[Catalyst] RESTful perl implementations...

[Garrett Goebel]

On Sep 26, 2006, at 10:08 AM, John Napiorkowski wrote:

> -- Garrett Goebel <ggoebel at goebel.ws> wrote:
>
>> I'm sure this is too late to be useful. But here is
>> my subclassed
>> Catalyst::Action to implemented RESTful request
>> method based
>> dispatching. [I've been working on this in my free
>> time. Which I have
>> precious little of lately. My apologies.]
>>
>> The references to $c->request->path_parameters are
>> to support RoR
>> style processing of urls (foo.com/person;create)
>> where
>> Catalyst::Dispatcher->perpare_action is overridden
>> to match url paths
>> separate from path parameters and file extensions.
>>
>> I also override Catalyst::Dispatcher->setup_actions
>> to change the
>> default method_action_class to my subclass of
>> Catalyst::Action. I was
>> surprised that default method_action_class and
>> action_container_class
>> appear to be hard coded...
>>
> [snipped content]
>
> You can do this more easily as a Action Class, (see
> http://search.cpan.org/~mramberg/Catalyst-Runtime-5.7003/lib/ 
> Catalyst/Manual/Actions.pod
> for some documentation).  This was released with
> Catalyst 5.7x and easy to miss if you don't have time
> to watch the mailing list closely.

I don't really like :ActionClass. I'd rather my action method  
attributes declare what they match... not how.

>
> I wrote something similar and shared it with the list
> but there didn't seem to be much interest.  I've
> attached what I did for you to check out.  Maybe if
> this kind of this is useful to you we can work
> together to fix it up and write some tests and release
> it as a real module.  Let me know if you have time for
> that.
>
> I kind of wanted this to be more than just an action
> to match on the method type, I was hoping to have it
> automatically in/deflate the request body to something
> like XML::Simple or other XML object to make it easier
> to build true RESTful apps.  If would be cool if
> (something like)

I think there is considerable overlap in what we are trying to  
achieve. Unfortunately I don't think I have the time available right  
now to partake in an ongoing dialogue. I.e., I'd just slow you down  
at this point.

cheers,

Garrett