<html>
<body>
At 12:46 PM 12/22/2008, Ragi Y. Burhum wrote:<br>
<blockquote type=cite class=cite cite="">
<dl>
<dd>Date: Mon, 22 Dec 2008 09:37:15 -0500<br>
<dd>From: Kevin Webb
<<a href="mailto:kfw4@cornell.edu">kfw4@cornell.edu</a>><br>
<dd>Subject: [gdal-dev] OGR and GEOS<br>
<dd>To:
<a href="mailto:gdal-dev@lists.osgeo.org">gdal-dev@lists.osgeo.org</a><br>
<dd>Message-ID:<br>
<dd>
<<a href="mailto:6.2.1.2.2.20081222085228.03fe8f40@postoffice7.mail.cornell.edu">
6.2.1.2.2.20081222085228.03fe8f40@postoffice7.mail.cornell.edu</a>><br>
<dd>Content-Type: text/plain; charset="us-ascii";
format=flowed<br><br>
<dd>Greetings!<br><br>
<dd>The documentation for the OGRGeometry class mentions the GEOS library
in a<br>
<dd>few of the<br>
<dd>method definitions, but I wonder if there is a "stronger"
OGR dependency on<br>
<dd>the GEOS<br>
<dd>library than the documentation suggests. Perhaps GEOS should be a
required<br>
<dd>library.<br><br>
<dd>I had done some processing using OGRGeometry::Intersects() without
the GEOS<br>
<dd>library<br>
<dd>installed, I then installed GEOS to run the same tests using<br>
<dd>OGRGeometry::Distance().<br>
<dd>After installing GEOS, but before running the
OGRGeometry::Distance()<br>
<dd>tests, I re-ran the<br>
<dd>code using OGRGeometry::Intersects() and noticed different
intersection<br>
<dd>results.<br><br>
<dd>Before-GEOS, the Intersects() method indicated intersection with 2
layer<br>
<dd>polygons. After-GEOS,<br>
<dd>the Intersects() method indicated intersection with 1 layer polygon.
Note:<br>
<dd>After-GEOS results<br>
<dd>for both Intersects() and Distance() == 0 were the same.<br><br>
<dd>Should the GEOS library be a requirement for deterministic OGR
processing?<br><br>
<br>
<dd>Thank you.<br><br>
<br>
<dd>KFW<br><br>
</dl>With GEOS and without GEOS you will get "deterministic"
results. The difference is that the algorithm to calculate intersections
without GEOS is simpler - just bounding box intersections. You want GEOS
installed.<br>
<br>
- Ragi</blockquote><br>
Thank you for your reply Ragi.<br><br>
You are correct, I should not have used to word
"deterministic;" with-GEOS or without-GEOS I produced the same
output<br>
result for each processing run. <br><br>
As Frank suggested in his post, I think more clarity is needed in the
write-ups for the methods that use alternate<br>
processing logic with-GEOS installed when compared to a similar operation
without-GEOS installed.<br>
<br><br>
<br>
KFW<br><br>
<br>
<blockquote type=cite class=cite cite="">
_______________________________________________<br>
gdal-dev mailing list<br>
gdal-dev@lists.osgeo.org<br>
<a href="http://lists.osgeo.org/mailman/listinfo/gdal-dev" eudora="autourl">
http://lists.osgeo.org/mailman/listinfo/gdal-dev</a> </blockquote>
<x-sigsep><p></x-sigsep>
-----------------------------------------<br>
Kevin Webb<br>
Programmer/Analyst<br>
Cornell Lab of Ornithology<br>
159 Sapsucker Woods Rd.<br>
Ithaca, NY 14850<br>
Tel: 607.254.2103<br>
Fax: 607.254.2415<br>
kfw4@cornell.edu<br>
-------------------------------------------<br>
</body>
</html>