Open inorton opened 6 years ago
Do you know if the original CUnit has been officially abandoned? It looks like there has not been a release since 2014, so perhaps that is the case. I needed CUnit to run the tests for NFFT (nfft-feedstock), but do not otherwise use it, so I don't have a strong opinion one way or the other.
conda-forge typically only provides builds of officially released versions as opposed to development builds from a particular branch. I don't know if there is any established criteria for making decisions on when a fork should take precedence over the original project. Perhaps someone from @conda-forge/core can comment on the issue? You might also try raising an issue over at staged-recipes if you don't get a response here.
@inorton - you may be infringing on the original cunit's trademark, and we currently can't build cunity/cunit and call it cunit. However, we'd be happy to take your package (under a disambiguated name) as a separate package. Simply submit a new PR to conda-forge/staged-recipes, as with all new packages.
Hi,
A few months ago I decided for fork cunit to fix some old compiler warnings and to finish off it's junit xml report format. I've also recently fixed windows vs2015+ support.
I maintain "cunity cunit" over on gitlab at https://gitlab.com/cunity/cunit which is CI built for windows and 32/64 linux and if I get time perhaps x64 solaris 11 soon.
I have a "classic-cunit" branch which is compatible with the last release of cunit from sourceforge and has only had minor code fixes. Or there is the master branch which will be "stable" with new features coming along soon.