Carsten Ziegeler wrote:
Stephen McConnell wrote:I think we are saying the same thing.
Carsten Ziegeler wrote:
a) I think a block should also be able to expose resources directly, e.g.Carsten:
images.
Do you think it would be reasonable for access to a resource to be
expressed as a service. I.e. the client goes though an interface exposed
as a service to access the resource. This present the benefit that the
block (as the resource repository) does not need to be exposed to the
client. It also means that we are always dealing with the same
conceptual service access model.
Sorry, perhaps we are talking about the same thing, but I lost you here. I think, it makes sense to access resources (= files) via the block protocol, e.g.: <map:read src="block:skin://images/logo.jpg"/> Now, we could say that "images/logo.jpg" is resolved using a sitemap and therefore implemented as a pipeline, so resources are not served directory. Do you mean this (I guess not ;) )?
:-)
Assuming that we don't want to expose blocks in a classloader to the client - instead what we want is a expose of the services that a block provides to the client. However, as pointed out in several of the messages on this topic, there is a requirement for the overloading of the result of URI based resource resolution. Given the principal of seperation of the block (in terms of client accessibility), then a service is needed against which resources can be resolved based on supplied URIs (as distinct from classloader based access to resources).
Cheers, Steve.
Carsten
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, email: [EMAIL PROTECTED]
-- Stephen J. McConnell OSM SARL digital products for a global economy mailto:mcconnell@;osm.net http://www.osm.net --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, email: [EMAIL PROTECTED]
