On Mon, Apr 2, 2012 at 08:54, Nigel Kersten <ni...@puppetlabs.com> wrote:
> So to translate from the bug report, it sounds like we're going to (go > back to?) representing :undef as nil inside Ruby? I think that is the right decision, but I wouldn't hold it completely decided. Someone could remember the horrible root cause of the problem that made us move away from that or something. > Whether that be templates or the Ruby DSL? > is that right? I think it is right from a language perspective, from a template perspective, and ultimately for our users. Having something that is "almost but not quite exactly" Ruby in the templates and Ruby DSL is horrible for someone coming to Puppet the first time. > Are we going to deprecate this for all of Telly and then change it in > the next major release? Given the amount of anecdotal pain we've had > in this area, is it feasible to admit we got this wrong and start > deprecating in 2.7.x ? I am inclined to say that we start to deprecate it through Telly, and by Telly + 1 we have nothing but `nil`. Really, this is enough pain and shock for people that I would almost count it as a bug we are fixing. Supporting both versions isn't too hard for users, either, since `if @foo.nil? or @foo == :undef then` is going to keep working as we shift away from `:undef` in the public interfaces. If we wanted to I would totally support pushing that back into 2.7, too, but I don't think it is necessary - especially because the reach of that patch release is never going to be that great. -- Daniel Pittman ⎋ Puppet Labs Developer – http://puppetlabs.com ♲ Made with 100 percent post-consumer electrons -- 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.
