Skip Menu |

This queue is for tickets about the Math-Prime-Util CPAN distribution.

Maintainer(s)' notes

The latest version is in https://github.com/danaj/Math-Prime-Util

    git clone --origin upstream git://github.com/danaj/Math-Prime-Util.git Math-Prime-Util

Comments, issues, and patches are welcome at either RT or github, or send me email.

Report information
The Basics
Id: 110785
Status: rejected
Priority: 0/
Queue: Math-Prime-Util
People
Owner: DANAJ [...] cpan.org
Requestors: SREZIC [...] cpan.org
Cc:
AdminCc:
Bug Information
Severity: (no value)
Broken in: 0.56
Fixed in: (no value)

History Show all quoted text
  Wed Dec 30 12:24:54 2015 SREZIC [...] cpan.org - Ticket created
Subject: t/81-bignum.t fails if newest Math::BigFloat/Math::BigInt is installed
Statistical analysis suggests that version 1.999712 of Math::BigFloat and/or Math::BigInt is causing the failures in t/81-bignum.t (negative theta is bad): **************************************************************** Regression 'mod:Math::BigFloat' **************************************************************** Name Theta StdErr T-stat [0='const'] 1.0000 0.0000 12418154965301672.00 [1='eq_1.9991'] -0.0000 0.0000 -0.73 [2='eq_1.9993'] 0.0000 0.0000 0.61 [3='eq_1.9997'] -0.0000 0.0000 -0.28 [4='eq_1.999707'] -0.0000 0.0000 -0.63 [5='eq_1.999709'] -0.0000 0.0000 -0.61 [6='eq_1.999710'] 0.0000 0.0000 0.00 [7='eq_1.999712'] -1.0000 0.0000 -10951783260150046.00 R^2= 1.000, N= 55, K= 8 **************************************************************** **************************************************************** Regression 'mod:Math::BigInt' **************************************************************** Name Theta StdErr T-stat [0='const'] 1.0000 0.0000 8630839465939284.00 [1='eq_1.89_01'] -0.0000 0.0000 -1.69 [2='eq_1.9993'] -0.0000 0.0000 -0.45 [3='eq_1.9997'] -0.0000 0.0000 -1.07 [4='eq_1.999707'] -0.0000 0.0000 -0.69 [5='eq_1.999709'] -0.0000 0.0000 -0.90 [6='eq_1.999710'] -0.0000 0.0000 -0.93 [7='eq_1.999712'] -1.0000 0.0000 -8073411060447944.00 R^2= 1.000, N= 55, K= 8 ****************************************************************
  Wed Dec 30 13:37:08 2015 DANAJ [...] cpan.org - Correspondence added
On Wed Dec 30 12:24:54 2015, SREZIC wrote: Show quoted text
> Statistical analysis suggests that version 1.999712 of Math::BigFloat > and/or Math::BigInt is causing the failures in t/81-bignum.t (negative > theta is bad):
Thanks. It looks like it is a combination of the newest Math::BigInt in combination with Math::BigInt::GMP. It looks like the latest Math::BigInt has broken that as well, so not a big surprise that we fail if that module doesn't work. $ perl -Mntheory=:all -E 'use Math::BigInt lib=>"GMP"; use Math::BigFloat lib=>"GMP"; my $n = Math::BigInt->new(31415926535897932384); $n = Math::BigFloat->new($n); say $n; my $result = RiemannR($n)+0.5; say $result; say $result->copy->bfloor->bstr; say Math::BigInt->new($result->copy->bfloor->bstr)' 31415926535897900000 716115441142293915.0210866059126536709524 716115441142293915.0000000000000000000000 0 or more to the point: perl -E 'use Math::BigInt lib=>"GMP"; say Math::BigInt->new("716115441142293915.0000000000000000000000")' 0 perl -E 'use Math::BigInt lib=>"GMP"; say $Math::BigInt::VERSION; say $Math::BigInt::GMP::VERSION;' 1.999712 1.46 I get the same result with 1.47 (which doesn't pass tests). Looks like something to fix in Math::BigInt::GMP.
  Wed Dec 30 13:37:09 2015 The RT System itself - Status changed from 'new' to 'open'
  Wed Dec 30 13:37:09 2015 DANAJ [...] cpan.org - Taken
  Wed Dec 30 14:24:19 2015 DANAJ [...] cpan.org - Correspondence added
This is a bug in Math-BigInt-1.999712: https://rt.cpan.org/Ticket/Display.html?id=110792 This ticket was useful in tracking the issue down and filing that ticket. I don't know if PJACKLAM was already fixing it, but I think the info should be helpful. Thanks!
  Wed Dec 30 14:24:21 2015 DANAJ [...] cpan.org - Status changed from 'open' to 'rejected'
  Wed Dec 30 14:34:37 2015 SREZIC [...] cpan.org - Reference to ticket #110792 added