New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

CI test to make sure bokeh works without heavy dependencies #2596

Closed
mattpap opened this Issue Jul 21, 2015 · 2 comments

Comments

3 participants
@mattpap
Copy link
Contributor

mattpap commented Jul 21, 2015

There should be a separate test that makes sure that bokeh works without ~~~numpy~~~, scipy, pandas, etc. The minimal environment should contain python (obviously), six and preferably nothing more. Make sure to test not only if bokeh imports like this, but test as many plotting APIs as feasible

removed comments about charts

@mattpap mattpap added this to the 0.9.2 milestone Jul 21, 2015

@damianavila damianavila modified the milestones: 0.9.2, 0.9.3 Jul 24, 2015

@damianavila damianavila modified the milestones: 0.9.3, 0.9.4, 0.10.0 Aug 28, 2015

@damianavila damianavila modified the milestones: 0.10.0, 0.11.0 Sep 28, 2015

@bryevdv bryevdv modified the milestones: short-term, 0.11.0 Dec 25, 2015

@bryevdv bryevdv modified the milestones: 0.12.7, short-term May 9, 2017

@bryevdv bryevdv modified the milestones: 0.12.8, 0.12.7 Jun 30, 2017

@bokeh bokeh deleted a comment from canavandl Jun 21, 2018

@bokeh bokeh deleted a comment from damianavila Jun 21, 2018

@bryevdv

This comment has been minimized.

Copy link
Member

bryevdv commented Jun 21, 2018

I think I am inclined to just expand this test (or make a new one just like it):

https://github.com/bokeh/bokeh/blob/master/tests/test_no_tornado_common.py

to include pandas scipy and call it a day. I stopped trying to run the unit tests after having to install boto, colorama, requests, and PIL, and it's not clear to me that trying to run API code is that valuable. As a general rule we import at the top of modules so this kind of test should go a long way.

@bryevdv

This comment has been minimized.

Copy link
Member

bryevdv commented Jun 21, 2018

Well that doesn't quite work since we do conditionally import pandas if it is present in some places. We would need an actual different environment. I'm not sure it's worth it, we have not needed this so far and I think a pandas or scipy import would be noticed in review.

Other idea though, we could add a NO_PD_SP flag to the py3.5 unit test run, and have that rest explicity uninstall pandas and scipy before running unit tests

@bryevdv bryevdv referenced this issue Jun 22, 2018

Merged

test with minimal packages #8016

2 of 2 tasks complete

@bryevdv bryevdv modified the milestones: 0.13.x, 1.0 Jun 22, 2018

@bryevdv bryevdv moved this from Triage to Done in Focus: Build and Test Automation Sep 13, 2018

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment