Bug #88130 for Module-CPANTS-Analyse: Module::CPANTS::SiteKwalitee is missing?

Sun Aug 25 13:56:42 2013 ether [...] cpan.org - Ticket created

Subject:

Module::CPANTS::SiteKwalitee is missing?

Some comments refer to Module::CPANTS::SiteKwalitee, but this doesn't seem to exist: e.g. in Module::CPANTS::Kwalitee::Distname: # NOTE: The metrics in this module have moved to # Module::CPANTS::SiteKwalitee because these require an archived # distribution which you don't have while testing local Kwalitee # with Test::Kwalitee.

Tue Aug 27 15:01:08 2013 ishigaki [...] cpan.org - Correspondence added

On Mon Aug 26 02:56:42 2013, ETHER wrote: Show quoted text

> Some comments refer to Module::CPANTS::SiteKwalitee, but this doesn't > seem to exist: > > e.g. in Module::CPANTS::Kwalitee::Distname: > > # NOTE: The metrics in this module have moved to > # Module::CPANTS::SiteKwalitee because these require an archived > # distribution which you don't have while testing local Kwalitee > # with Test::Kwalitee.

It's github-only. https://github.com/cpants/Module-CPANTS-SiteKwalitee I'll provide links to github when things are more settled.

Tue Aug 27 15:01:09 2013 The RT System itself - Status changed from 'new' to 'open'

Tue Aug 27 15:43:15 2013 ether [...] cpan.org - Correspondence added

On 2013-08-27 12:01:08, ISHIGAKI wrote: Show quoted text

> On Mon Aug 26 02:56:42 2013, ETHER wrote:

> > Some comments refer to Module::CPANTS::SiteKwalitee, but this doesn't > > seem to exist: > > > > e.g. in Module::CPANTS::Kwalitee::Distname: > > > > # NOTE: The metrics in this module have moved to > > # Module::CPANTS::SiteKwalitee because these require an archived > > # distribution which you don't have while testing local Kwalitee > > # with Test::Kwalitee.

> > It's github-only. > > https://github.com/cpants/Module-CPANTS-SiteKwalitee > > I'll provide links to github when things are more settled.

a ha! FWIW, it might be worthwhile to split the Module::CPANTS::Kwalitee:: namespace out into a separate dist too -- that will really help cut down the MCA bug queue, as most of those tickets are related to specific kwalitee metrics, not the engine itself.

Tue Aug 27 21:30:30 2013 ishigaki [...] cpan.org - Correspondence added

On Wed Aug 28 04:43:15 2013, ETHER wrote: Show quoted text

> On 2013-08-27 12:01:08, ISHIGAKI wrote:

> > On Mon Aug 26 02:56:42 2013, ETHER wrote:

> > > Some comments refer to Module::CPANTS::SiteKwalitee, but this > > > doesn't > > > seem to exist: > > > > > > e.g. in Module::CPANTS::Kwalitee::Distname: > > > > > > # NOTE: The metrics in this module have moved to > > > # Module::CPANTS::SiteKwalitee because these require an archived > > > # distribution which you don't have while testing local Kwalitee > > > # with Test::Kwalitee.

> > > > It's github-only. > > > > https://github.com/cpants/Module-CPANTS-SiteKwalitee > > > > I'll provide links to github when things are more settled.

> > a ha! > > FWIW, it might be worthwhile to split the Module::CPANTS::Kwalitee:: > namespace out into a separate dist too -- that will really help cut > down the MCA bug queue, as most of those tickets are related to > specific kwalitee metrics, not the engine itself.

I believe splitting would do more harm than help, at least for now, because quite a few metrics depend on the information analyzed elsewhere in Module::CPANTS::Kwalitee/Analyse, and otherwise, it'll be ineffective as each metric file would need to traverse the directory etc (it actually does the traversal several times now, and I really don't like that). Besides, where are you going to open tickets to request new metrics after we split the distribution? :)

Wed Aug 28 12:34:19 2013 ether [...] cpan.org - Correspondence added

On 2013-08-27 18:30:30, ISHIGAKI wrote: Show quoted text

> I believe splitting would do more harm than help, at least for now, > because quite a few metrics depend on the information analyzed > elsewhere in Module::CPANTS::Kwalitee/Analyse, and otherwise, it'll be > ineffective as each metric file would need to traverse the directory > etc (it actually does the traversal several times now, and I really > don't like that).

Ok, interesting, I was actually wondering about that (what dependencies existed between metrics and what analysis needed to be done first). I wonder if those dependencies could be codified, so if a particular analyzer hadn't run, then a metric can make sure it gets run? Show quoted text

> Besides, where are you going to open tickets to request new metrics > after we split the distribution? :)

I was figuring the core metrics would stay together, in Module::CPANTS::Kwalitee, so that would be the gathering place for most tickets.