Skip Menu |

This queue is for tickets about the Crypt-DSA CPAN distribution.

Report information
The Basics
Id: 72371
Status: open
Priority: 0/
Queue: Crypt-DSA
People
Owner: Nobody in particular
Requestors: james4765 [...] gmail.com
Cc:
AdminCc:
Bug Information
Severity: (no value)
Broken in: (no value)
Fixed in: (no value)

History Show all quoted text
  Fri Nov 11 19:13:02 2011 james4765 [...] gmail.com - Ticket created
Subject: CPAN install missing dependency on Math::Pari
Date: Fri, 11 Nov 2011 19:12:53 -0500
To: bug-Crypt-DSA [...] rt.cpan.org
From: Jim Nelson <james4765 [...] gmail.com>
System: RHEL6 Perl version: 5.10 When installing from CPAN, there is no warning about needing Math::Pari, and the install fails with a cryptic error mesage if it is not installed.
  Mon Feb 25 18:12:04 2013 TODDR [...] cpan.org - Correspondence added
On Fri Nov 11 19:13:02 2011, james4765@gmail.com wrote: Show quoted text
> System: RHEL6 > Perl version: 5.10 > > When installing from CPAN, there is no warning about needing Math::Pari, > and the install fails with a cryptic error mesage if it is not installed.
Yeah, I actually had to add these deps to get it working. Math::BigInt::GMP: 0 Math::BigInt::Pari: 0
  Mon Feb 25 18:12:05 2013 The RT System itself - Status changed from 'new' to 'open'
  Mon Aug 26 17:37:07 2013 DANAJ [...] cpan.org - Correspondence added
Crypt::DSA 1.17 installed on a fresh perl 5.19.3 with no problem. I'm concerned about any dependency on Math::Pari -- I believe that is not desirable and certainly shouldn't be required by any of the code in 1.17. For performance it would be nice to install one of Math::BigInt::GMP or Math::BigInt::Pari, but other than adding an entry for recommend (which I believe all current tools ignore), I'm not sure what the best idea is. I haven't tested or looked in depth at Crypt::DSA, but my experience with Alt::Crypt::RSA::BigInt was that it works with the Calc/FastCalc backend, but is *very* slow. Ideally at install time we'd check if GMP is available and tack on Math::BigInt::GMP to the dependency list if so, but I never found a good way to do that.