jackrabbit-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Alexander Klimetschek" <aklim...@day.com>
Subject Re: webdav integration
Date Thu, 17 Jul 2008 16:16:27 GMT
On Thu, Jul 17, 2008 at 5:06 PM, Michael Harris
<michael.e.harris@gmail.com> wrote:
> thanx for all your help.
>
> So there are two things here -- one is the webdav component of the app.  JR
> isn't going to be involved in this; we are using it just for prototyping,
> although we will use the JR-webdav client to access the sharepoint repo the
> organization uses.
>
> the second is the storage of our app content.  This content is xml.  the app
> is basically a web based xml editor.  we were thinking of using JR to store
> the xml file, as we have requiremens around versioning and locking.  So, if
> a repository is bundled with the app (deployment method one), do servlets
> even come into play?  We are using Struts as our web tier, and were going to
> wrap services around the repository functionality (get a file, lock a file,
> version a file, save a file, lock a file, etc).  In that approach, would the
> service layer be able to access the repo?  I assume we would have to build
> some functionality around managing the sessions that provide access ot the
> repo.

Yes, no problem, your services use the JCR API and map the service calls to it.

But I could imagine that using the WebDAV server already gives all you
need (since you mentioned your services work with files). In that case
you "only" need a WebDAV client - you don't need to do much on the
server side then.

You should also have a look at Sling, which is a thin web application
framework on top of JCR. It provides XML and JSON views of the nodes
and properties in a RESTful way. Files in the repository (nt:file) are
delivered as in a static webserver. And developing with Sling makes
much more fun than Struts ;-):

http://incubator.apache.org/sling/

http://incubator.apache.org/sling/site/links.html


> or am i totally off here?
>
> Every "deploy in app server" wiki entry assumes oracle.  they all have setup
> jndi with oracle parts.  Do we need a db?  Doesn't the repo run on disk (or
> i should say, can't jackrabbit just run on the filesystem?)

You can configure a so-called PersistenceManager for storing the data.
You can use the Derby bundle persistence manager (default
configuration in most repository.xmls) that uses an embedded Apache
Derby database. Have a look at

http://jackrabbit.apache.org/jackrabbit-configuration.html

and

http://wiki.apache.org/jackrabbit/PersistenceManagerFAQ

and

http://wiki.apache.org/jackrabbit/DataStore

(DataStore is a separate and faster storage for binary data, such as
files stored in the repository).

Regards,
Alex

-- 
Alexander Klimetschek
alexander.klimetschek@day.com

Mime
View raw message