Cross-posting my reply on GitHub:
To compute a full noding, use IntersectionAdder instead of
IntersectionFinder. IntersectionFinderAdder only adds
*proper*intersections. (This is stated in the Javadoc, but I agree
that its
implication is not obvious, and that the name of the class is not
particularly descriptive. However, if you look at the JTS source code,
you'll see that IntersectionAdder is used for general noding, while
IntersectionFinderAdder is only used for snap-rounding).
To make all this clearer, I am going to copy IntersectionFinderAdder to a
new class ProperIntersectionAdder, and deprecate IFA.
On Sun, Aug 11, 2013 at 3:04 PM, Ryan Clark <[email protected]> wrote:
> I actually ran into this issue in working with JSTS (
> https://github.com/bjornharrtell/jsts), but it was suggested that I ask
> about it here.
>
> Basically, it appears that if I start with two LineStrings that share a
> common vertex, the noding algorithm (using MCIndexNoder) does not split the
> lines into segments at the shared vertex. I saw similar behavior where a
> line "touches" the other line but does not cross it.
>
> I described the issue here with pictures:
> https://github.com/bjornharrtell/jsts/issues/137
> I wrote an example script (again, in JavaScript) here:
> https://gist.github.com/rclark/6205437
>
> I'm checking to see if this is expected behavior, if it has to do with
> JSTS specifically, or if there's another way I should approach the problem?
> The "touches" situation seems like it must be some kind of precision issue,
> but I don't understand the situation where the two initial lines cross at a
> shared vertex.
>
> If you want to try it out, here is a
> map<http://rclark.github.io/make-polys/> that
> lets you draw lines, select a set of them, and then run them through the
> noding algorithm before running a polygonizer operation. You can easily
> demonstrate the "touches" situation there -- lines that "touch" but don't
> cross will not define polygons, because the nodigin algorithm will not
> produce the appropriate SegmentStrings.
>
> Thank you!
> ____________________
>
> Ryan Clark
> Arizona Geological Survey
> [email protected]
> (520) 302-4871
> facebook.com/ModernGeologist <https://www.facebook.com/ModernGeologist>
> @worbly <https://twitter.com/worbly>
>
>
>
>
>
>
>
>
>
>
>
> ------------------------------------------------------------------------------
> Get 100% visibility into Java/.NET code with AppDynamics Lite!
> It's a free troubleshooting tool designed for production.
> Get down to code-level detail for bottlenecks, with <2% overhead.
> Download for free and get started troubleshooting in minutes.
> http://pubads.g.doubleclick.net/gampad/clk?id=48897031&iu=/4140/ostg.clktrk
> _______________________________________________
> Jts-topo-suite-user mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/jts-topo-suite-user
>
>
------------------------------------------------------------------------------
Get 100% visibility into Java/.NET code with AppDynamics Lite!
It's a free troubleshooting tool designed for production.
Get down to code-level detail for bottlenecks, with <2% overhead.
Download for free and get started troubleshooting in minutes.
http://pubads.g.doubleclick.net/gampad/clk?id=48897031&iu=/4140/ostg.clktrk
_______________________________________________
Jts-topo-suite-user mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/jts-topo-suite-user
