On Jun 3, 3:44 am, Daniel Pittman <[email protected]> wrote:
> On Thu, Jun 2, 2011 at 10:42, James Turnbull <[email protected]> wrote:
> > Daniel Pittman wrote:
> >> On Thu, Jun 2, 2011 at 10:09, Jacob Helwig <[email protected]> wrote:
>
> >>> Just my personal opinion:  I think we're actually better off
> >>> implementing #4113[2] in the long-run.
>
> >> When I spoke to Luke about those options, his expectation would be
> >> that this was a single parameter on the package type, which would take
> >> a set of boolean options.  You could add a little language to that
> >> (like, say, the environment) that allowed options with arguments to be
> >> passed, and let each provider parse it themselves, but ... that sounds
> >> like an invitation to unmanageable data to me.
>
> >> In solving that I would strongly encourage you to address additional
> >> per-provider top level parameters and properties available, rather
> >> than just packing complex data into strings.
>
> > +1 - plus what Daniel said.
>
> I should clarify: I think that options, as Luke described them, do
> solve some genuine problems and might well be valuable in and of
> themselves.  I just don't want to see 'proxy=...', or worse,
> 'enablerepo=foo,bar,baz' packed into them.
>

Hello,

>From reading this thread it seems as though most people seem to think
#4113 is the way to go.

It would probably be wise to then close 2247 if this is the case.

I may throw my 2c into the 4113 ticket then. I may have a crack
at that one instead, however my ruby learning is just beginning
so it may be out of my scope for a while.

Nathan

-- 
You received this message because you are subscribed to the Google Groups 
"Puppet Developers" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to 
[email protected].
For more options, visit this group at 
http://groups.google.com/group/puppet-dev?hl=en.

Reply via email to