Closed xrotwang closed 4 years ago
It may still become a bit of a problem, if many datasets trigger warnings about things that cannot be fixed, e.g. about bookkeeping languoids with no alternative. But I guess, we'll have to see how this pans out in practice.
I like it!
Oh that’s cool. Everything done in 2m or so. This is going to be really helpful!
Looking forward to seeing this with >100 datasets on the server :)
And I think at ~100 lines of config, that's still fairly maintainable.
Just need to make sure we have lots of cores and we can just spawn a worker per dataset :)
Hitting "build all" in the evening and fixing warnings during the day is probably ok, too, to push out a release within a couple of days :)
I think it would be easier to integrate
cldfbench lexibank.check
withcldf-buildbot
if it returned a non-zero exit status, e.g. by raising an exception, if any warnings have been issued during the checks.