> The interface was just an idea I threw in there, because there were concerns
> about ReferencedEnvelope3D not being a ReferencedEnvelope.
>
> But the idea of not making separate classes and make the current ones support
> 3D, I think is quite tricky too.
> The algorithms aren't necessarily the same with 2d and 3d. Basically, every
> method would need to have a check whether there is a 3rd dimension or not.
Agreed; it would not be fun. We do have a couple implementations of the opengis
Envelope interface to show us the way however.
Still I would rather have one class that is tricky for us to write; then make
something that is difficult for everyone else to use.
Notes:
- this single class solution prevents you from performing an instance of check
to determine if 2D or 3D is supported - will that be a problem for your code?
- A clean way to implement would be to have an optional internal "delegate"
opengis Envelope, if it is non null we can delegate to that implementation. If
it is null we can delegate to the JTSEnvelope superclass.
Jody
------------------------------------------------------------------------------
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
