On Sep 18, 2009, at 4:42 PM, Luke Kanies wrote:
> I've been getting the feeling recently that we haven't been keeping > the language updated well enough with problems in the domain, so being > more flexible about what problems we should be trying to solve can > help resolve some of that. > > Any other opinions? As the instigator of this, I'll chime in and say the existing behaviour is OK as long as (a) the precedence rules are more clear (James has already added them to the relevant wiki pages) and (b) it's made more explicit that anything beyond trivial static nodes really ought to be pushed out to an external classifier. IMO it would greatly help (b) if such a classifier were included in source, a'la this thread from last year regarding nodify: http://www.mail-archive.com/puppet-dev@googlegroups.com/msg00038.html Unless I misssed it somewhere, nodify doesn't actually exist. There's a couple of classifier examples on the ExternalNodes wiki page and the book, but they're a bit too simplistic to be useful (plus they're not in Ruby). Since this thread made it clear I was going about nodes the wrong way, I've written an ENC that should hopefully be general enough for other people to use. I'll work on getting it cleaned up and submitted if others agree it'd be helpful. -=Eric --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "Puppet Developers" group. To post to this group, send email to puppet-dev@googlegroups.com To unsubscribe from this group, send email to puppet-dev+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/puppet-dev?hl=en -~----------~----~----~----~------~----~------~--~---
