On May 28, 2010, at 12:46 PM, Markus Roberts wrote:
> Comments?
Yes.
A few thoughts I had on first read through: Is this a use case for
hashes? Should it be a passthrough / override setup (so that the
module write can provide defaults for values the user does not
provide)? At what epistemological level is the data path determined?
I'm sure hashes could be used here, but you still need external data,
regardless of the format.
Defaults should probably be shipped as a file in the module, IMO, but
I could see other arguments. Other options might be specifying the
defaults inline, but I would expect the defaults themselves to have
branching, so I would expect to actually see multiple default files
(e.g., one each for debian and red hat).
There are two ways to define data path: Where you look for data, and
how you decide which data files to load. E.g., maybe you look for
data in $confdir/data then $module/data, which is one kind of data
path. Also, though, you might look for data in $os-$osversion.yaml
then $os.yaml then $default.yaml.
The former should be an option in puppet.conf; the latter needs to be
configurable per-class, unfortunately; maybe:
class foo loads_from(["$os-$osversion", "$os", "default"]) { }
Or something similar.
--
Man is ready to die for an idea, provided that idea is not quite
clear to him. --Paul Eldridge
---------------------------------------------------------------------
Luke Kanies -|- http://puppetlabs.com -|- +1(615)594-8199
--
You received this message because you are subscribed to the Google Groups "Puppet
Developers" group.
To post to this group, send email to puppet-...@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.
