You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The current version of python-control is incompatible with scipy-1.0.0 and so all checks are failing. In this build script we do explicit checks against scipy-0.19.0 and scipy-1.0.0. This is related to issue Compatibility issue with SciPy 1.0 #164.
The way that tests were being done against slycot, they were actually not checking to make sure slycot was installed correctly => they 'succeeded' if slycot failed to install. The new version of the build script separates out the slycot versus non-slycot builds so that this is more obvious. Issue Slycot installation failing in Travis CI #168 describes the issue.
In addition to these changes, I updated the python version numbers to check against to 2.7, 3.5, and 3.6 (versions 3.3 and 3.4 seem old enough that we can stop checking against them).
If this PR works correctly, it should have a revised set of Travis CI checks that succeed on the 0.19.1 version of scipy with slicot not installed, but fail on all other cases.
Update PR to include new rules for building slycot from source, which allows proper checking against the slycot version of the library. A couple of notes on the implementation:
Need to be careful about how to install gfortran and lapack. Install both via apt-get in the root environment (not the test-environment) to make sure that compilers use the right paths (found this out the hard way, but trying almost every other combination of things).
Python3 implementations need to have the LIBRARY_PATH environment variable set up properly, otherwise slycot will fail to link properly. For this, you need to link to the version in the local environment, so that you get the right version of python.
While I was at it, I made a few other changes:
Reordered the build cases to start with the most up-to-date case (python-3.6, scipy-1.0, slycot) gets tested version.
Added more comments throughout to explain what is going on.
Don't bother to install conda-build since it is not needed any more.
Some of the Travis builds will still fail for this version since the corrections for scipy-1.0 from PR #164 are not yet included.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
This PR addresses two related issues:
The current version of python-control is incompatible with
scipy-1.0.0and so all checks are failing. In this build script we do explicit checks againstscipy-0.19.0andscipy-1.0.0. This is related to issue Compatibility issue with SciPy 1.0 #164.The way that tests were being done against slycot, they were actually not checking to make sure slycot was installed correctly => they 'succeeded' if slycot failed to install. The new version of the build script separates out the slycot versus non-slycot builds so that this is more obvious. Issue Slycot installation failing in Travis CI #168 describes the issue.
In addition to these changes, I updated the python version numbers to check against to 2.7, 3.5, and 3.6 (versions 3.3 and 3.4 seem old enough that we can stop checking against them).
If this PR works correctly, it should have a revised set of Travis CI checks that succeed on the 0.19.1 version of scipy with slicot not installed, but fail on all other cases.