY!===
MVMArray​: Index out of bounds

STD gives 'syntax OK" for those three. Especially it seems to parse '$​::' the same way as '$'.

From @coke

On Tue Oct 28 15​:16​:43 2014, bartolin@​gmx.de wrote​:

The internal error message has changed slightly (but is still LTA)​:

$ perl6-m -e 'for our $​:: ()'
===SORRY!===
MVMArray​: Index out of bounds

$ perl6-p -e 'for our $​:: ()'
===SORRY!===
QRPA​: index out of bounds

$ perl6-j -e 'for our $​:: ()'
===SORRY!===
VMArray​: Index out of bounds

STD complains about a missing block​:

$ viv -c -e 'for our $​:: ()'
===SORRY!===
Missing block at (eval) line 1 (EOF)​:
------> for our $​:: ()⏏<EOL>
expecting initializer
Parse failed

But it looks as if the problem ('Index out of bounds') is not related
to the for loop​:

$ perl6-m -e '$​::'
===SORRY!===
MVMArray​: Index out of bounds

With other sigils the same error occurs​:

$ perl6-m -e '@​​::'
===SORRY!===
MVMArray​: Index out of bounds

$ perl6-m -e '&​::'
===SORRY!===
MVMArray​: Index out of bounds

STD gives 'syntax OK" for those three. Especially it seems to parse
'$​::' the same way as '$'.

New behavior​:

15​:55 < [Coke]> r​: for our $​:: ();
15​:55 <+camelia> rakudo-{moar,jvm} 6a1879​: OUTPUT«===SORRY!=== Error while
  compiling /tmp/tmpfile␤Signatures as constraints on variables
  not yet implemented. Sorry. ␤at /tmp/tmpfile​:1␤------> for our
  $​:: ()⏏;␤»
15​:55 < GLRelia> rakudo-moar 287360​: OUTPUT«(timeout)»

15​:56 < [Coke]> m​: $​::
15​:56 <+camelia> rakudo-moar 6a1879​: OUTPUT«===SORRY!=== Error while compiling
  /tmp/qyCyFL1Z6t␤Variable '$' is not declared␤at
  /tmp/qyCyFL1Z6t​:1␤------> <BOL>⏏$​::␤»
15​:56 < GLRelia> rakudo-moar 287360​: OUTPUT«===SORRY!=== Error while compiling
  /tmp/chCtyvb9XT␤Variable '$' is not declared␤at
  /tmp/chCtyvb9XT​:1␤------> <BOL>⏏$​::␤»
--
Will "Coke" Coleda

From @smls

Is this still considered a BUG/LTA?

Isn't the error messages it prints now (see last comment by coke) pretty reasonable?

From @skids

On Tue, 22 Aug 2017 15​:35​:08 -0700, smls75@​gmail.com wrote​:

Is this still considered a BUG/LTA?

Isn't the error messages it prints now (see last comment by coke)
pretty reasonable?

I added a reminder to also handle anonymous vars in RT#​73104. The lingering issues
are just a sub-case of that problem.

Since we should not add tests for an X​::NYI on the signature constraint, and the
presence of the X​::NYI serves as ample codebase notation that the feature is
missing, I don't think keeping this ticket open serves any purpose. So I'm moving
it to 'resolved' status.

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