jackrabbit-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Julio Castillo" <jcasti...@edgenuity.com>
Subject RE: JCR Webdav Server -1.4.1 -looking for a client
Date Fri, 21 Mar 2008 23:10:33 GMT
Thank you Angela,
I have a couple of questions regarding your last statement:

"if you have jcr2spi configured to use the spi2dav SPI implementation, you
must make sure you have the corresponding webdav server running. spi2dav is
still part of the client side.

steps to go there:
- deploy the JCRWebdavServerServlet (part of the jackrabbit webapp)
- make sure your jcr2spi is started with spi2dav
- configuration properties of spi2dav are (see
repositoryStubImpl.properties):
   > location of the server
   > default workspace name"

1) When you say, "make sure your jcr2spi is started with spi2dav", where
exactly is this configured? Are you talking on the servlet side? Or are you
talking of whatever package I put together for the client side?

2) Related to the above question. I did find several references to the
repositoryStubImpl.properties file in the main trunk and sandbox directory.
Somehow I have to make sure that is read by the client package that I put
together?

Thanks for your feedback

Julio Castillo 

-----Original Message-----
From: Angela Schreiber [mailto:anchela@day.com] 
Sent: Wednesday, March 19, 2008 1:13 AM
To: users@jackrabbit.apache.org
Subject: Re: JCR Webdav Server -1.4.1 -looking for a client

hi julio

> What does it actually mean to use this servlet 
> (JCRWebdavServerServlet) to "remote JCR170 calls"?

it means, that you have a JCR client that communicates with the persistent
layer (server) using the webdav protocol.

> Does it mean that the WebDav interface was "extended" to support 
> specific JCR calls?

the JCR API maps pretty well (and surprisingly natural) to WebDAV. there
were only a couple of areas that were are not covered by any of the WebDAV
RFCs. This includes:

- observation
   -> extension
- clone
   -> unresolved
- series of transient modifications (everything between
   2 Session.save() calls).
   -> custom Lock type/scope.

the mapping is described in
http://jackrabbit.apache.org/JCR_Webdav_Protocol.doc

> Per the online documentation a corresponding JCR client refers to the 
> "JCR to SPI" and "SPI to WebDAV" in the sandbox. I found the 
> following: a package
> called: jackrabbit-jcr2spi/* on the regular release and spi2dav/* on 
> the sandbox.

that's it. jcr2spi represents the JCR API and it requires an SPI
implementation. spi2dav is one example.

> Any preference on either approach? 
> How do I bootstrap a client?

which approach? you lost me here... sorry.

> On a brief look at the spi2dav files I believe this is the path to follow:
> Instantiate a copy of JCR2SPIRepositoryStub which in turn will 
> instantiate a copy of RepositoryServiceImpl.
> The starting point for my code would then start with the 
> JCR2SPIRepositoryStub.

> I couldn't not find a starting point however with just the code in the 
> main trunk within the jackrabbit-jcr2spi/*. Any ideas?

if you have jcr2spi configured to use the spi2dav SPI implementation, you
must make sure you have the corresponding webdav server running. spi2dav is
still part of the client side.

steps to go there:
- deploy the JCRWebdavServerServlet (by default its
   part of the jackrabbit webapp).
- make sure your jcr2spi is started with spi2dav
- configuration properties of spi2dav are (see
   repositoryStubImpl.properties):
   > location of the server
   > default workspace name

hope that helps
regards
angela

> Julio Castillo


Mime
View raw message