Tim, I do agree with all your suggestions. Actually I can just go on holidays and wait you to do the job for us, I'm sure it will be better that whatever I can do alone. ;-)
I'll go on holidays tomorrow (15 days) but I'm available for the debate today. Let me know if you want to chat about it. #openlayers ? I think we can already debate about a timetable and see if and where I can help you to integrate your work. Let's try to see who else want to join (seems Erik has already something in his hands). I suggest to use your Wiki page to manage this: http://trac.openlayers.org/wiki/Future/MapContext see you later Lorenzo Tim Schaub wrote: > A few ideas: > It would be good to start working towards a JavaScript structure for the > map context. For OL, this structure would contain all relevant layer > information and the current map center/resolution. On the WMC side, we > should discuss how to standardize the extensions to WMC to account for > OL specific stuff. > > This structure should be serialized/deserialized with a JSON parser. > http://dev.openlayers.org/sandbox/tschaub/geojson/lib/OpenLayers/Format/JSON.js > (for example - and you can do format.write(obj, pretty) too) > > In addition to translating this structure into a WMC doc, we should be > able to translate it into a short string - though this will be lossy, it > can be used to replace the permalink logic. > > All this could be handled by a control like OpenLayers.Control.Context > or OpenLayers.Control.ApplicationState. This control would use the JSON > format and a (yet to be built) WMC format. > > I'll be working on this sort of stuff over the next couple months. I > look forward to collaborating. > > Tim > > >> Click on the bottom-right button "WMC Exporter" to see a pop-up with the >> JSON code of the WMC. >> Naturally you can alter the map with WMSManager and the WMC "should" be >> updated. >> >> as you can see: >> - the structure is not a perfect WMC. I've took a look from the 1.1 >> specs but there are things to be cleaned up. >> - some values are empty or null. Still have to decide if it make sense >> to use all of them. We can decide together which optional elements not >> to support. >> - I've added a "visibility" attribute to Layer object because WMC is not >> providing it. Some hint how to avoid this? I'm sure you understand how >> this is important for OL to define base or overlay layers. >> >> as you can't see: >> - not all values are dinamically set. Some are still hardcoded. >> >> >> there's another problem about attributes of Layers with multiple WMS >> layers inside. Which title, abstract to use? does it make sense to >> create a comma (or something else) separated list of values? >> >> >> I'm still not sure if the MapContext should be part of the OL.Map object >> or an external Tool. >> I'll go on with the external tool and hope to create soon a WMCParser to >> import WMCs. >> >> >> My hope was to offer something better before leaving for holidays but >> many meeting/debate/ecc at work delayed coding time. >> I'll leave in 2 days and be back for the third week of August. In the >> meantime I'll be available for debate about this solution. >> >> ciao >> Lorenzo >> >> _______________________________________________ >> Dev mailing list >> [email protected] >> http://openlayers.org/mailman/listinfo/dev >> >> !DSPAM:4033,46adec84154021431913854! >> >> > > _______________________________________________ > Dev mailing list > [email protected] > http://openlayers.org/mailman/listinfo/dev > _______________________________________________ Dev mailing list [email protected] http://openlayers.org/mailman/listinfo/dev
