Open p6rt opened 9 years ago
say sprintf('%f %f %f %f', Mu, Any, Nil, NaN); # output: rakudo-moar ed47bb: OUTPUT«0.000000 0.000000 0.000000 0.000000» # rakudo-jvm ed47bb: OUTPUT«java.lang.NumberFormatException: For input string: "nan"» # NaN should numify to NaN for sure and so may Mu, Any and Nil
On Wed Sep 16 04:45:58 2015, gfldex wrote:
say sprintf('%f %f %f %f', Mu, Any, Nil, NaN); # output: rakudo-moar ed47bb: OUTPUT«0.000000 0.000000 0.000000 0.000000» # rakudo-jvm ed47bb: OUTPUT«java.lang.NumberFormatException: For input string: "nan"» # NaN should numify to NaN for sure and so may Mu, Any and Nil
The output for NaN is NaN now, the rest is unchanged:
$ perl6-m -e 'say sprintf("%f %f %f %f", Mu, Any, Nil, NaN);' 0.000000 0.000000 0.000000 NaN
The RT System itself - Status changed from 'new' to 'open'
My idea is to not allow Mu
as a value to sprintf
: each value should be Cool
in the RakuAST implementation of sprintf
in 6.e.
This would mean that a Mu
would cause a typecheck error.
Migrated from rt.perl.org#126067 (status was 'open')
Searchable as RT126067$