Re: [geos-devel] Ownership of userData

2019-01-29 Thread Martin Davis
FWIW I sympathize with the problem. This actually felt like a bit of a hack when I implemented it, but since it's not much of an issue in Java-land I ploughed ahead. Isn't there *always* going to be an ownship problem for anything returned? The user knows they own any returned geometry, which is

[geos-devel] Ownership of userData

2019-01-29 Thread Paul Ramsey
I am going through the various tickets and PRs and trying to clean them up. This one seemed straightforward at first, but the issue of ownership of userData is pretty ugly in GEOS land. https://github.com/libgeos/geos/pull/115#issuecomment-458649362 Basically we seem to currently assume that whom