Hi,
On 1/4/07, Fredrik Jonson <[EMAIL PROTECTED]> wrote:
I'm new to JCR and jackrabbit, and as far as the top down perspective
goes - from a content application perspective - I think I've understood
the basic concepts.
Now I wonder what's involved in publishing a JCR adapter - I hope I've
at least got the term right, I mean the back end interface to the data -
from a legacy content store?
Take any generic java webapp that has a custom database design with users
and data in a few simple database tables. What is the normal approach to
make a such backend store available through a JCR interface? Say at JCR
level 1.
Do I need to implement the (full?) JCR api from scratch? Or does
Jackrabbit supply any generic adapters that can be extended or something
similar? Where do I start?
While doable, implementing JCR adapters for legacy content stores is
by no means trivial. The SPI contrib project is currently building a
framework that should simplify the task.
In your case I would rather consider migrating to a full JCR
implementation like Jackrabbit instead of implementing a custom
adapter, but I understand that in many cases such drastic changes are
not justifiable.
BR,
Jukka Zitting
