I was just about to put together a PR to whitelist the test_mplot3d.py so that Travis would do these tests by default, when I discovered that the test wasn’t even available via the packaged install.
In setupext.py, we have a mpl_toolkits OptionalPackage as well as a tests OptionalPackage, which are tests for mpl proper, and not mpl_toolkits. Should the tests for mpl_toolkits be considered a separate OptionalPackage with dependencies on mpl_toolkits and tests?
I am already pushing the amount of free time I have to work on this, unfortunately.
Cheers!
Ben Root
Pinging on this issue now that we are about to include a new graph type in mplot3d.
My thinking is that the determination of inclusion of this new OptionalPackage should be made based on whether the regular tests package was specified to be included and if the mpl_toolkits OptionalPackage was specified to be included. Not exactly sure how that would get implemented, but I am just getting my thoughts down on paper…
Cheers!
Ben Root
···
On Fri, Feb 28, 2014 at 10:34 AM, Benjamin Root <ben.root@…553…> wrote:
I was just about to put together a PR to whitelist the test_mplot3d.py so that Travis would do these tests by default, when I discovered that the test wasn’t even available via the packaged install.
In setupext.py, we have a mpl_toolkits OptionalPackage as well as a tests OptionalPackage, which are tests for mpl proper, and not mpl_toolkits. Should the tests for mpl_toolkits be considered a separate OptionalPackage with dependencies on mpl_toolkits and tests?
I am already pushing the amount of free time I have to work on this, unfortunately.
Cheers!
Ben Root