jackrabbit-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Felix Meschberger <fmesc...@gmail.com>
Subject Re: Jackrabbit 3: repository requirements
Date Wed, 10 Feb 2010 14:34:17 GMT

On 10.02.2010 15:16, Alexander Klimetschek wrote:
> On Tue, Feb 9, 2010 at 23:04, Guo Du <mrduguo@gmail.com> wrote:
>> Architecture for modular design
>> * Other way to configure repository than repository.xml.
> See https://issues.apache.org/jira/browse/JCR-1438
>> * Get implementation/resource/service from OSGi environment
> There is some stuff happening wrt to LoginModules in Sling. And it
> should be done alongside JCR-1438. I added a note to the bug.

Actually I am working on such a thing in a Jackrabbit sandbox at

The goal is to have a jackrabbit-core library which is also a bundle and
which supports the same extensibility as we currently have in the
Jackrabbit Server bundle over at Apache Sling.


>> Scalability
>> * Scale to complexity. Add graph style operation, e.g. a blog post may
>> have multiple path to retrieve: date, author, tag.
> Did you look at shareable nodes in JCR 2.0?
> I think Jackrabbit should be careful and not add new API-level
> features that are not standardized.
>> Connector interface to other system
>> * Expose other system via jcr api. e.g. S3, google, source control system.
> This is something for which jcr spi is intended for. The jcr2spi
> provides the transient client implementation and you "only" have to
> implement the simpler spi interface. Also, via spi a lot of common
> stuff (node type templates, namespace parsing, etc.) is already
> available to simplify implementation from scratch.
> I don't think this is necessarily important for jackrabbit core.
> Regards,
> Alex

View raw message