[Libreoffice-qa] On the topic of Backporting bugs fixed on master

Robinson Tryon bishop.robinson at gmail.com
Mon Feb 2 09:09:30 PST 2015

[ moved to better forum from
https://bugs.documentfoundation.org/show_bug.cgi?id=89039 ]

Comment # 5 on bug 89039 from V Stuart Foote
Regards resolving WFM, guess that is OK if that is the QA consensus.

But the logic escapes me then of even retaining MABs (or tracking Metas). If a
bug shows as resolved in the MAB listing (or any tracking Meta)--it does not
get further reviewed--and then we have to drag BZ for the issues.

Sure. That's us running into the limitations of Bugzilla, really.

Which then
requires an advanced query against Whiteboard field to match regular expression
"backportRequest" or similar strings.

Queries are here, for reference:

With 4.4.0 just out of the gate, valid new MABs are going to bubble up against
the branch--I just don't see a reason to close them so quickly and then to
depend on someone doing a "reverse" bibisect and back-port requests.

Regardless of what we mark the bugs in Bugzilla, our fastest tool for
figuring out which commit fixed a bug is by a reverse-bibisect. And
the backportRequest:X.y tag is a very specific way for us to
communicate our intentions with the developers.

We don't currently have a way to indicate a status for each branch
that's active, but if we were to add that, then we could say NEW for
4.4 and RESOLVED for 4.5. I'm not sure of the best way to add that
without making things more confusing... Bugzilla isn't quite that

can't even be done for OS X or Windows issues.

Well, we can do some limited bibisecting on both platforms. I really
hope that in 2015 we can make bibisecting much more doable for both
Win and OSX.


Robinson Tryon
QA Engineer - The Document Foundation
LibreOffice Community Outreach Herald
qubit at libreoffice.org

More information about the Libreoffice-qa mailing list