Yes, that's the same data pattern and the same problem.
Sandro, as you say it's fairly easy to optimize *intersects* to avoid the
problem (and the JTS PreparedGeometry does this). It's quite a bit harder
to optimize *intersection*.
I have been working on a different approach to overlay which avoids some of
the issues in the current JTS algorithm. In particular, unlike the current
the algorithm the new approach does not have to build the entire
intersection graph in memory before computing the result. Out of interest
I ran the test case using this code. It completed in about 160 s, with
reasonable memory use. The result contained about 2M nodes, compared to the
12K in the input. So this is essentially an O(n^2) explosion in data
volume, which is why the problem fails under the current JTS algorithm.
On Tue, Jan 20, 2015 at 12:32 PM, Sandro Santilli <[email protected]> wrote:
> On Tue, Jan 20, 2015 at 09:38:51AM -0800, Martin Davis wrote:
> > Wow, that's one crazy LineString! It has an area where it has a huge
> > number of self-intersections (see attached image). This causes a huge
> > number of intersection nodes to be created, which is flooding memory.
>
> Reminds me of a similar shape reported here:
> http://trac.osgeo.org/geos/ticket/708
>
> > PreparedGeometry doesn't (yet) support overlay methods, so not sure what
> > you mean when you say it "works".
>
> I guess we're talking about the RelateComputer here, see also the ticket.
>
> --strk;
>
> () Free GIS & Flash consultant/developer
> /\ http://strk.keybit.net/services.html
>
> --
>
> what comes below this line is just spam, dont bother scrolling...
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
> still here ?
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
------------------------------------------------------------------------------
New Year. New Location. New Benefits. New Data Center in Ashburn, VA.
GigeNET is offering a free month of service with a new server in Ashburn.
Choose from 2 high performing configs, both with 100TB of bandwidth.
Higher redundancy.Lower latency.Increased capacity.Completely compliant.
http://p.sf.net/sfu/gigenet
_______________________________________________
Jts-topo-suite-user mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/jts-topo-suite-user
