On Mon, Aug 6, 2012 at 10:45 AM, Niels Charlier <ni...@scitus.be> wrote: > The current patch and proposal does not include these features. However, I'd > understand that WFS post requests need to at least support the same features > as wfs get requests. Please let me know the minimum requirements with > regards to this to get the patch approved.
The GeoTools part is not affected, but in order to commit the GeoServer part you will need to make sure POST requests are supported as well. Only some time ago Justin complained that some WFS functionalities have been added on GET requests only and that moving forward we should try to implement everything on both sides. > Yes, you have a point here, this is a weak part. I stole the code from your > fastbbox, but it is indeed an ugly hack. I would be happy to change the > filter visitors instead, but I was trying to minimise changes to the core. > I am sorry I didn't point this out in the proposal, I will do something > about it. The hack makes sense in the FastBBOX case because it's just a faster alias to the BBOX filter, it's not a new type of filter, yours is a brand new one that implementors should be aware of. > The method checkCoordinateReferenceSystemDimension is identical to the one > in its superclass, ReferencedEnvelope. (In fact, I should remove this > copy-pasted method and instead make it protected instead of private in its > superclass, this is left-over from changing the class structure). > So the existing implementation already checks the srs's dimensionality > against this. This means that is now forbidden to apply a 2D bbox against a > 3D SRS. It only seemed logical for me to extend the same logic to situation > the other way around. > I would personally recommend to use a 3D SRS if you have 3D geometries, even > in postgis, this seems to me like the best way to work. If this is not > desired behaviour, I think it is not just my changes, but the existing code > of the ReferencedEnvelope class that must be reviewed. > Perhaps I could change the check to test if the dimension of SRS < than the > dimension of the bbox? Does that make any sense at all? Hmm... I don't have a good answer to this one off the top of my head. While I agree that in theory one should just use a 3d system, the practice is difference and here we are not doing research, but industrial systems that must work in the real world. Whatever path is chosen imho it should allow to write 3d filters against a PostGIS store that has a 2d crs but has 3 dimensions in the geometry columns table. SRS < than the dimensions of the bbox seems a bit hacky but indeed it might well be the only solution. > >> I find it suprising that 3D reprojection works indeed. Afaik to do it one >> needs proper support >> for vertical CRSs, which is contained in referencing3D. I though that >> module was not complete, >> have you looked into it and found otherwise? > > > As far as I know that module isn't functional. > I admit I am not 100% knowledgeable about everything in this field, I have > only been learning this SRS stuff the past few months, so there is still a > lot I don't know. > I am only needing to reproject point and linestring geometries with 3d > coordinates. Perhaps that explains it. Afaik this works the way I explained before, that is, the Z dimension is left unaltered during the reprojection. Which is not true reprojection, but what we can afford given that the vertical transformation module is not working Cheres Andrea -- == Our support, Your Success! Visit http://opensdi.geo-solutions.it for more information. == Ing. Andrea Aime @geowolf Technical Lead GeoSolutions S.A.S. Via Poggio alle Viti 1187 55054 Massarosa (LU) Italy phone: +39 0584 962313 fax: +39 0584 962313 mob: +39 339 8844549 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/ _______________________________________________ GeoTools-Devel mailing list GeoTools-Devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/geotools-devel
