<div dir="auto"><div><div class="gmail_extra"><div class="gmail_quote"><blockquote class="quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Something I forgot to add in my original reply - instead of investing<br></blockquote><blockquote class="quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
time into utilising CGAL instead of GEOS, I'd rather see us<br>
investigate swapping out GEOS calls to Boost::geometry calls. The<br>
boost geometry algorithms are really nicely implemented, and their<br>
interface avoids the need to convert between different object types<br>
(ie QgsGeometry->GEOS). This could potentially be a big<br>
performance/memory usage win.</blockquote></div></div></div><div dir="auto"><div class="gmail_extra"><div class="gmail_quote"><blockquote class="quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><font color="#888888"></font></blockquote></div></div></div><div dir="auto"><br></div><div dir="auto">I think that keeping GEOS for all the Simple Features stuff (geoms, predicates and ops) is important to remain aligned and keep comparability with most of the other GFOSS sw based on GEOS/JTS.</div><div dir="auto"><br></div><div dir="auto">The only relevant exception is GRASS GIS, but it's different in so many ways that it is barely comparable to anyone else...</div><div dir="auto"><br></div><div dir="auto">Giovanni</div></div>