I haven't done tests to explore the space, but I didn't see anything in that example to say that it wouldn't work the way a reasonable person might expect _in_addition_to_ the convoluted way shown in that example. My reading is that this is a very contrived (and pointless) example intended solely to demonstrate how you _could_ do something that you'd never really want or need to do.
To me (opinion subject to revision if it turns out that the straight forward form doesn't actually work) the real question is: can we think of a valid use case for referencing a defined type's metaparameters as variables? If so, let's use that as the example instead of this; if not, we should reconsider covering the "feature" in the tutorial in the first place, on the principle that there are an infinite number of such "you could do X but there's absolutely no reason to*" features and if we tried to cover even a small fraction of them it would rapidly become unwieldy, so we had best omit all of them. -- Markus * E.g. You could use the names of dead composers for your variable names; you could use word-wrap and right-justify your manifests; you could have an exec that notes that the run is about to start, and have all your resources require it; etc. They sound pointless (because they are) but so far as I can see they are no worse than this $require example. -- 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.
