In order to support T102920: What to do with old open patches for unmaintained/inactive repositories when not even the uploader responds, patches with Verified -1 should not be counted in our code review metrics, regardless of their CR value.
Just wondering, is there a Verified -2 value?
Just wondering, is there a Verified -2 value?
Depends on each repo's configuration but no "verified: -2" exists in our Zuul configuration currently.
@Qgil clarified "In the same way we filter CR-1; Jenkins can vote in a different and automatic way so we have some patchsets that are e.g. CR-0 but V-1". So these patches are not waiting for a reviewer but they are waiting for an author/submitter to update/rebase.
@Dicortazar said "shouldn't be too hard because it's a global filter but we should have a look, because of KPIs in another path in the library".
@Dicortazar to check if such changesets are currently accounted for author/submitter.
Wondering if having that Verified -1 in mind, the definition of 'open' is the definition of 'patchsets waiting for a submitter action'.
The question is: do you want to mark as 'open' only the changesets waiting for a submitter action? (so they were either reviewed or negatively verified).
At http://korma.wmflabs.org/browser/gerrit_review_queue.html we have "New" and "Waiting for review", and then "Time from last patchset" which is calculated from the 'waitig for review'.
Verified -1 would still count for "New", but not for "Waiting for review", and therefore not for "Time from last patchset".
I guess we've worked on that in the past. It seems and I think it is already fixed. Checking the code in GrimoireLib:
Any evidence this is failing or something?
Ah, thanks!
Looks to me like you're right from a quick test: I tried with "Gadgets" on http://korma.wmflabs.org/browser/gerrit_review_queue.html?page=22 listed as zero items waiting for review. That repo has one open patch in total in Gerrit and that one patch is V=-1 and CR=0 and from Oct 9.