On Mon Jun 04 08:27:46 2012, diakopter wrote:
> On Sun Jan 15 08:12:30 2012, jn...@jnthn.net wrote:
> > On Thu Nov 24 08:48:55 2011, gfldex wrote:
> > > > my $a; $a>>.();
> > > Method 'isa' not found for invocant of class 'Undef'
> > >
> >
> > Fixed:
> >
> > > my @a = -> { say 1 }, -> { say 2 }; @a>>.()
> > 1
> > 2
> >
> > Tagging as testneeded.
> >
> > Thanks,
> >
> > /jnthn
>
> broken in most recent rakudo. at the REPL, segfaults in linux if you
> press Ctrl-d after typing the above: my $a; $a>>.(); exits with error
> in windows repl if you type Ctrl-z<enter> after typing the above.
> unmarking testneeded; marking bug.
Now this runs fine (again). I tested on the command line and in the REPL on
Linux and FreeBSD. There are also the two following passing tests in
S03-metaops/hyper.t:
# RT #77670
{
is ( { 1 + 1 }, { 2 + 2 } ).>>.(),
(2, 4),
'.>> works with .()';
is ( { 1 + 1 }, { 2 + 2 } ).>>.(),
( { 1 + 1 }, { 2 + 2 } )>>.(),
'.>>.() means the same as >>.()';
}
Since I don't know how to write a proper test for the REPL, I'll mark the
ticket testneeded again.
As a side note: The command given by jnthn above has a different output now
$ perl6 -e 'my @a = -> { say 1 }, -> { say 2 }; @a>>.()' ## output was 1\n2
2
1
AFAIU this happens since commit
https://github.com/rakudo/rakudo/commit/1d4eb481ef5dc4e772042c896cda4a5b7b510d0c
and is intenionally (cmp. the comments in said commit). The output of the
following command shows the non-sequential ordering:
$ perl6 -e 'my @a = 1..6; @a>>.say;'
6
4
2
5
3
1
