TestInterop.test_splrep_errors seems to be the single failing test in a variety of POSIX and Windows wheel build scenarios

This appears to be more prone to failure in wheel builds on Windows or POSIX that use older versions of NumPy in combination with new OpenBLAS.

I don't know why, yet.

On Wed, 6 Feb 2019 at 16:15, Tyler Reddy <tyler.je.reddy@gmail.com> wrote:

1.2.1 wheel builds are now running, including the OpenBLAS 0.3.5.dev version bump from Chuck in the wheels repo.

Fingers crossed!

On Mon, 17 Dec 2018 at 17:44, Charles R Harris <charlesr.harris@gmail.com> wrote:

On Mon, Dec 17, 2018 at 12:15 PM Tyler Reddy <tyler.je.reddy@gmail.com> wrote:
Currently waiting on 1.2.0 wheel builds as of this morning. Appveyor can take about 4.5 hours to handle that.

I like to start the builds in the evening, then go to bed :)

<snip>

Chuck
_______________________________________________
SciPy-Dev mailing list
SciPy-Dev@python.org
https://mail.python.org/mailman/listinfo/scipy-dev