Great, Thanks! I am looking forward for more improvements in the filter chain. WanMil
Hi WanMil,
thanks for the propsals. I've changed the code and committed it. I kept the extra lines for the spike detection but moved them to the beginning. You are right that the area test would detect that as well, but this test is much simpler.
Gerd P.S. The change in the order of points was indeed not intended.
WanMil wrote
Hi Gerd,
filter_shapes_v1.patch: I did not test it but I made a little code review. That looks good. Anyhow I have some comments:
You want to filter shapes only? (The comments also talk about filtering lines). When working with shapes only I think in the end of the doFilter method it should better be if (newPoints.size() > 3) ... instead of newPoints.size() > 1.
Maybe it's also better for quick understanding to cast to MapShape only instead of working with a MapLine line object.
I stumbled a bit on the following part: int last = newPoints.size()-1; areaTest[0] = newPoints.get(last); areaTest[1] = newPoints.get(last-1); areaTest[2] = newP; It is correct but for a better reading I would leave the points in the original order: int last = newPoints.size()-1; areaTest[0] = newPoints.get(last-1); areaTest[1] = newPoints.get(last); areaTest[2] = newP;
I wonder if the following part is required: if (newPoints.get(newPoints.size()-2).equals(newP)){ // detected simple spike newPoints.remove(newPoints.size()-1); continue; } I think the simple spike is also covered with the area==0 test?
Maybe you could introduce a boolean flag if anything was changed by the filter and create a new object only if that's true. Otherwise each shape is recreated.
WanMil
Hi,
I got no feedback on these posts regarding patches:
http://gis.19327.n5.nabble.com/Why-do-we-have-the-Area-too-small-to-split-at...
http://gis.19327.n5.nabble.com/PATCH-v1-Optimize-filters-for-shapes-tp574414...
I used both of them and I think they are ok. Is it okay to commit them?
Gerd
-- View this message in context: http://gis.19327.n5.nabble.com/uncommitted-patches-for-mkgmap-tp5749600.html Sent from the Mkgmap Development mailing list archive at Nabble.com. _______________________________________________ mkgmap-dev mailing list
mkgmap-dev@.org
_______________________________________________ mkgmap-dev mailing list
mkgmap-dev@.org
-- View this message in context: http://gis.19327.n5.nabble.com/uncommitted-patches-for-mkgmap-tp5749600p5749... Sent from the Mkgmap Development mailing list archive at Nabble.com. _______________________________________________ mkgmap-dev mailing list mkgmap-dev@lists.mkgmap.org.uk http://lists.mkgmap.org.uk/mailman/listinfo/mkgmap-dev