1112574 - Edge-swiping to the current app while the keyboard is up results in half of the screen being blank

Status: VERIFIED FIXED

(Firefox OS Graveyard :: Gaia::System::Window Mgmt, defect)

Description

[Chris Lord [:cwiiis]]

If launch two apps, focus a text entry in one (to bring the keyboard app), then begin edge swiping, but don't actually switch apps, the keyboard will disappear but the space that the keyboard occupied will remain blank (instead of the application resizing to fill the space).

Either the keyboard should remain and disappear on focusing the new app or the application should resize correctly when the keyboard disappears.

n?Patryk for UX comment on whether the keyboard should remain or not during edge-swiping.

Comment 1

[Patryk Adamczyk [:patryk] UX]

Francis would be best to comment on this as he owns the feature.

Comment 2

[Tim Guan-tin Chien [:timdream] (please needinfo)]

This looks like a regression to input mgmt -- we have the described behavior implemented back in v2.1 -- the keyboard should disappear and the app should resize correctly when you start the edge swipe.

qawanted for checking v2.1. Flagging v2.2 as affected per comment 0.

:mnjul, could you help here?

Comment 3

[Duane Dixon [:ddixon]]

Branch Check 

Issue DOES NOT occur in Flame 2.1, 2.0 (shallow flash, 319 MB memory, engineering). 

Actual Results: Edge swipping between apps functions correctly while keyboard is up. The app windows resize correctly when the keyboard collapses.  

Device: Flame 2.1
BuildID: 20141218131205
Gaia: 6af3d029bae3a14f400fec0926f0f8ad7b579b4b
Gecko: d41f6bd64343
Version: 34.0 (2.1) 
Firmware Version: v188-1
User Agent: Mozilla/5.0 (Mobile; rv:34.0) Gecko/34.0 Firefox/34.0
--------------------------------------------------
Device: Flame 2.0
BuildID: 20141218131305
Gaia: ce83ea7b8e3fa2d1c3fd771fc22b654c18b3c381
Gecko: f57067657bfa
Version: 32.0 (2.0) 
Firmware Version: v188-1
User Agent: Mozilla/5.0 (Mobile; rv:32.0) Gecko/32.0 Firefox/32.0

Comment 4

[Jayme Mercado [:JMercado] (Inactive 3/4/2016)]

Chris would you say this is the same as bug 1106651?

Comment 5

[John Lu [:mnjul][Please NI or I'll miss]]

Yeah I'll take a look. Keeping NI for reminding.

Comment 6

[John Lu [:mnjul][Please NI or I'll miss]]

Attachment: Screenshot 2014-12-19 4.45.38 PM.png

I tend to think this might be a WindowMgmt issue (or timing issue between WindowMgmt & InputMgmt).

I've added some ad-hoc console.log traces [1]; when this bug reproduces, I get the log trace as attached. Note that |keyboardhide| is correctly fired from InputMgmt, and AppWindowManager thinks that the active app is the Messages app -- both behaviors are correct. But the Messages app AppWindow thinks it's not active because it's still |will-become-inactive|, as in the log, AW.isActive checkpoints through 3 but not 4.

By simply commenting out the line on [2] this bug will no longer repro.

George, as offline discussed, could you take the bug from here and investigate further? Much appreciated :)

[1] https://github.com/mnjul/gaia/commit/f6ab892709b227270b3555b3738272487c1ecb7c . Note that AppWindow only emits the ad-hoc logs when it thinks it is the Messages app.

[2] https://github.com/mnjul/gaia/blob/f6ab892709b227270b3555b3738272487c1ecb7c/apps/system/js/app_window.js#L318

Comment 7

[(Inactive after June) George Duan [:gduan] [:喬智]]

Thanks John, this information is very helpful. Take this bug.

Comment 8

[Tim Guan-tin Chien [:timdream] (please needinfo)]

2.2? for regression.

Comment 9

[howie [:howie]]

Triage: regression, blocking.

Comment 10

[(Inactive after June) George Duan [:gduan] [:喬智]]

After bisect, I found it's a regression of bug 1096704.

Comment 11

[(Inactive after June) George Duan [:gduan] [:喬智]]

Hi Etienne,
could you help on this bug?
bug 1096704 will check will-become-inactive className when querying app.isActive, which will cause previous app cannot be resized to normal size after StackManager try to call goPrev/goNext (_queueBroadcast -> queueHide -> add will-become-inactive className).

Comment 12

[Etienne Segonzac (:etienne)]

(In reply to George Duan [:gduan] [:喬智] from comment #11)
> Hi Etienne,
> could you help on this bug?
> bug 1096704 will check will-become-inactive className when querying
> app.isActive, which will cause previous app cannot be resized to normal size
> after StackManager try to call goPrev/goNext (_queueBroadcast -> queueHide
> -> add will-become-inactive className).

Thanks for figuring this out! Working on a patch :)

Comment 13

[Etienne Segonzac (:etienne)]

Attachment: Gaia PR

Comment 14

[(inactive after 6/18) Alive Kuo [:alive]]

Comment on attachment 8540099 [details] [review]
Gaia PR

shouldResize() is a good start, r=me

Comment 15

[Etienne Segonzac (:etienne)]

https://github.com/mozilla-b2g/gaia/commit/ed2a43d0927f158d22dac0ac0baf994f9ac0d678

Comment 16

[Hermes Cheng (inactive after July 27, 2015)]

verified...

*2.2
Build ID               20150326164141
Gaia Revision          6d0174e28576f2f93e696a43d1ac3b03340117f6
Gaia Date              2015-03-26 21:47:33
Gecko Revision         https://hg.mozilla.org/releases/mozilla-b2g37_v2_2/rev/820c2f2e817a
Gecko Version          37.0
Device Name            flame
Firmware(Release)      4.4.2
Firmware(Incremental)  eng.cltbld.20150326.201153
Firmware Date          Thu Mar 26 20:12:05 EDT 2015
Bootloader             L1TC000118D0

*master
Build ID               20150326160206
Gaia Revision          525c341254e08f07f90da57a4d1cd5971a3cc668
Gaia Date              2015-03-26 16:34:16
Gecko Revision         https://hg.mozilla.org/mozilla-central/rev/59554288b4eb
Gecko Version          39.0a1
Device Name            flame
Firmware(Release)      4.4.2
Firmware(Incremental)  eng.cltbld.20150326.193247
Firmware Date          Thu Mar 26 19:32:58 EDT 2015
Bootloader             L1TC100118D0