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
Right now I think it's possible to introduce lint failures (or Python unit test failures) by committing directly without Travis checks (example).
AFAICT, we're not currently continuously running these tests for the master branch, but I've seen for other GitHub repos that they can have a current build status indicator (https://docs.travis-ci.com/user/status-images/).
Is this right? If so, how hard is it to enable continuous runs for some things? (I imagine stability checks would take way too long, but lint and Python tests might be reasonable?)
Right now I think it's possible to introduce lint failures (or Python unit test failures) by committing directly without Travis checks (example).
Admins can push without passing Travis checks, though we rarely do. We could potentially just require admins to pass Travis checks first? It's sufficiently rarely used it's unlikely to make much difference. @jgraham?
AFAICT, we're not currently continuously running these tests for the master branch, but I've seen for other GitHub repos that they can have a current build status indicator (https://docs.travis-ci.com/user/status-images/).
qyearsley-zz
changed the title
Run continuous (not per-PR) Travis checks
Run continuous Travis checks (edit: Make failing status more obvious?)
Aug 10, 2017
Right now I think it's possible to introduce lint failures (or Python unit test failures) by committing directly without Travis checks (example).
AFAICT, we're not currently continuously running these tests for the master branch, but I've seen for other GitHub repos that they can have a current build status indicator (https://docs.travis-ci.com/user/status-images/).
Is this right? If so, how hard is it to enable continuous runs for some things? (I imagine stability checks would take way too long, but lint and Python tests might be reasonable?)
@gsnedders @bobholt
The text was updated successfully, but these errors were encountered: