Thorsten Scherler wrote:
> El jue, 10-08-2006 a las 21:10 +0100, Ross Gardler escribió:
>
>>Thorsten Scherler wrote:
>>
>>>El jue, 10-08-2006 a las 04:09 +0000, crossley@apache.org escribió:
>>>
>>>
>>>>Author: crossley
>>>>Date: Wed Aug 9 21:09:40 2006
>>>>New Revision: 430235
>>>>
>>>>URL: http://svn.apache.org/viewvc?rev=430235&view=rev
>>>>Log:
>>>>One demonstration of using Cocoon "running modes" to get a property into xconf.
>>>>See FOR-917
>>>
>>>
>>>I am not sure whether it is a good idea to have yet another properties
>>>system.
>>>
>>>The example you added seems to be a wee bit different then the
>>>ForrestConfModule regarding the usage within xconf files. With the
>>>"running modes" for example Cyriaque could implement the dynamic
>>>serializer configuration that he wanted to implement the other day.
>>>
>>>However we have now an overkill on properties systems (lm, defaults,
>>>project and runnig modes, did I forgot some?) in forrest and I do not
>>>think that is helping to make forrest simpler and easier to understand.
>>>
>>>Shouldn't we consider to focus one only one or two, document them well
>>>and harmonize their usage?
>>
>>+1, but which one?
>>
>
>
> In regards to my other mail I say:
> - locationmap for all properties that need to point to existing source
> - properties module (merge of defaults and project) aka forrestConf for
> all string based props.
I've not had time to digest that mail yet. Will respond in that thread
later (initial response seems positive).
Ross
|