From @AlexDaniel

To understand the issue, let's take a look at examples that *do* work.

Code​:
my @​a = <a b c d e>; say @​a[2..*]

Result​:
(c d e)

Code​:
my @​a = <a b c d e>; say @​a[2..*]​:exists

Result​:
(True True True)

Code​:
my @​a = <a b c d e>; say @​a[*-2..*]

Result​:
(d e)

Alright? Now with “​:exists”​:

Code​:
my @​a = <a b c d e>; say @​a[*-2..*]​:exists

Result​:
(hangs forever, *-2..* is being reified)

From @zoffixznet

On Mon, 12 Dec 2016 11​:25​:01 -0800, alex.jakimenko@​gmail.com wrote​:

To understand the issue, let's take a look at examples that *do* work.

Code​:
my @​a = <a b c d e>; say @​a[2..*]

Result​:
(c d e)

Code​:
my @​a = <a b c d e>; say @​a[2..*]​:exists

Result​:
(True True True)

Code​:
my @​a = <a b c d e>; say @​a[*-2..*]

Result​:
(d e)

Alright? Now with “​:exists”​:

Code​:
my @​a = <a b c d e>; say @​a[*-2..*]​:exists

Result​:
(hangs forever, *-2..* is being reified)

Here are some comments from jnthn​: https://irclog.perlgeek.de/perl6/2016-12-12#i_13729493

FWIW, I'm not seeing a problem. You've contorted a piece of code into trying to reify an infinite number of elements...

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

From @AlexDaniel

Sooo… Are you saying that the behavior in provided examples is consistent?

On 2016-12-13 09​:25​:36, cpan@​zoffix.com wrote​:

On Mon, 12 Dec 2016 11​:25​:01 -0800, alex.jakimenko@​gmail.com wrote​:

To understand the issue, let's take a look at examples that *do*
work.

Code​:
my @​a = <a b c d e>; say @​a[2..*]

Result​:
(c d e)

Code​:
my @​a = <a b c d e>; say @​a[2..*]​:exists

Result​:
(True True True)

Code​:
my @​a = <a b c d e>; say @​a[*-2..*]

Result​:
(d e)

Alright? Now with “​:exists”​:

Code​:
my @​a = <a b c d e>; say @​a[*-2..*]​:exists

Result​:
(hangs forever, *-2..* is being reified)

Here are some comments from jnthn​:
https://irclog.perlgeek.de/perl6/2016-12-12#i_13729493

FWIW, I'm not seeing a problem. You've contorted a piece of code into
trying to reify an infinite number of elements...

From @zoffixznet

On Tue, 13 Dec 2016 11​:10​:38 -0800, alex.jakimenko@​gmail.com wrote​:

Sooo… Are you saying that the behavior in provided examples is consistent?

As far as I can see, yes. The first two are lazy Ranges. The second two are non-lazy WhateverCodes. And as jnthn++ pointed out "The array subscripter decides whether to auto-viv for not based on .is-lazy".

The reason it does so is so things like `my @​a; @​a[3..5] = 1 xx *; say @​a` autovivicate, while still letting you do stuff like `@​a[2..*]` to get a truncated result.

In the example you point out as problematic, you're exploiting the autovivication feature to create an infinite loop, hence the hang.

Is there still an issue on this ticket not covered by those explanations?

From @AlexDaniel

Yes. Why the behavior with and without :exists is different? Why does it have to be different?

On 2016-12-14 08​:02​:48, cpan@​zoffix.com wrote​:

On Tue, 13 Dec 2016 11​:10​:38 -0800, alex.jakimenko@​gmail.com wrote​:

Sooo… Are you saying that the behavior in provided examples is
consistent?

As far as I can see, yes. The first two are lazy Ranges. The second
two are non-lazy WhateverCodes. And as jnthn++ pointed out "The array
subscripter decides whether to auto-viv for not based on .is-lazy".

The reason it does so is so things like `my @​a; @​a[3..5] = 1 xx *; say
@​a` autovivicate, while still letting you do stuff like `@​a[2..*]` to
get a truncated result.

In the example you point out as problematic, you're exploiting the
autovivication feature to create an infinite loop, hence the hang.

Is there still an issue on this ticket not covered by those
explanations?