Mobile/Notes/14-Aug-2013

From MozillaWiki
< Mobile‎ | Notes
Jump to: navigation, search

Details

  • Wednesdays - 9:30am Pacific, 12:30pm Eastern, 16:30 UTC
  • Dial-in: Audio-only conference# 99998
    • People with Mozilla phones or softphones please dial x4000 Conf# 99998
    • US/Toll-free: +1 800 707 2533, (pin 4000) Conf# 99998
    • US/California/Mountain View: +1 650 903 0800, x4000 Conf# 99998
    • US/California/San Francisco: +1 415 762 5700, x4000 Conf# 99998
    • US/Oregon/Portland: +1 971 544 8000, x4000 Conf# 99998
    • CA/British Columbia/Vancouver: +1 778 785 1540, x4000 Conf# 99998
    • CA/Ontario/Toronto: +1 416 848 3114, x4000 Conf# 99998
    • UK/London: +44 (0)207 855 3000, x4000 Conf# 99998
    • FR/Paris: +33 1 84 88 37 37, x4000 Conf# 99998
    • Gmail Chat (requires Flash and the Google Talk plugin): paste +1 650 903 0800 into the Gmail Chat box that doesn't look like it accepts phone numbers
    • SkypeOut is free if you use the 800 number
  • irc.mozilla.org #mobile for backchannel
  • Mobile Vidyo Room

Schedule

  • Next merge: 2013-09-16

Topics for This Week

Call for Fig Testing
Lucas posted to the mailing list about getting some additional testing on Fig Nightlies. See his post for details.

Friends of the Mobile Team

Give a shoutout/thanks to people for helping fix and test bugs. Make sure friends also get awarded a badge.

  • Jinsu (heyjinsu) landed a patch for bug 818987 - Links to media files should have an "Open in app" option
  • Michael Boon (michaeltpb) landed a patch for bug 888982 - fennec should use channel-specific build defines rather than MOZ_UPDATE_CHANNEL
  • Capella filed and fixed bug 903871 - JS crash in SelectionHandler during longtap text select on target page

Stand ups

Suggested format:

  • What did you do last week?
  • What are working on this week?
  • Anything blocking you?

Please keep your update to under 2 minutes!

James W. (snorp)

Kats

  • Last week
    • bug 896547 - (landed) fix a regression in the dynamic toolbar code
    • bug 902888 - (landed) flash elements not sized properly on HiDPI pre-honeycomb devices
    • bug 901962 - (discussing) can't click outside the CSS viewport area
    • bug 890280 - (mostly landed, one patch pending) fix up the input event transformations for multi-APZC
    • bug 732971 - (waiting review) make setResolution work on subdocument presShells
    • helping botond, reviews, etc.
  • Next week
    • more work on input event transformations for multi-apzc, how it interacts with event fluffing and guiding botond and/or implementing the focus model code that will be needed for dynamic toolbar on B2G and proper pinching behaviour.

Chris Lord (cwiiis)

Shane Tully

  • Still tracking down cause of two intermittent test failures with the contacts API on Android 2.2 Annoying the A-team
    • Should be getting a physical tegra to test on today though
  • bug 880118 - Package GeckoView into an Android library project
    • Last week:
      • Pages loading (actually working pretty well!)
      • Patches starting to be reviewed
    • This week:
      • Integration into build system
      • Landing patches

Next week:

  • Last week here! :(
  • Working on an intern presentation I guess

JChen

  • Past week
    • Finished patch+test for bug 897221 - dynamically updatable UA override mechanism
      • Getting reviewed
    • Fixed bug 904646 - GDB doesn't see shared libraries from custom linker on Android 4.3
    • Fixed bug 899233 - Text input is screwy in textarea
    • Doing Science Lab review
    • Investigated blassey's restart crash
  • Next
    • UA override follow up
    • Some more Eideticker issues
    • Get back to profiler work

GCP

  • Last week:
    • SafeBrowsing Download protection reviews
    • bug 902000 Service to monitor the system load
  • Next week:
    • Finish 902000
    • bug 877954 Adapt video encode resolution & framerate according to available bandwidth and CPU use

Botond

  • Done
    • Tested patch for bug 895905 - "Contents needs to inform APZC about scroll events on B2G". Waiting for one minor issue to be resolved before it's ready to land.
    • Finished and tested patch for bug 898580 - "Contents needs to inform APZC about scroll events on Metro". Waiting for bug 895905 to land before this one can land. Also, Brian Bondy is experiencing an issue which I can't reproduce and which may or may not be relatd to this.
    • Began working on bug 860812 - "[Browser] Make hiding address bar work more like Fennec Nightly".
  • Next

Brian N

  • Done
    • Cruise, booze, and flus
    • Bug 901903 - The guest profile is created at every startup
    • Bug 899141 - URLs are not opened from other apps if the "Don't keep activities" option is set
    • Created etherpad to come up with a plan for multi-engine suggestions: https://etherpad.mozilla.org/wMC375bLW3
    • Trying to get some session restore tests working, but was hitting bug 901811
  • Next
    • Get those doggone session tests working already

Sriram

  • bug 902064 - Folders on BookmarksListView throws IllegalArgumentException.
  • bug 902038 - Use a multi type cursor adapter for our adapters.
  • bug 880525 - Search suggestions are misaligned on tablets.
  • bug 888905 - Reading list for tablets.
  • bug 904840 - Lazily inflate history page.
  • bug 862801 - Exploring about:home snippets.
  • More and more reviews.
  • Blog on grilist view.

WesJ

  • Bug 808636 - Update visual style of mobile error pages
  • Bug 901426 - Web Apps can not be opened from Firefox for Android notification and crashes Firefox . lots of bisecting
  • Bug 788073 - Platform touch fluffing. Bounced, debugging.
  • Bug 889620 - [FIG] Transition for creating a new tab from the tab tray
  • Bug 840989 – favicon doesn't refresh with connectivity after no connectivity)
  • Playing with writing some new color/date/time pickers in my spare time. Very WIP: https://github.com/wesj/Pickers

Shilpan

  • r+/landed
    • bug 894077: [FIG] Tablet layout for visited page
    • bug 891183: [FIG] Change Title strip to Tabs strip for tablets (with custom animation!)
  • In progress
    • bug 900234: Force reader to always be the left most icon in page actions
    • bug 899376: PageAction for helper apps

LucasR

Last week

  • bug 905162 - Misc empty view tweaks
  • bug 871651 - Janky animation to enter editing mode
  • bug 904689 - Implement empty screen state for "Most Visited" page
  • bug 899550 - Disable editing mode animations on pre-Honeycomb devices
  • bug 905139 - Fix alpha setter for pre-HC AnimatorProxy
  • bug 905088 - Properly handle default URL on BrowserApp's enterEditingMode
  • bug 871652 - Disable toolbar's context menu while in editing mode
  • bug 903448 - Update testHistory for new about:home
  • bug 903478 - Add testMostRecentPage for new about:home
  • bug 897483 - Add testMostVisitedPage for new about:home
  • bug 897481 - Update testShareLink for new about:home
  • bug 899183 - Update testBookmarklets for new about:home
  • Tons of patch reviews
  • Prioritizing fig bugs, defining readiness parameters for the merge
  • Proposed merge plan:

Next week

  • Gradually land all ported tests in fig
  • General prep work for the about:home merge to m-c
  • More about:home hacking

Liuche

  • Fig empty pages:
  • uplift things to aurora, reviews

Margaret

  • Done
    • TRIBE (if you're interested, you can learn more/sign up on the TRIBE wiki page)
    • bug 899393 - [fig] Switch-to-tab is broken
    • Reviews
  • In Progress
    • bug 897772 - [fig] "Add to Home Screen" shortcut doesn't have an icon sometimes
    • Started investigating bug 848242 - Add support for widget/tiles to homepage customization
  • Next
    • Help do whatever needs to get done to merge fig into m-c

Chris Kitching

  • This week
    • bug 897772 - Got catapaulted into this bug. Still not entirely sure which direction it's going to go in. Working on implementation of native encode/decode of JSON. Native decode implementation seemingly working, but causing absolutely every unit test to segfault (Some value of "working"). Unclear how to debug, since try doesn't test debug builds and segfault doesn't seem to occur locally. It IS somewhat faster, although the majority of the speedup is expected from the native encode anyway, which I'm yet to write.
    • bug 794981 - Dragging on longer than I'd hoped. Now essentially working, being thwarted by different types of nsString being used in different places in AndroidBridge. Generated code has no way to decide which such type to use. Seems to necessitate refactoring all callers to use a consistent flavour of string. Hopefully finishing up today.
    • bug 709230 - Still blocked on the above...
    • bug 904279 - This appeared. Initially thought to be (Yet another) regression from prior search config bug. Turns out to be a rather subtle bug that we just shipped. Whoops. Wrote an almost entirely unsuccessful patch to correct the issue (Seems my ability to reproduce this is somewhat lesser than Margaret's). Looks like the multi-engine suggestion thing Brian N just mentioned might accidentially fix this anyway..
    • bug 892125/bug 892136 - Work continues - slowly for now - added missing JS helper script to dump needed information into Java on request, should now be straightforward to add UI to list addons and provide common configuration utilities. Remaining interesting part of this task is the addition of a system to parse the addon-defined xul files into Android preference screens. Hopefully will be less frantic to land than the last UI feature I added..
    • bug 895423 - Filed what is hopefully the final needed version of the patch for this one. Turns out Android styling was actually even more strange than I had first thought. I was presented with a Samsung Galaxy Ch@at to test on - this device is precisely as insane as its name suggests - found a bunch of weird edge cases in my patch making the icons not display correctly. Ended up essentially rewriting FaviconView. Icon display now consistent across all devices within easy reach from my desk. Pending review.
  • Next
    • Finish bug 794981, then bug 892125/bug 892136, followed by bug 709230 and bug 897772.
    • Next project - It was suggested that it'd be nice to get the Javadoc tool to work on the Fennec source code to produce something approximating documentation. Others are welcome to steal this.. Should be straightfowardish if you point it at the post-preprocessed code where applicable (Preliminary work suggested that the main problem with this was it getting confused by our preprocessed Java files. It's also possible that bug 890586 blocks this...)

mcomella

  • Past
    • bug 900289: Enable foreign key pragma on FHR database
    • bug 902039: [fig] Correct about:home back-button/swipe UX
      • (WORKSFORME) New UX is not viable
  • Present
    • bug 904122: (f+) [fig] Scrolling on search suggestions page does not dismiss the keyboard
    • bug 870925: (f+) Add submissions info to FHR documents
    • bug 870171: (f+) Expire old FHR data
  • Future

RNewman

  • Talking to dozens of people.
  • Sync.next roadmapping.

Scott (jwir3)

Last Week:

  • bug 900564 : Reflow-on-zoom now removed from settings screen for beta.
  • bug 794693 : Crash in ContainingBlock reflow with font-inflation.
  • Reflow on zoom now disabled on <video>, <image>, etc...

This Week:

  • Still working on bug 878935: suppress repainting during reflow.

BLassey

  • bug 894313 - GeckoThread should own its own static instance, landed
  • bug 880259 - Firefox should use GeckoView, landed, bounced, new patch waiting for review from finkle
  • continuing to look at restarts

MFinkle

  • Setting the Stage
  • PiCL planning: Getting to part where we code
  • More added to Chromecast support in bug 901803
  • Scanning upcoming Android conferences for Mozilla/Firefox opportunities

Ian Barlow

This week:

  • Unloading my brain onto Arun!
  • Thinking about what we can do with third party "save for later" providers
  • Thinking about First Run UX
  • Thinking about Chromecast stuff
  • Ongoing Awesomescreen polish. It's nice!

Security/MGoodwin

WebRT

PiCL

Performance

A-Team

Feature Focus

Fx24
Fx25
Fx26

Round Table

User Advocacy

We are still watching the Google Play issue, appears to be an infrastructure issue.

Now that we have fond the large issues in 23 (4.3 support and google play) we are beginning to dig into the other feedback to look for smaller issues.