[geos-devel] Gsoc 2013

vishal tiwari hi.vishal123 at gmail.com
Sun Mar 31 00:35:30 PDT 2013


Hey,
I found the code, it wasn't there with the initial search result on git. :P

Now i understand what you mean by what was missing. I can see a bunch of
files that are not in triangulation, so i suppose the project is to write
these missing files, which correspond to the Voronoi diagram generation.
Tell me if i am going wrong.

Thanks,
Vishal


On 31 March 2013 12:21, vishal tiwari <hi.vishal123 at gmail.com> wrote:

> Hey,
> i was looking at port of Delaunay triangulation. and came to know some of
> these stuff:
> i looked at the method signatures from the javadoc for JTS, and saw these
> signatures in the headers files 'libgeos/include/geos/triangulate'. And
> their implementation in 'libgeos/src/triangulate'. And also noticed some of
> the methods in QuadEdgeSubdivision.h were missing. I think there were
> mainly related to voronoi cell.
>
> I was searching for JTS's source code on git hub, and didn't find the code
> for Delaunay triangulation. in 'com.vividsolutions.jts' directory i
> couldn't find the triangulate directory. Where can i find that?
> Can you provide me a link where i can find JTS triangulation source code?
>
> And yes i understand the memory management problems in c/c++ that can
> happen.
>
> Thanks,
> Vishal
>
>
> On 29 March 2013 18:12, Sandro Santilli <strk at keybit.net> wrote:
>
>> On Fri, Mar 29, 2013 at 04:09:17PM +0530, vishal tiwari wrote:
>> > hey Sandro, thanks for replying. I want to know details of the project.
>> The
>> > ideas page of GEOS talks very less about it. I want to know what exactly
>> > needs to be done in the project that is proposed.
>>
>> Figuring it out is part of the project :)
>>
>> > From the statement what i have understood is, that we have to port the
>> > classes of Voronoi functionality from JTS to GEOS. JTS are java API,
>> and we
>> > need to port it to c++ classes. I want some guidance on how java classes
>> > can be ported to C++.
>>
>> The last port has been for the Delaunay triangulation, you could start to
>> see how that was done. There's a git mirror of the codebase on github if
>> you're more comfortable with that (offline history helps).
>>
>> The main problem you'll have to deal with is memory management:
>> the Java version relies on automatic garbage collection while in GEOS we
>> have to do that manually (no refcounting, no reachable graphs).
>>
>> It's usefu to start by looking at what's missing.
>> The GEOS codebase organization reflects the JTS one pretty closely.
>>
>> --strk;
>> _______________________________________________
>> geos-devel mailing list
>> geos-devel at lists.osgeo.org
>> http://lists.osgeo.org/mailman/listinfo/geos-devel
>>
>
>
>
> --
> Vishal Tiwari
> Undergraduate
> Computer Science and Engineering
> Lab for Spatial Informatics
> IIIT-Hyderabad
>



-- 
Vishal Tiwari
Undergraduate
Computer Science and Engineering
Lab for Spatial Informatics
IIIT-Hyderabad
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/geos-devel/attachments/20130331/b21ac7c1/attachment.html>


More information about the geos-devel mailing list