Re: [Geoserver-devel] Property changes of Geoserver's nested XXInfo objects aren't notified to the

SourceForge Headquarters 225 Broadway Suite 1600 San Diego, CA 92101 +1 (858) 454-5900

Hi guys,
sorry for the late reply.
Thanks for your feedbacks... I have already taken a quick look to them and
I'll check them again as soon as possible.

Cheers,
Daniele

On Thu, Jul 19, 2012 at 7:00 PM, Gabriel Roldan <gr...@op...> wrote:

> Since JAIInfo is a plain data type (i.e. GeoServerInfo _owns_ it), I
> think it's just easier to use a clone for the JAIInfo wicket model and
> then set it on save. See attached patch on my previous reply?
>
> On Thu, Jul 19, 2012 at 1:53 PM, Justin Deoliveira <jde...@op...>
> wrote:
> > Hi Daniele,
> >
> > Yeah, the way we deal with nested objects like this is kind of weird. So
> I
> > think there are a couple of things things going on here.
> >
> > The first is that JAIInfo and CoverageAcecssInfo don't extend from the
> Info
> > interface, so modification proxy doesn't treat them like any differently
> > than a "regular" object ,in that they generally don't get picked up as
> dirty
> > unless they are set with a setter. So one way to fix this problem would
> be
> > to have them extend from the Info interface.
> >
> > WIth that I believe that modification proxy will actually treat them as
> > "dirty" when you do something like this:
> >
> > GeoServerInfo global = geoServer.getGlobal();
> > JAIInfo jai = global.getJAI();
> > jai.setSomething(...);
> > geoServer.save(global);
> >
> > SO what you will get is a property changed event which reports that the
> > "JAI" property has changed. The trouble is that as currently implemented
> the
> > new value and the old value will be the same object. And of course it
> > doesn't give you any notion of what properties in the nested object has
> > changed.
> >
> > I am not so sure how to fix this. Obviously the change of hanging the
> > interfaces extend Info is a trivial one but how to report the changes to
> the
> > nested objects isn't. One thing we could do is instead of just reporting
> > that the "jai" object changes is use a nested notation to report what
> > properties actually changed, and pass them into the listener method.  SO
> > instead of af an event that looks like this:
> >
> >   postModifed(GeoServerINfoImp@..., ["jai"], [JAIInfoImpl@...],
> > [JAIInfoImpl@...])
> >
> > You would get one like:
> >
> >   postModifed(GeoServerINfoImp@..., ["jai.something"], ["oldSomething"],
> > ["newSomething"])
> >
> > If that makes sense? This would require some changes to modification
> proxy.
> >
> > An alternative would be to ensure we throw multiple events in this case.
> Not
> > sure which one is better.
> >
> > -Justin
> >
> > On Thu, Jul 19, 2012 at 2:33 AM, Daniele Romagnoli
> > <dan...@ge...> wrote:
> >>
> >> Hi all,
> >> yesterday I was debugging geoserver to figure out why any global change
> is
> >> triggering a JAI initialization and a coverageAccess initialization.
> >> Long story short the handleGlobalChange and the handlePostGlobalChange
> of
> >> the custom ConfigurationListenerAdapter simply invoke the initialization
> >> method without doing any check of the changed properties/values.
> >> Beside of that, I have noticed that the propertyNames, oldValues and
> >> newValues lists are empty (even when changing some params from the JAI
> page
> >> or the CoverageAccess page). The only changing thing seems the
> >> "updateSequence".
> >> After a brief meeting with Andrea, the outcome of the discussion is that
> >> the properties changes won't be reported to the listener in case the
> >> properties belong to an object which is nested in the main object (as
> >> CoverageAccessInfo and JAIInfo are nested into GeoServerInfo).
> >> Having these properties changes being notified would allow to check them
> >> and potentially avoid some initializations (JAI TileCache settings,
> >> TileScheduler settings, Tile Recycling , CoverageAccess
> threadPoolExecutor
> >> settings) when they aren't needed.
> >>
> >> How can we deal with that?
> >>
> >> Best Regards,
> >> Daniele
> >> --
> >> ==
> >> Our support, Your Success! Visit http://opensdi.geo-solutions.it for
> more
> >> information.
> >> ==
> >>
> >> Ing. Daniele Romagnoli
> >> Senior Software Engineer
> >>
> >> GeoSolutions S.A.S.
> >> Via Poggio alle Viti 1187
> >> 55054  Massarosa (LU)
> >> Italy
> >> phone: +39 0584 962313
> >> fax:   +39 0584 962313
> >>
> >> http://www.geo-solutions.it
> >> http://twitter.com/geosolutions_it
> >>
> >> -------------------------------------------------------
> >>
> >>
> >>
> ------------------------------------------------------------------------------
> >> Live Security Virtual Conference
> >> Exclusive live event will cover all the ways today's security and
> >> threat landscape has changed and how IT managers can respond.
> Discussions
> >> will include endpoint security, mobile security and the latest in
> malware
> >> threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
> >> _______________________________________________
> >> Geoserver-devel mailing list
> >> Geo...@li...
> >> https://lists.sourceforge.net/lists/listinfo/geoserver-devel
> >>
> >
> >
> >
> > --
> > Justin Deoliveira
> > OpenGeo - http://opengeo.org
> > Enterprise support for open source geospatial.
> >
> >
> >
> ------------------------------------------------------------------------------
> > Live Security Virtual Conference
> > Exclusive live event will cover all the ways today's security and
> > threat landscape has changed and how IT managers can respond. Discussions
> > will include endpoint security, mobile security and the latest in malware
> > threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
> > _______________________________________________
> > Geoserver-devel mailing list
> > Geo...@li...
> > https://lists.sourceforge.net/lists/listinfo/geoserver-devel
> >
>
>
>
> --
> Gabriel Roldan
> OpenGeo - http://opengeo.org
> Expert service straight from the developers.
>

-- 
==
Our support, Your Success! Visit http://opensdi.geo-solutions.it for more
information.
==

Ing. Daniele Romagnoli
Senior Software Engineer

GeoSolutions S.A.S.
Via Poggio alle Viti 1187
55054  Massarosa (LU)
Italy
phone: +39 0584 962313
fax:   +39 0584 962313

http://www.geo-solutions.it
http://twitter.com/geosolutions_it

-------------------------------------------------------

Re: [Geoserver-devel] Property changes of Geoserver's nested XXInfo objects aren't notified to the

Re: [Geoserver-devel] Property changes of Geoserver's nested XXInfo objects aren't notified to the globalChanges event listeners registered to the related initializers (CoverageAccessInitializer, JAIInitializer, ...)