Bug #111813 for Math-MPFR: t/LongDouble.t fails with latest Math::LongDouble

Fri Feb 05 16:01:42 2016 SREZIC [...] cpan.org - Ticket created

Subject:

t/LongDouble.t fails with latest Math::LongDouble

If Math::LongDouble 0.18 is installed, then t/LongDouble.t fails: $man: 9.99999999999999999950 $man: 1.00000000000000000005 t/LongDouble.t ............ Failed 2/11 subtests

Fri Feb 05 18:54:10 2016 sisyphus [...] cpan.org - Correspondence added

Thanks Slaven. Math::LongDouble now, by default, prints out extra digits. Prior to version 0.16 it printed out LDBL_MANT_DIG (most commonly 18) digits, but I've changed that to the more generally accepted 1 + ceil(P*log(2)/log(10)) - where P is the precision of the long double (most commonly 64). The value of that expression is usually 21 - and t/LongDouble.t needs to be amended to cater for this change in Math::LongDouble behaviour. This should be fixed in github: https://github.com/sisyphus/math-mpfr I've yet to run the final checks on the github version of Math::MPFR and get it uploaded to CPAN. Cheers, Rob

Fri Feb 05 18:54:11 2016 The RT System itself - Status changed from 'new' to 'open'

Mon Feb 08 06:31:44 2016 sisyphus1 [...] optusnet.com.au - Correspondence added

Subject:	Re: [rt.cpan.org #111813] t/LongDouble.t fails with latest Math::LongDouble
Date:	Mon, 8 Feb 2016 22:30:58 +1100
To:	<bug-Math-MPFR [...] rt.cpan.org>
From:	<sisyphus1 [...] optusnet.com.au>

Show quoted text

-----Original Message----- From: Sisyphus via RT Sent: Saturday, February 06, 2016 10:54 AM To: sisyphus1@optusnet.com.au Subject: [rt.cpan.org #111813] t/LongDouble.t fails with latest Math::LongDouble

> I've yet to run the final checks on the github version of Math::MPFR and > get it uploaded to CPAN.

t/LongDouble.t required some additional tweaks. All now done, and Math-MPFR-3.31 has been uploaded to CPAN. Cheers, Rob

Mon Mar 23 02:50:22 2020 SREZIC [...] cpan.org - Correspondence added

On 2016-02-05 18:54:10, SISYPHUS wrote: Show quoted text

> Thanks Slaven. > > Math::LongDouble now, by default, prints out extra digits. > Prior to version 0.16 it printed out LDBL_MANT_DIG (most commonly 18) > digits, but I've changed that to the more generally accepted 1 + > ceil(P*log(2)/log(10)) - where P is the precision of the long double > (most commonly 64). > The value of that expression is usually 21 - and t/LongDouble.t needs > to be amended to cater for this change in Math::LongDouble behaviour. > > This should be fixed in github: > https://github.com/sisyphus/math-mpfr > > I've yet to run the final checks on the github version of Math::MPFR > and get it uploaded to CPAN.

Recent versions of Math-MPFR do not seem to show this failure anymore (checked http://matrix.cpantesters.org/?dist=Math-MPFR%204.13 ), so I assume this issue may be resolved?

Mon Mar 23 03:56:35 2020 sisyphus [...] cpan.org - Correspondence added

On Mon Mar 23 02:50:22 2020, SREZIC wrote: Show quoted text

> On 2016-02-05 18:54:10, SISYPHUS wrote:

> > Thanks Slaven. > > > > Math::LongDouble now, by default, prints out extra digits. > > Prior to version 0.16 it printed out LDBL_MANT_DIG (most commonly 18) > > digits, but I've changed that to the more generally accepted 1 + > > ceil(P*log(2)/log(10)) - where P is the precision of the long double > > (most commonly 64). > > The value of that expression is usually 21 - and t/LongDouble.t needs > > to be amended to cater for this change in Math::LongDouble behaviour. > > > > This should be fixed in github: > > https://github.com/sisyphus/math-mpfr > > > > I've yet to run the final checks on the github version of Math::MPFR > > and get it uploaded to CPAN.

> > Recent versions of Math-MPFR do not seem to show this failure anymore > (checked http://matrix.cpantesters.org/?dist=Math-MPFR%204.13 ), so I > assume this issue may be resolved?

Yes - sorry about the oversight. I think this ticket will be closed as soon as I send this reply .... not entirely sure .... Cheers, Rob

Mon Mar 23 03:56:36 2020 sisyphus [...] cpan.org - Status changed from 'open' to 'resolved'