Created attachment 553570 [details] [diff] [review] Patch v1 The title says it all.
Attachment #553570 - Flags: review?(sagarwal)
Hmm, I think this should be the other way round -- we shouldn't print rich failures unless we set an env var called (e.g.) MOZMILL_RICH_FAILURES. Then we can update buildbot and set the variable.
Comment on attachment 553570 [details] [diff] [review] Patch v1 yeah, let's do it the other way round unless you have an objection.
Attachment #553570 - Flags: review?(sagarwal) → review-
Yeah sure I just need to find the time to figure out how to talk to the buildbots so that they have the right env variable exported, that's all :)
Created attachment 554455 [details] [diff] [review] Patch v2
Created attachment 554456 [details] [diff] [review] Patch v3 Sorry, I'd left an extra print in there.
Comment on attachment 554456 [details] [diff] [review] Patch v3 Buildbot patches coming up. We'll need to get those in first.
Attachment #554456 - Flags: review?(sagarwal) → review+
Oh, so you need to define USE_RICH_FAILURES on top as None so that if that code doesn't get run (due to an error perhaps) then python doesn't complain even further about USE_RICH_FAILURES not being defined.
... as None or False.
Created attachment 554513 [details] [diff] [review] Patch v4 Do you want to check this in at the same time you update the build bots? (I don't know if that's the same repo or not).
It doesn't need to be simultaneous -- the buildbots can go first, since all we're doing is setting a currently unrecognized envvar.
Mark, gozer, could we get the buildbot patches in sometime?
buildbot patches (thanks!) have been applied to production, try, and try-unittest. these buildbot instances have been reconfig'd to make the changes live. Now you just need to apply Patch v4 to the tree(s)
Status: ASSIGNED → RESOLVED
Last Resolved: 7 years ago
Resolution: --- → FIXED
Target Milestone: --- → Thunderbird 9.0
You need to log in before you can comment on or make changes to this bug.