Auto-tools/Projects/Mozbase/Automation-Refactor/Meetings/2013-06-18
From MozillaWiki
< Auto-tools | Projects | Mozbase | Automation-Refactor
Contents
Details
June 18th, 2013 @ 11:30am PDT / 2:30pm EDT
dial-in:
- vidyo: Jgriffin's vidyo room
Backchannel #ateam on irc.mozilla.org
Take meeting notes here: https://etherpad.mozilla.org/great-automation-refactor Then copy and paste them below afterwards
Meeting Notes
Quick Status Updates
- [ahal] using in-tree mozbase for test harnesses now (bug 855893)
- [ahal] import mozbase by adding modules to sys.path instead of crazy hack
- [jhammel] make check locally works, try run coming soon, optimistic it'll stick this time
- [chmanchester] starting on the javascript side (log4moz.js): see https://bugzilla.mozilla.org/show_bug.cgi?id=884397
- [jmaher] close to finishing manifests for mochitests (bug 852416)
- [ted] working on mozinfo integration with test harnesses, added find_and_read_json method
Issues of the Two Weeks (since you looked at me)
- Agenda for mozbase work week? Propose items below
- brainstorm the creation of a mozbase "manifesto" -- both a high-level statement vision of the purpose of the project, and a detailed roadmap of how to make it live up to that vision [jhammel: a big +1][nalexander: another big +1 here]
- what components are we missing, if any?
- where should shared code live that doesn't belong in a mozbase component?
- [jhammel: I would (OTTOMH) go for a manifesto with a clear, short exec summary==vision statement and then a longer (not very long) high-level explanation of how all this fits together: an aggressive plan of attack to the stars]
- I would tend
- machine readable logging - how to get off to a running start? [chmanchester]
- figure out and prototype approach to multiprocessing in tests? likewise get off to running start [mineahdb]
- we are starting with xpcshell apparently: https://bugzilla.mozilla.org/show_bug.cgi?id=660788
- reftests are also being worked on in bug 813742.
- strategizing on how to write tests for ffledging's gsoc project? [jhammel: +1; i'd be willing to triage andor audit/help triage andor audit to this end]
- finish up mozrunner refactor and make sure it is a viable replacement for mozb2g
- mozb2g/eideticker, can we replace it? What's the best path forward?
- moztest? common code etc.
- manifestdestiny/mozprocess lots of work to do (tracking bugs?)
- brainstorm the creation of a mozbase "manifesto" -- both a high-level statement vision of the purpose of the project, and a detailed roadmap of how to make it live up to that vision [jhammel: a big +1][nalexander: another big +1 here]
Decisions for /the future/
- CI failures: to note on bug?
- add to policy :)
- mock, etc, libraries in testing?
- leaning towards not using external dependencies unless absolutely necessary
Mozbase, Packaging, Mirroring and the Future of M-C
Areas of Development
Mozbase has essentially two domains in need of development effort: improvement of the mozbase software and integration of the mozbase packages with (largely legacy) targetted consumers.
- manifestdestiny
- mozprocess
- tests (ffledgling)