On 10/4/2017 12:14 AM, Mateusz Loskot wrote:
As I explained on IRC last night, IMHO, dropping GEOS by osm2pgsql and switch to libosmium turned to be good. The GEOS API breakage just catalised the switch, which IMO was inevitable anyway.
osm2pgsql dropping GEOS is unrelated to its API switch. The idea has been around since before GEOS 3.5 and development work started before we were aware of GEOS 3.6 issues.
We looked at switching to the C API but that wasn't considered a worthwhile use of time. We also didn't have any patches which would support GEOS 3.6 without breaking support for earlier versions. There were also questions on if we could fit those changes into a patch release.
_______________________________________________ geos-devel mailing list geos-devel@lists.osgeo.org https://lists.osgeo.org/mailman/listinfo/geos-devel
