I am unsure what's different here vs. staged recipes. We need to solve this issue to get a successful upload of the win-64 package (osx and linux are available now: https://anaconda.org/conda-forge/libignition-rendering4)
The check_ tests are just a Python-based check that the regular tests produced the expected XML result files. Probably we just need to add Python or some tool to the build dependencies, let me check.
For some reason half the tests currently fail on Windows (all tests that start with
check_...
while the corresponding tests withoutcheck_
seem to be fine. (https://dev.azure.com/conda-forge/feedstock-builds/_build/results?buildId=268364&view=logs&jobId=a70f640f-cc53-5cd3-6cdc-236a1aa90802&j=a70f640f-cc53-5cd3-6cdc-236a1aa90802&t=f5d15007-a01c-5ad8-c9ce-4d519d3b275f)Any ideas @JShep1 ?
I am unsure what's different here vs. staged recipes. We need to solve this issue to get a successful upload of the win-64 package (osx and linux are available now: https://anaconda.org/conda-forge/libignition-rendering4)