Stephan Bergmann wrote on Wed Nov 29 11:16:24 UTC 2017,

>Is there any good reason why a failed Gerrit/Jenkins build does not
>set the corresponding Gerrit change's Verified flag to -1 [...]

This is Gerrit/Jenkins integration bug, obviously. Norbert may know
more, but my impression is that this is a left over from the test
period, where CI verification in gerrit and voting VRFY+1/+1 was
flaky, and the main purpose of skipping of voting VRFY-1 in
gerrit was to avoid false negative votes.

Still, even if the CI build still considered to be experimental today,
(is it?) I would swap the logic and do the opposite: fail the build,
and ask change owner (with a help of maintainers) to remove VRFY-1
vote in case the negative vote wasn't justified.

_______________________________________________
LibreOffice mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Reply via email to