Gaia/Meeting/2014-02-18

From MozillaWiki
Jump to: navigation, search

Weekly Gaia Development Meeting

Announcements & Notices

   We need a host for next week
   [bajaj] In case you missed the email on "Landing Procedure for 1.3 blocker bugs (1.3+)", please note that all the blocking bugs require explicit approval (approval-mozilla-b2g28+ / approval-gaia-v1.3+). Also refer : https://wiki.mozilla.org/Release_Management/B2G_Landing#v1.3.0

QA

reporter: tony chung

  • Continue to validate MWC devices and builds this week. ongoing test results: https://etherpad.mozilla.org/mwc-demos2014-qa
  • 1.3
    • 1.3+ bugs opened (29, down from 36 last week)
    • MWC Demo testing, ongoing: https://etherpad.mozilla.org/mwc-demos2014-qa
      • See etherpad for notable bugs.
    • Latest Smoketest Result -
    • Burirun 1.3-3
      • 3390 passed, 220 failed, 0 blocked, 89 invalid / 4364 cases total.
      • Coverage is at 84% (QAnalysts)
    • L10N run - first half completed Feb 10:
      • List of L10N bugs identified: mzl.la/LBCMNb (226 bugs in total as of now)
      • Second half of l10n testing goes Feb 20 - 26
    • Wrapping up 1.1->1.3 and 1.2->1.3 FOTA testing. Bug rollup:
  • Bug list
   [V1.0.1 to V1.3]
   Existed Bugs - 
   Bug#958942 [OTA][Helix] "Order by last name" setting is inconsistent after update from v1.2 to v1.3.
   Bug#944351 [OTA][Data Migration][Internet sharing] "Wi-Fi hotspot" and "USB tethering" is disabled after update from v1.1 to v1.2
   Bug#958945 [OTA][Helix] Configuration of app permissions for each app is inconsistent when device updates from v1.2 to v1.3
   New Found Bug - 
   Bug#971764 [OTA][Data Migration] "Wi-Fi usage" is disabled after updating from v1.0.1 to v1.3
   [V1.1 to V1.3]
   Existed Bugs -
   Bug#958942 [OTA][Helix] "Order by last name" setting is inconsistent after update from v1.2 to v1.3.
   Bug#944351 [OTA][Data Migration][Internet sharing] "Wi-Fi hotspot" and "USB tethering" is disabled after update from v1.1 to v1.2
   Bug#958945 [OTA][Helix] Configuration of app permissions for each app is inconsistent when device updates from v1.2 to v1.3
   New Found Bug -
   Bug#971816 [OTA][Data Migration] Voicemail number is empty after updating from v1.1 to v1.3

Support

   https://marketplace.firefox.com/app/mozilla-support new app for offline support articles

TEAM UPDATES

Productivity

Talking this week: Sprint tracking wiki: https://wiki.mozilla.org/FirefoxOS/productivity Notes: https://etherpad.mozilla.org/fxos-productivity The Team: asuth, evanxd, ianliu, gaye, lightsofapollo, doliver, jrburke, mcav, jhford, cserran, pdol, tony (qa), william hsu (qa), robmac

   E-mail compose UI's body composition area now is a contenteditable and scrolls with the rest of the compose UI. https://bugzil.la/837251 props to Kevin Grandon
   E-mail manual setup enhancements: we validate the e-mail address better https://bugzil.la/948580, and we support separate SMTP credentials and are friendlier about propagating settings in the manual setup UI https://bugzil.la/805501
   E-mail and calendar drawer visual refreshes fairly imminent
   Calendar visual refresh under way, but will hold off landing until given the okay to avoid communications apps looking outdated
   Clock alarms get volume control, https://bugzil.la/939197
   Calendar is moving to using a custom icon font for vector icons https://bugzil.la/972871 after extensive performance and memory comparisons of SVG and png options by :millermedeiros

Media front end

Talking this week: djf

   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
   Camera api changes landed in gecko and gaia. Beware of version skew issues
   Diego and Wilson travelling to Korea next week to work with LG engineers
   Ringtones work ongoing
   NFC file transfer code has landed in Gaia, but it still requires a custom Gecko build on an NFC-capable device like a nexus-4

Performance

Talking this week: mchang

   TEAM: bkelly, geo, hub, huseby, jhylands, jld, kgrandon, mchang, mlee, wlach, Walter Chen
   Sites
   Wiki: https://wiki.mozilla.org/B2G/Performance
   Sprints: http://scrumbu.gs/p/fxos-perf/
   Updates:
   Found a large performance improvement in the keyboard app - bug 971651
   150 ms sync reflow in keyboard - bug  

Comms app

Talking this week: Anthony Only two 1.3+ open. Contacts: NFC sharing close to land. Bulk delete is coming soon. Dialer: First Dual SIM patches coming together. Call log should open way faster when we land bug 969996. Messages: Read Reports for MMS landed. (1.4)

System front end

Talking this week: Gregor

   Sprint tracking wiki: https://wiki.mozilla.org/FirefoxOS/systemsfe
   Daily standups: https://etherpad.mozilla.org/fxos-systems-frontend1-4

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

   Rocketbar + haida pushed to 1.5 but still being worked on
   Download manager polish for 1.4

OPTIONAL UPDATES

System platform

RIL

Media recording

Device

Round Table

   note: some long standing bugs were fixed in the test-agent:
   shouldn't hang as often anymore: https://bugzilla.mozilla.org/show_bug.cgi?id=892048
   the full test-run now shows the progress as the test run https://bugzilla.mozilla.org/show_bug.cgi?id=907621
   Awesome !!
   note: we still have the Min reporter in Travis (means: no output), we might want to use Spec yet again.
   localStorage alternatives?
   bent would like to see a profile
   Kevin: Thoughts on ~~() function.
   https://github.com/mozilla-b2g/gaia/pull/16414/files#diff-da3cc7904e3f38b57a6bfa93d66123b0R35
   (bkelly) So we prefer |0 for forcing to int? ~~ is for Math.floor? It also forces to integer.  They both force to integer and truncate.
   Why not Math.round() instead of |0?
   Won't they both drop us down to 32 bits? ex: 4 billion max, versus the full 52 bits that float can get us under the hood before the number goes exponential?
   (bkelly) I see Math.floor() guarantees an integer return, so that works for me.  Just asking for clarification because I have mainly used ~~ for integer coercion in the past.  (I think emscripten uses |0, etc, but its the same thing.)
   (asuth) yeah, but emscripten is trying to force integer super fast-path, and it's required for asm.js now too.  I think in the meeting it was agreed if you need super duper fast path super-clever is fine.
   (bkelly) I'm cool with Math.floor().
   Group hug! \o/
   Julien: db migrations
   Who's hosting next week? Gregor!