At 05:08 PM 2/8/2007 -0800, Jeffrey Harris wrote:
Hi Phillip,
Morgen and I were just discussing that we're going to need a little more
infrastructure to make modification export work smoothly.
It's possible to have a recurring event whose master is just an event,
but one of whose modifications is, say, a task.
Morgen's current plan is to have only master events passed in to the
translator for export. The translator should then yield appropriate
modification records (if there are any).
So it seems the event_export method should have some pluggable mechanism
for generating all modifications, so if the event is recurring,
event_export can iterate over modifications and pass them off to be
checked for stamp modification record exporters.
Does this seem reasonable?
Well, you could have a second Translator for that. Your translator's
__init__ could create an instance of the other translator and store it in
your translator. Then your event exporter could pass items to the other
translator.
That might be a bit of overkill though, because it seems to me that you
could probably just call back to the *same* translator (i.e. self), and
detect whether you should be issuing modification records or regular
records based on whether the item is a master event or not.
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
Open Source Applications Foundation "chandler-dev" mailing list
http://lists.osafoundation.org/mailman/listinfo/chandler-dev
