From MozillaWiki
Jump to: navigation, search

Weekly Gaia Development Meeting

Announcements & Notices

   1.3[t]/1.4 blockers - a few unowned 1.4 blockers, don't everyone pile on now
   Daily testing of master for great success
   Welcome Eli!
   Welcome Sam!
   Welcome Shel!
   Performance team in Taipei next week


reporter: jason smith

  • Priority focus has been on 1.3/1.3T testing, then 1.4 testruns/bugs. 1.5 focus will resume next week after we discontinue 1.3 smoketest runs.
  • handling branch-landing patches only, that dont land in m-c/master (ie. Bug 982530). how are we keeping this in sync?
    • Discussed: reopen the bugs and they should only switch status to RESOLVED FIXED when it has landed in master/m-c.
  • We are still behind on hardware. Flame has WSOD issues and many releng builds aren't available yet for shipping devices (ie. Open C). Note that bugs being filed are against builds that may have differences than self-built builds.
  • 1.3
    • Smoketests are passing - 48/48
    • Ongoing QA Wanted & Verification Support As Needed
  • 1.3T
    • Latest smoketest Result - 49/49 cases passing (note, still waiting on releng builds)
    • Tarakorun 1: (whiteboard "1.3tarakorun1"): 812 passed, 45 failed, 0 blocked, 11 invalid / ~2640 cases total. Coverage is at 29%
  • 1.4
    • Latest Smoketest Result - 48/48 tests passing
    • Burirun 1.4-2: (whiteboard "Burirun1.4-2") 2114 passed, 107 failed, 0 blocked, 72 invalid / 4404 cases total. Coverage is at 50%
  • 1.5
  • Gaia UI Automation

131 tests ran in 8395 seconds. 126 passed, 4 skipped, 1 failed, 2 errors. 2 expected failures, 0 unexpected passes. Test failures: - Bug 987024 - [Contacts] Can not make a call from Contacts - Bug 987145 - Investigate failure in - Bug 976028 - test_play_ogg_video failing on TPE CI Tests blocked by existing bugs : - Bug 946130 - Flick and other actions not working on System app - Bug 932804 - Tapping on select all button doesn't work on Gmail or Outlook frame, before importing contacts




Talking this week: Gareth Sprint tracking wiki: Notes: The Team: asuth, evanxd, gaye, lightsofapollo, doliver, jrburke, mcav, jhford, cserran, pdol, tony (qa), william hsu (qa), jhuang (ux), harly (ux) [email] asuth, jrburke

   email blockers for 1.3, 1.4 having to do with notifications, ui breakages, apz/scrolling
   e-mail integration test refactoring completion, trying to overlap as much as possible with blockers

[calendar] gaye, millemedeiros, evanxd

   calendar integration test work, lots of reported blockers which turned out to be gecko regressions
   Visual refresh
   Jshint compliancy

[clock] mcav

   they just keep coming, the older 1.3 gets the more blockers we have
   second pass on review from :miller about py2js test refactoring
   i'm getting an intern!

Media front end

Talking this week: read-only

   Our   team: Dave Hylands, David Flanagan, Mike Habicher, Jim Porter, Hema   Koka, Dominic Kuo, John Hu, Diego Marcos, Wilson Page, Punam Dahiya   (part-time), Justin D'Arcangelo (part-time) Product: Sri Kasetti Ux:  Rob  MacDonald, Patryk Adamczyk EPM: Candice Serran QA: Marcia Knous
   Entire team busy with getting madai camera features (12+ features) ready to land on master from camera branch in the next day or two. Please help test and file bugs! (
   Lot of incoming gallery and music (audio channel) related bugs and enhancements for tarako. Engineers in Taipei are also helping fix these issues (divide and conquer! thank you!!)
   Continuing to work on other blocker issues for 1.4


Talking this week: mchang

   TEAM: bkelly, eli, geo, hub, huseby, jhylands, jld, kgrandon, mchang, mlee, wlach, Walter Chen
   Flame (mozilla reference device) 
   has a switchable USB charger, so we can take power measurements while the phone is plugged in via USB
   also has a unique serial number, which allows multiple devices to be plugged into the same host PC
   these two features make it possible to use it for power regression testing on a large-scale basis
   HEADS UP!! I conclusively showed last week that mercurial's topographic sort for bisecting BREAKS when the "oldest" end of the bisect range isn't a common ancestor.  So if you get a message telling you to use "hg bisect --extend" you have to re-run your entire bisect using the common ancestor it gives you.  See Bug 960224 for details.  I have a blog post in the works to explain all of the details (with pretty graphs) and why mercurial breaks. - talk to :huseby if you hit this probem
   BTW if you want to bisect using git, there's a master "map file" that maps hg <---> git revisions that can be found here:

Comms app

Talking this week: ? Past week we had our Paris Comss working week. Several meetings including topics:

 - Spliting comms into independent apps.
 - Visual refactor for all comms apps.
 - Spliting sms and contacts into several documents.
 - Using DS in contacts.
       - Multiple delete in contacts landed
       - We landed absolutely all bugs blocking DSDS!

System front end

Talking this week: Michael H

   Sprint tracking wiki:
   Daily standups:

The team: cserran, gwagner, ladamski, kyle m, michael h, aus, sam, alexandre, francis, jason, atsai, pdol, benfrancis, daleharvey, naoki, john, tef

   Download Manager/API fixes, improvements
   B2G-Desktop OOP, fixing up unit tests
   Offline experience just landed
   Replaceable Homescreen prototype


System platform


Media recording


Round Table