Bug #85284 for Statistics-CaseResampling: Do not bundle Mersenne Twister implementation

Tue May 14 08:08:05 2013 ppisar [...] redhat.com - Ticket created

Subject:

Do not bundle Mersenne Twister implementation

Current code bundles Mersenne Twister generator. This is not good for maintenance (e.g. upstream provides newer and faster versions). Please use Math::Random::MT or Math::Random::MT::Auto instead.

Tue May 14 12:27:05 2013 smueller [...] cpan.org - Correspondence added

Thanks, applied. Will hopefully cut a release soon. Feel free to remind me if I forget.

Tue May 14 12:27:05 2013 The RT System itself - Status changed from 'new' to 'open'

Tue May 14 12:27:06 2013 smueller [...] cpan.org - Status changed from 'open' to 'resolved'

Tue May 14 12:36:32 2013 smueller [...] cpan.org - Correspondence added

Okay, I actually wanted to close the other bug. I blame slow-as-molasses web interfaces via 3G... ;) The reply to THIS ticket is a bit different: I understand the reason why Linux distribution dislike any form of code duplication. I even sympathize. But your remark that this module should just use another completely misses the fact that this is using the MT at the C level. I will keep the ticket open because I can pull in the updated version of the MT. Furthermore, I am happy to accept patches to make this use an MT implementation that lives elsewhere and does so *efficiently*. More specifically, if a given Linux distro ships a system library that this can use, I'm happy to have this use said library if available - but right now, I don't have time and interest to do the appropriate fuzzing with Makefile.PL. Best regards, Steffen

Tue May 14 12:36:32 2013 smueller [...] cpan.org - Status changed from 'resolved' to 'stalled'

Tue May 14 12:37:14 2013 smueller [...] cpan.org - Severity Wishlist added