Closed pfefferle closed 4 years ago
Thanks for the PR!
I like trading off the amount of revisions listed for reporters.
Am thinking we should still have some limit on reporters to avoid having a limitless query. Maybe we can pick something arbitrarily larger that should still work, but might be a good time to re-evaluate how they are retrieved?
Good point! I changed the code, so that it displays 15 items on the main page and all items on the "single revision" page. We could also try to sort them by "errors first", but perhaps in a separate merge request.
Ah, I like that direction. I had been thinking setting both to something ridiculous like 100(+), so that everyone gets listed, but things are protected from edge cases.
On the latter note, a couple of thoughts:
test-results
(with no individual results at all in the table), then show all host results (with some very high limit to avoid edge cases) when clicking through. This option would make it clear what particular revision(s) might have broken/fixed things.Sorry this has taken so long! Wanted to be sure to update: @aaroncampbell offered to see if we can get some test data so as to be able to test more easily. If that's possible -- awesome! If not, I'll still go ahead with making a mocked set of test users + results.
Thanks much!
Thanks much!
Wanted to make sure to props @dd32 and @aaroncampbell for test content from make. Thank you!
fix #62