Closed johnbley closed 2 months ago
Blergh. Apparently there is some combination of (a) things that lack of set -e
was masking which were failing before, and (b) things where I should have been more careful applying set -e
and that are expected to fail. Will dig deeper when I have more time.
I'm going to close this and hopefully come back to do it piecewise (file-/directory- at a time). It certainly seems like a scheduled daily "full CI" might help with reducing bitrot/churn here as some of these tasks don't seem to have been run for weeks/months.
I'm going to close this and hopefully come back to do it piecewise (file-/directory- at a time). It certainly seems like a scheduled daily "full CI" might help with reducing bitrot/churn here as some of these tasks don't seem to have been run for weeks/months.
@johnbley
Thanks.
This repository indeed has a lot of bitrot, no idea what was wrong with the failing tests earlier.
set -e
for almost all shell scripts (one exception which is commented)shellcheck --severity=error
shellcheck