Thanks for both responses. Comments below:

David Nuescheler wrote:

hi robert,

i assume that there is a reason why a WebDAV network share is not
feasible for your usecase. is that right?

Yes, that's right .. because I want to support a local-copy, disconnected model. But, my question was implicitly directed towards the WebDAV, RMI and related efforts.

since using webdav would allow you to "simulate" the filesystem right
"inside" the repository and would avoid the necessity to sync it up altogether.


Yeah, this approach will solve a lot of my problems, but not all. I'm actually writing three web-based apps (2 for work/1 personal) which will require general filesystem-JSR170 sync features so I figured this would be a common CMS problem.

The personal app (photo app) is a bit different from the other two, but similar enough to get an idea. I want a local service which listens to a 'My Albums' directory. Then, when the likes of my mother creates a sub-directory in Explorer and drops in images; she essentially creates a web-based album (no GUI involved). So, this client app in question picks up the changes to this directory, scales the images and creates the 'album' (mainly a mirrored view of the album dir) in CMS. If the local system is not connected, it will just queue up the changes. I want to scale images before the sync mainly to save server diskspace. So, this client app would be based on some filesystem listener/filtering service and some JSR170 remote framework (WebDAV/RMI)

The other two apps do no processing - just pure syncing - which is why I wanted to start with a general solution.

Thanks again for the responses.

-Robert



you can test the repository webdav features by pointing your
webdav client (eg. ms webfolders) at:
http://jsr170tools.day.com/crx/repository/

regards,
david

On 9/27/05, Michael Wechner <[EMAIL PROTECTED]> wrote:
Robert Ritchy wrote:

I'd like to write/find an app that will keep the contents of an entire
directory (i.e. My Documents) in sync with a JSR170 repository (I'm
using jackrabbit). In other words, when I save a file to this 'hot'
directory, it will get auto-synced to the corresponding location in my
repository. Likewise, all sub-folders will be created as folder nodes,
etc., etc..  In the beginning, I plan to keep it simple because my
needs are simple; but the syncing strategy can obviously get
complicated very quickly.

Has someone already done or started a similar effort?
I am not sure if this might help, but it seems to me that the efforts of
Christian Jaeger with http://pflanze.mine.nu/ by using a virtual file system
should do the trick. His website is a bit outdated, but I think he is still
active and you might just want to contact him

HTH

Michi

Thanks for any info.
Robert


--
Michael Wechner
Wyona      -   Open Source Content Management   -    Apache Lenya
http://www.wyona.com                      http://lenya.apache.org
[EMAIL PROTECTED]                        [EMAIL PROTECTED]



Reply via email to