Adam Wulkiewicz wrote:
I'd rather choose option number one (thank you again) but I'll wait a while for the feedback from the comunity.
You should search the archives of the dev list because we have discussed spatial indexing several times in the context of a failed GSOC project from 2006 and the design of my library as well as Boost.Geometry. Both libraries were criticized during review for lacking spatial query data structures. Extention to n-dimensions was discussed at length several times. There is no doubt that there is interest in a library such as you propose. At this stage we would probably like to see your code and also have you describe the interfaces you have in mind along with example code for how they would be used. The majority of interest will be in the interfaces and how the data structures are used and not the internal details of how they are implemented. Regards, Luke