PyPI skeleton not coping well with environment markers?

[msarahan]

hdf5storage is failing in our test suite right now, with errors like

Traceback (most recent call last):
  File "/home/travis/build/conda/conda-build/tests/test_api_skeleton.py", line 150, in test_setuptools_test_requirements
    api.skeletonize(packages='hdf5storage', repo='pypi')
  File "/home/travis/build/conda/conda-build/conda_build/api.py", line 281, in skeletonize
    recursive=recursive, config=config, **kwargs)
  File "/home/travis/build/conda/conda-build/conda_build/skeletons/pypi.py", line 274, in skeletonize
    setup_options=setup_options)
  File "/home/travis/build/conda/conda-build/conda_build/skeletons/pypi.py", line 737, in get_package_metadata
    sys.exit("Error: Could not parse: %s" % dep)
SystemExit: Error: Could not parse: numpy ; python_version == '2.7'

The requirements coming out look like:

"numpy < 1.12.0 ; python_version == '2.6'", "numpy ; python_version == '2.7'", "numpy<1.6.0 ; python_version == '3.0'", "numpy<1.8.0 ; python_version == '3.1'", "numpy<1.12.0 ; python_version == '3.2'", "numpy<1.12.0 ; python_version == '3.3'", "numpy ; python_version >= '3.4'", "h5py>=2.1 ; python_version == '2.6'", "h5py>=2.1 ; python_version == '2.7'", "h5py>=2.1,<2.4 ; python_version == '3.0'", "h5py>=2.1,<2.7 ; python_version == '3.1'", "h5py>=2.1,<2.7 ; python_version == '3.2'", "h5py>=2.1 ; python_version >= '3.3'"

these need to be translated into deps that conda-build understands.

[github-actions]

Hi there, thank you for your contribution!

This issue has been automatically locked because it has not had recent activity after being closed.

Please open a new issue if needed.

Thanks!