You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The source code is now tight to Zenodo releases. However, the Zenodo release does not do a git clone --submodule.
Hence, due to the dependency on thirdparty submodules LabelLib cannot be built with the Zenodo release alone.
I know, this is not very elegant, however, to make it possible to build LabelLib using the Zenodo release alone and to address the issue that conda-forge wants releases to be build from tarballs as opposed to git repositories, we should consider coping the dependencies instead of using submodules.
This would solve issues with conda-forge and zenodo alike.
For the next release we could/should link labellib to Zenodo (see: link). This way, LabelLib gets a DOI and could be cited.
Since thirdparty/Eigen and thirdparty/pybind11 are only used as fallback, it should be sufficient to explicitly copy pcg-cpp, which we totally could do. And for conda-forge we could use automatic github tarballs, which are generated for any github release.
I just submitted labellib to conda-forge. There was in issue with the windows build. Hence, I skip the Windows build for now. see conda-forge/staged-recipes#15497
The source code is now tight to Zenodo releases. However, the Zenodo release does not do a git clone --submodule.
Hence, due to the dependency on thirdparty submodules LabelLib cannot be built with the Zenodo release alone.
I know, this is not very elegant, however, to make it possible to build LabelLib using the Zenodo release alone and to address the issue that conda-forge wants releases to be build from tarballs as opposed to git repositories, we should consider coping the dependencies instead of using submodules.
This would solve issues with conda-forge and zenodo alike.
For the next release we could/should link labellib to Zenodo (see: link). This way, LabelLib gets a DOI and could be cited.
Originally posted by @tpeulen in #11 (comment)
The text was updated successfully, but these errors were encountered: