jackrabbit-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Adrian Custer <acus...@gmail.com>
Subject Building a repo using existing Db Schemas
Date Sat, 10 Nov 2007 11:23:30 GMT
Hey all,

This is, no doubt a heretical topic: I'd like to build a repository on a
database which uses an existing schema and then use both the repository
and the database. I see, a priori, no semantic reason one could not do
this if one restricts the use of the underlying database to be
read-only. And perhaps one could even ease that restriction...


Can any of you point me to literature (terminology, docs, threads, ...)
that discusses a hybrid repository/database system where the repository
is built against a Database for which we already have a schema? 

This, of course, goes against the grain of what content repository
systems like jackrabbit are about. Nonetheless you all probably deal
with this kind of question regularly and may have some stock answers, if
only to tell us not to mix water and oil. I've seen some hints on this
topic in the various docs I've read so far but have not yet found a good
analysis of the issues involved. 

We work on free software for geographic information systems (GIS). These
generally involve massive databases. So far, our libraries, apps and
methods are being built and 'optimized' for those databases and their
schemata (e.g. PostGIS on PostgreSQL). So we are not yet ready to switch
gears right now to a JCR repository system. However, we also are
starting to need a storage system for various metadata/runtime-info
classes so we may soon start to use both systems to have data on the
database and metainfo in a repository. During my current exploration I
am wondering about a possible, eventual, deeper integration of these two
content groups but I currently do not even have the words or concepts to
see how to examine the question further.


Any links or related information would be greatly appreciated,

--adrian


Mime
View raw message