From @ilyabelikin

grammar G {
  token TOP { '​:' <foo>+ };
  token foo { \d };
}

my $a = "​:123" ~~ G​::TOP;
say +$/; # 0 But match success. :(
say +$a; # 1

say $/.WHAT; # Match
say $a.WHAT; # Hash (!?)

# Hash, hmmm. I expect Match object.

say $/.perl; # {"foo" => [{}, {}, {}]}
say $a.perl; # {"foo" => [{}, {}, {}]}

# Strange hash

say $/<foo>.WHAT; # List
say $a<foo>.WHAT; # List

From @chrisdolan

I'm seeing this too. It happened in the last 3-5 days, I think, about
the same time as the ObjectRef overhaul. I thought it was a mistake I
introduced in my own code, so I did not report it...

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

From @pmichaud

On Mon, Nov 10, 2008 at 03​:21​:47AM -0800, Ilya Belikin wrote​:

grammar G {
token TOP { '​:' <foo>+ };
token foo { \d };
}

my $a = "​:123" ~~ G​::TOP;

The official syntax for matching against a rule or grammar
has not yet been determined. So, I'm marking this ticket
as "stalled" until we have that.

In the meantime, I suggest​:

  my $a = G​::TOP("​:123");

Pm

@pmichaud - Status changed from 'open' to 'stalled'

From @chrisdolan

I think I'm getting closer to the root cause of this regression -- the
smart match is not part of the problem. Instead there appears to be
something wrong in the recent ObjectRef overhaul. Consider this​:

  ./perl6 -e 'grammar Foo {token TOP {'foo'};}; say Foo​::TOP("foo").WHAT'
  Match

  ./perl6 -e 'grammar Foo {token TOP {'foo'};}; my $m=Foo​::TOP("foo");
say $m.WHAT'
  Hash

even more revealing​:

  ./perl6 -e 'say PGE​::Match.new("foo").WHAT'
  Match

  ./perl6 -e 'my $m = PGE​::Match.new("foo"); say $m.WHAT'
  Hash

Any guesses whether this is a bug in assignment or in PGE​::Match? I'm
guessing the latter, so I'll start looking there.

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

From @chrisdolan

More news​: using 'infix​::=' works around the regression.

  ./perl6 -e 'my $m := PGE​::Match.new("foo"); say $m.WHAT'
  Match

  ./perl6 -e 'my $m = PGE​::Match.new("foo"); say $m.WHAT'
  Hash

I now suspect the flaw is in assign.pir. I stuck a ton of debugging
"say" lines in there. The following code is in the 'infix​:=' method​:

  $I0 = can source, 'Scalar'
  unless $I0 goto have_source
  source = source.'Scalar'()

The $m is an ObjectRef and it does has a 'Scalar' method. After the
source.'Scalar'() invocation, the resulting source returns true to "isa
source, 'Perl6Hash'".

If I comment out the "source = source.'Scalar'()" line, then the code works

  ./perl6 -e 'my $m = PGE​::Match.new("foo"); say $m.WHAT'
  Match

From @pmichaud

This issue should now be fixed in r32597 so that PGE​::Match objects
correctly return an ObjectRef and assigns properly into scalar types.

Also, in order to help diagnose such issues in the future, I've added a
.PARROT method to rakudo that can be used to better diagnose what is
happening internally when difficulties arise. This should be a more
reliable indicator than trying to use .WHAT, which often hides such
differences. For example

  $ ./parrot perl6.pbc
  > my $a = 'foo' ~~ /./; say $a.WHAT; say $a.PARROT;
  Match
  ObjectRef->PGE;Match
  > my $a = 'foo' ~~ /./; say ($a).WHAT; say ($a).PARROT;
  Str
  String

Here we can see that even though .WHAT returns the expected Perl 6
protoobject, .PARROT shows us what is really happening beneath the surface.

I expect that most of the remaining discrepancies will resolve
themselves when we get HLL mapping in place.

Resolving ticket, thanks!

Pm

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