Timeline for Fixing non-noded intersection problem using PostGIS
Current License: CC BY-SA 3.0
4 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jan 17, 2019 at 21:26 | comment | added | jpmc26 | You mention snapping the geometry to the 1e-6 grid, but I'm sitting here wondering if snapping to a power of 2 would be better. PostGIS (and GEOS) using floating point numbers, so snapping to a power of 10 might not actually truncate the coordinates very much since the number may not have a finite length binary representation. But if you snap to say 2^-16, I believe that would be guaranteed to truncate any fractional part to just 2 bytes. Or am I thinking wrong? | |
| Mar 21, 2016 at 13:48 | history | edited | dbaston | CC BY-SA 3.0 |
fix typo
|
| Dec 2, 2014 at 15:20 | comment | added | John Powell | That's a clever idea. I have often found that intersection issues come from linestrings appearing during combinations of unions, differences, buffers, etc, which can be fixed by buffering everything, or dumping everything and only selecting Polygons/Mutlipolygons. This is an interesting approach. | |
| Mar 19, 2014 at 23:04 | history | answered | dbaston | CC BY-SA 3.0 |