Hi Martin and all,

I've now had the chance to do some test runs of the "direction
constrained nearest neighbour" routine on a multi-core machine and it
seems to be working very well. The concurrent STRtree queries were
fine.

I spent a bit of time tweaking the code involved in working out which
geometries a given sector can "see" and then calculating distances to
them. The code to test if a geometry is seen follows the logic that
you suggested last month: run through segments testing if an end-point
can be seen or, if not, testing for intersection with a sector ray.
For each geometry that is visible, the routine then collects the
visible segments and calculates a distance using the relevant
CGAlgorithms method. I belatedly realized that a bit of caching to
avoid doubling up between the "can see" and "get distance" tasks makes
a big difference to the running time.

Let me know if you still want to do a blog post on the algorithm
Martin - I'd be happy to contribute text and/or pics.

Michael

------------------------------------------------------------------------------
CenturyLink Cloud: The Leader in Enterprise Cloud Services.
Learn Why More Businesses Are Choosing CenturyLink Cloud For
Critical Workloads, Development Environments & Everything In Between.
Get a Quote or Start a Free Trial Today. 
http://pubads.g.doubleclick.net/gampad/clk?id=119420431&iu=/4140/ostg.clktrk
_______________________________________________
Jts-topo-suite-user mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/jts-topo-suite-user

Reply via email to