Luke Kanies <[EMAIL PROTECTED]> writes:
> On Jul 21, 2008, at 1:24 PM, Adam Jacob wrote:
>> On Sun, Jul 20, 2008 at 5:02 PM, Daniel Pittman  
>> <[EMAIL PROTECTED]> wrote:
>
>> OK.  I don't quite understand the basis of your objection here; it
>> would help me if you could explain why you object.
>>
>> As far as I can see there is no sensible definition of, for example,
>> "puts", "fork" or "raise" during the template expansion process.
>>
>> Is your objection that you can see a meaningful use of those  
>> functions,
>> or simply that you don't like the (relatively) ugly steps required to
>> provide a clean evaluation environment in Ruby?
>>
>> It's both - I think that if I could forsee why someone wants to use
>> those functions in the template, I would be Psychic. :)

Fair enough.  I am, personally, always wary of any language feature
built on the basis of "someone might" that causes trouble for "someone
does", but ...

[...]

>> What about methods like 'rand', or split?

...these are certainly reasonable and valid objections.

>> Ruby's templating language *is* Ruby - anything that makes it less
>> like Ruby is harmful, in my opinion. 

My point of view on this is that we are talking about the *puppet*
scripting language, not the Ruby scripting language.  For the later,
Ruby, I completely agree with you.

For puppet I am less decided: a more restrictive language might make
sense, because it is not for general use -- it is for use in a specific,
restricted problem domain.

However...

[...]

>> Essentially, the behavior I would like to see is the equivilant of  
>> Erubis' context object support, with the context populated with all  
>> the available variables.  That would get you everything you're  
>> looking for, with the exception of throwing an exception if a  
>> variable is missing/undefined.
>
> I concur -- if there's a change, I think this should be the change.

... as noted, I will write appropriate testing for my instance variable
branch and resubmit that.[1]

Regards,
        Daniel

Footnotes: 
[1]  ...as noted elsewhere, I will probably also ask for Ruby to develop
     a hook to catch access to undefined member variables, so that this
     facility can be added back to Puppet with a sufficiently recent
     version of the interpreter.


--~--~---------~--~----~------------~-------~--~----~
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
-~----------~----~----~----~------~----~------~--~---

Reply via email to