Accepting a Context allows us to use custom resolvers for variable values... such as the ObjectContext will provides the ability to use java getters for the values. I'd suggest that we go ahead and keep it.

- James

Dan Diephouse wrote:
I'm looking at this code again. Does Route.expand really need to take a Context? It seems like all it should ever really need is a Map.

James - thoughts?
- Dan

Jim Ancona (JIRA) wrote:
[ https://issues.apache.org/jira/browse/ABDERA-95?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12564498#action_12564498 ]
Jim Ancona commented on ABDERA-95:
----------------------------------

The problem with my fix is that the third parameter to RouteManager.urlFor() is 
an Object and my fix only handles the case where it is a Map. If you look at 
RouteManager.getContext(), you'll see that it handles two other cases (Contaxt 
and Object). I couldn't see  a general way to add the extra parameters to the 
Context. (Contexts don't even have the equivalent of Map.getKeys(), so you 
can't find out what in them in order to merge them.) My code will work in the 
current case, but would fail if a caller passed something other than a Map to 
RouteManager.urlFor() .

This look to me to be better than the servicesPattern approach. I had tried an 
approach using code that I wrote to parse a URI according to a URI template, 
but as James said, you can't do it in the general case, and besides the 
Principle of Least Power[1] applies: Routes are less powerful than templates 
but still solve the problem cleanly and therefore are more desirable in this 
case.

[1] - http://www.w3.org/2001/tag/doc/leastPower.html

DefaultProvider doesn't properly handle parameters in base routes
-----------------------------------------------------------------

                Key: ABDERA-95
                URL: https://issues.apache.org/jira/browse/ABDERA-95
            Project: Abdera
         Issue Type: Bug
           Reporter: Jim Ancona
        Attachments: CustomerAdapterTest.patch, RouteManager_workaround.patch


I spent some time trying to debug this last night, but didn't find the 
solution. I will attach a unit test patch that reproduces the problem.



--
Dan Diephouse
MuleSource
http://mulesource.com | http://netzooid.com/blog

Reply via email to