On 2017-08-23 02:46:46, SREZIC wrote:
Show quoted text> On 2017-08-22 16:33:21, SREZIC wrote:
> > See subject. Excerpt of test log:
> >
> > ...
> > # Failed test 'Eval expressions at user context and dump them'
> > # at t/27-cmd_e.t line 62.
> > # --- Got
> > # +++ Expected
> > # @@ -1,19 +1,19 @@
> > # '-e:0001 $x = 1;
> > # -e:0002 $x = [ a => 1 ];
> > # 1
> > # 1
> > # 1
> > # 2
> > # -e:0003 $x = { a => 1 };
> > # ["a", 1]
> > # -e:0004 @x = ( a => 1 );
> > # { a => 1 }
> > # -e:0005 %x = ( a => 1 );
> > # a 1
> > # 2
> > # ["a", 1]
> > # -e:0006 2;
> > # a 1
> > # -1
> > # +1/8
> > # { a => 1 }
> > # '
> > # Looks like you failed 1 test of 6.
> > t/27-cmd_e.t ...........
> > Dubious, test returned 1 (wstat 256, 0x100)
> > Failed 1/6 subtests
> > ...
>
> To be more precise, it fails on my systems since perl 5.25.3. 5.25.2
> is the last version with a PASS.
Probably it's the following change, from 5.26.0's perldelta:
"scalar(%hash)" return signature changed
The value returned for "scalar(%hash)" will no longer show information
about the buckets allocated in the hash. It will simply return the count
of used keys. It is thus equivalent to "0+keys(%hash)".
A form of backward compatibility is provided via
"Hash::Util::bucket_ratio()" which provides the same behavior as
"scalar(%hash)" provided in Perl 5.24 and earlier.
