Set up continuous integration/deployment, and conda binary install

sandialabs / Spitfire

Spitfire is a Python/C++ library for constructing tabulated chemistry models and solving differential equations.

Other

37 stars 9 forks source link

Set up continuous integration/deployment, and conda binary install #6

Closed michael-a-hansen closed 3 years ago

michael-a-hansen commented 3 years ago

A good intermediate here would be to call CMake to build the Griffon library directly from setuptools, which would also simplify the build for developers.

michael-a-hansen commented 3 years ago

Conda is being a pain and so this hasn't really gone anywhere, but a nice improvement is that CMake is no longer required and Griffon is compiled into an extension automatically with setup.py.

michael-a-hansen commented 3 years ago

Cantera support, at least 2.3 and 2.5, is necessary in CI.

michael-a-hansen commented 3 years ago

I'm giving up on conda install for the time. Travis CI is set up but can't find blas and lapack when linking the Griffon extension.

michael-a-hansen commented 3 years ago

Done! Travis CI's badges are unstable and it says it's failing despite everything showing successful builds. So whatever...