Matthew,

The short answer to this is that I had to patch gsconfig.py to support
cascaded layers. You may be able to get away with just using that and
using updatelayers.

https://github.com/jj0hns0n/gsconfig.py/tree/wms_store

Give that a try and let me know if you run into any problems or if it works.

Jeff

On Fri, Mar 16, 2012 at 1:38 PM, Matthew Hanson
<[email protected]> wrote:
> Jeff,
>
> Are you able to use updatelayers to update GeoNode with any WMS layers
> that are in GeoServer?
>
> Because for WMS layers that I *am* able to add to GeoServer
> (forgetting the above issues), updatelayers does not work.   I've
> traced the problem into gsconfig.py where it gets the list of stores.
>  I've got two stores: one PostGIS, one WMS (cascaded).   It returns
> only the PostGIS in the list of stores and not the wms layer.
>
> I sort of lost track of things when I saw that GeoServer was getting
> the list of stores from a datastores.xml file, but which didn't
> actually exist....I don't know enough about GeoServer to know what's
> going on, but it must create that on the fly from all the
> datastore.xml files that exist, in each of the store directories.
> However WMS stores do not have a datastore.xml file, instead they have
> a wmsstore.xml file.
>
> So it looks like this is a problem with gsconfig.py not retrieving all
> the stores (only dataStores and coverageStores...not sure of the
> difference there), and that the WMS stores are left out.
>
> matt
>
>
>
> On Fri, Mar 16, 2012 at 10:28 AM, Matthew Hanson
> <[email protected]> wrote:
>> Well since I'm already using the existing version, that should be
>> 2.1.3, so updating it won't help me with the cascading problem.
>>
>> On Thu, Mar 15, 2012 at 4:34 PM, David Winslow <[email protected]> wrote:
>>> On Thu, Mar 15, 2012 at 4:13 PM, Jeffrey Johnson <[email protected]>
>>> wrote:
>>>>
>>>> David,
>>>>
>>>> Can you describe these "other modifications". I've never run into
>>>> trouble doing it this way, and just saying its not a good idea to do
>>>> it this way is just not a good enough reason for me not to continue
>>>> doing it.
>>>
>>>
>>> I don't think the users' mailing list is the appropriate place for that.
>>>  Feel free to diff the sources of geoserver's web/app module with GeoNode's
>>> geoserver-geonode-ext.
>>>
>>>>
>>>> On Thu, Mar 15, 2012 at 1:07 PM, David Winslow <[email protected]>
>>>> wrote:
>>>> > I don't recommend just dropping the geonode-geoserver-ext JAR into a
>>>> > standard GeoServer installation.  The WAR file that we distribute has
>>>> > some
>>>> > modifications beyond just adding the geoserver-geonode-ext JAR.  WIthout
>>>> > them, it is non-deterministic whether or not the GeoNode integration
>>>> > with
>>>> > the GeoServer security system actually takes effect.  (Actually, it used
>>>> > to
>>>> > be non-deterministic but I think now it just doesn't work.)
>>>>
>>>> I did this as recently as monday, and it certainly did work, and in
>>>> fact, I carefully tested that the GeoNode based security most
>>>> certainly was working.
>>>>
>>>> > The GeoNode WAR is built from GeoServer nightly jars and has all
>>>> > bugfixes
>>>> > that have gone into GeoServer 2.1.3.  I also put together a branch to
>>>> > bring
>>>> > GeoNode up to the GeoServer 2.2
>>>> > series: https://github.com/dwins/geonode/tree/geoserver-2.2  Please try
>>>> > one
>>>> > of those instead of trying to "geonode-ize" a regular GeoServer.
>>>>
>>>> Is there a reason that this geoserver-2.2 branch hasnt been submitted
>>>> as a pull request and pulled onto master yet? Seems like it makes
>>>> sense for us to get GeoNode up to speed with current GeoServer
>>>> development sooner rather than later.
>>>
>>>
>>> When a stable release of GeoServer 2.2 is published by the GeoServer team, I
>>> will verify that this branch still compiles.  If so, we'll update to
>>> GeoServer 2.2 at that time.  For now though, GeoServer 2.2 is still
>>> considered unstable by the GeoServer developers and I don't think it's
>>> appropriate for inclusion as the default in GeoNode.
>>>
>>> --
>>> David Winslow
>>> OpenGeo - http://opengeo.org/

Reply via email to