Hi,

putting on my manifest design philosopher's hat :-)

On 06/18/2014 04:11 AM, Trevor Vaughan wrote:
> It does get hairy when you mix the callbacks with the reverse
> dependencies. I can see having to iterate over the stack (mesh?) and
> check for loops and correctness while doing so. It could certainly turn
> into quite the recursive memory hog if not done properly.

Yes, as Henrik mentioned, you'd basically need an advanced resolution
mechanism to it right. I predict that there may well be cases that the
system will not just get right.

> However, I really do think that this is needed in the language as it
> seems to be a place where people frequently stumble when putting
> together what should be a straightforward thought process.

While those are cool ideas (and I'm still rooting for a lazy
evaluator...anyone? no?) I'd like to contend the proposition that actual
support for this style of solving problems should be supported (and by
extension, recommended).

We are currently trying to teach users that in order to model a system,
they typically need to find a way to non-ambiguously define what does
and what does not need to be part of the catalog. This may often boil
down to introducing a flag value in Hiera or somesuch.

In other words, there seems to be a consensus that "reflection" (for
lack of a better term) in manifests should be avoided in favor of a
common data source in which to base your queries.

And yes, I do realize that there are edge cases where this is quite
painful in comparison to `if defined(Class[x])`, but I guess this is us
discussing if those are worth the burden of supporting the scheme.

Final aside (although I believe that you are aware of this, still got
the urge to point this out): 'Don't reverse these two' will work on the
node level, but as soon as you're two or three levels of `include` deep,
it's wishful thinking at best. One should never rely on evaluation
order. This is precisely why we need defined() gone.

...rereading, the whole message has a distinct Get Off My Lawn feel to
it - sorry. Guess I'm that dude by now :-)

Cheers,
Felix

-- 
You received this message because you are subscribed to the Google Groups 
"Puppet Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/puppet-dev/53A14820.3080907%40alumni.tu-berlin.de.
For more options, visit https://groups.google.com/d/optout.

Reply via email to