Re: [geos-devel] GEOS 3.7.0 release in 2 weeks

2017-09-01 Thread Roger Bivand
If this is just a patch release, do not do 3.7.0. If it is a major release, please respect the downstream developers and do a proper release with alpha, beta and RCs, to give us a reasonable chance to check whether any other changes don't break our stuff. Just because you need new functionality doe

Re: [geos-devel] GEOS 3.7.0 release in 2 weeks

2017-09-01 Thread Regina Obe
Roger, I believe most of the changes in GEOS 3.7.0 are bug fixes too involved to backport to GEOS 3.6. There are some new features but not all that many. If I were to release an alpha, I think it would be setting a precedent since I don't think an alpha has ever been released for GEOS. At bes

Re: [geos-devel] [GEOS] #832: LinearLocation::isValid and LinearLocation::normalize, faulty comparisons

2017-09-01 Thread GEOS
#832: LinearLocation::isValid and LinearLocation::normalize, faulty comparisons +--- Reporter: nila| Owner: geos-devel@… Type: defect | Status: closed Priority: major | Milestone: Component: Default |

Re: [geos-devel] GEOS 3.7.0 release in 2 weeks

2017-09-01 Thread Regina Obe
Just confirmed with Sandro. Just two functions added and a lot of bug fixes that were also present in 3.6.0 and 3.5.1. He needs to make another change so I'm going to wait another week before releasing a beta or RC. Now for GEOS 3.8 I think we'll probably want an alpha since I think Mateusz wa

Re: [geos-devel] GEOS 3.7.0 release in 2 weeks

2017-09-01 Thread Mateusz Loskot
On 1 September 2017 at 17:31, Regina Obe wrote: > > Now for GEOS 3.8 I think we'll probably want an alpha since > I think Mateusz wants to do some major refactoring and up the required GCC to > 4.8. GEOS has accepted to require C++11 [1] as minimum C++ standard. It means, we are free to use C++1