Ok, we have overcome the administrative hurdles to getting
CTS into the CVS - I think.
Now we need to consider how much and how to integrate CTS
with Eclipse. There's 2 parts to this, server side and client
side. For now, let's concentrate on the client side.
So, CTS client infrastructure. Though really, in the long term,
it's not *client* infrastructure it's *consumer* infrastructure.
As a developer, and CTS consumer, I think I know what I want. I
imagine the CTS providers are plug-ins, using a generic CTS
interface. Multiple different CTS plug-ins provide different CTS
services, and the user/administrator/whatever decides which CTS they
want to use and how to configure it. I simply ask
if a CTS is available, and create it, and then use it. As an
alternative, I know which CTS I want and how to configure it, so
I ask for that CTS with those parameters to be created.
This requires a double plug-in structure - a host CTS manager which
provides CTS management - registration, arbitration, creation,
configuration, and a CTS provider plug-in which registers with the CTS
manager.
I suspect it makes sense for there to be more than CTS active,
but if there was, I'm not sure how I decide which one to use?
Maybe there's some sort of meta-query interface where you ask
the manager for a list of active CTS's that recognise a given
(domain/value set/something)?
What do you think? is any of this making sense?
Grahame
------------------------------------------------------------------------
_______________________________________________
ohf-dev mailing list
ohf-dev@xxxxxxxxxxx
https://dev.eclipse.org/mailman/listinfo/ohf-dev