871402 - [STK] launch browser doesn't work properly

Status: RESOLVED FIXED

(Firefox OS Graveyard :: Gaia::Settings, defect, P1)

Description

[leo.bugzilla.gecko]

Attachment: attach file include main, radio log and video

From attached logs,  when the launch browser command is executed,  it seems that  browser is launched. However, settings app is immediately brought again to foreground. (may be due to some STK commands, not sure though)

Comment 1

[leo.bugzilla.gecko]

if you want to take a test about this case, you must have VIVO operator sim card.
After insert simcard and finish booting, you can see STK menu in setting app.

Comment 2

[Leo]

(In reply to leo.bugzilla.gecko from comment #1)
> if you want to take a test about this case, you must have VIVO operator sim
> card.
> After insert simcard and finish booting, you can see STK menu in setting app.

After finish booting, in the idle mode - settings - Movistar - launch browser - browser basico ( it is located in setting menu)

Comment 3

[Alex Keybl [:akeybl]]

New feature, unimplemented requirement, or regression? needinfo ffos-product

Comment 4

[Leo]

when we confirm this issue, we found additional information. 

<gaia/apps/settings/js/icc.js>

function init() {
......
icc.addEventListener('stkcommand', function do_handleSTKCmd(event) {
      handleSTKCommand(event.command);
    });
window.addEventListener('stkasynccommand',
      function do_handleAsyncSTKCmd(event) {
      handleSTKCommand(event.detail.command);
      });
......

Comment 5

[Leo]

when we confirm this issue, we found additional information. 

<gaia/apps/settings/js/icc.js>

function init() {
......
icc.addEventListener('stkcommand', function do_handleSTKCmd(event) {
      handleSTKCommand(event.command);
    });
window.addEventListener('stkasynccommand',
      function do_handleAsyncSTKCmd(event) {
      handleSTKCommand(event.detail.command);
      });
......

When the Launch Browser Command executed, the handleSTKCommand function is called twice.(icc EventListener and window EventListner).

Comment 6

[Leo]

this is regression issue. when we take a test about this issue, it operated properly in qe1 version. but the problem has occured in qe2 version.    

<QE1 version>
gaia     : 663101b6eb809383e5882d9bc3868a923a57998a
gecko    : 0b83d5e6a54df9bfccdbb5ba662527bf657f5381

<QE2 version>
gaia     : d5f7b2d3f7acc04c35abda8b001262d8912f9dfe
gecko    : 0ab3418551bfcefe1e14633f8f2b0422707a4c2f

Comment 7

[Lukas Blakk [:lsblakk] use ?needinfo]

Is this a dupe of 865985?

Comment 8

[Leo]

No, it is different 865985. although mobile phone has implented launch browser in STK menu, 
mobile phone still shows stk menu. In my opinion, ME should show browser app and stk menu is located in background.
plz confirm attached video clip.

Comment 9

[Wayne Chang [:wchang]]

Triage - Leo+, a regression according to comment 6.

Comment 10

[Leo]

When we analyze icc.js, icc_menu.js, icc_cache.js in setting app's code, there is another problem. Although browser is implemented one time, browser tab is three except for basic tab one. I think that browser is actually implemented twice.   

UICC                 -------------------->            Mobile phone
                proactive command(launch browser)

icc_cache.js receive stkcommand and send stkcommand event to icc.js. And launch browser is implemented. However, icc_menu.js also send stkasynccommand event to icc.js. 
Therefore, this is the reason launch browser is implemented twice.

Comment 11

[Alex Keybl [:akeybl]]

frsela - are you the right assignee for this leo+ bug?

Comment 12

[Fernando R. Sela (no CC, needinfo please) [:frsela]]

(In reply to Alex Keybl [:akeybl] from comment #11)
> frsela - are you the right assignee for this leo+ bug?

I'll check why the message is processed twice

Comment 13

[Leo]

(In reply to Fernando R. Sela [:frsela] from comment #12)
> (In reply to Alex Keybl [:akeybl] from comment #11)
> > frsela - are you the right assignee for this leo+ bug?
> 
> I'll check why the message is processed twice

How does it proceed about checking this problem??
I know that you have plan to code refactoring about simtoolkit code.(bugzilla : 875679)
is it possible to fix this problem, when you proceed code refactoring??

Comment 15

[Leo]

AU_LINUX_GECKO_ICS_STRAWBERRY.01.01.00.019.108

Mozilla build ID: 20130526070207

FFOS os version : 1.1.0.0 

I have attached log file.(Log_enable_debug.7z) you can check more detail log, 
because it is enabled debug value from system app (icc_cache.js file DEBUG variable) and settings app (utils.js DEBUG variable). 
plz confirm this log.

Comment 16

[Leo]

Attachment: log file .(enable debug value from false to true)

Comment 17

[Fernando R. Sela (no CC, needinfo please) [:frsela]]

(In reply to Leo from comment #13)
> (In reply to Fernando R. Sela [:frsela] from comment #12)
> > (In reply to Alex Keybl [:akeybl] from comment #11)
> > > frsela - are you the right assignee for this leo+ bug?
> > 
> > I'll check why the message is processed twice
> 
> How does it proceed about checking this problem??
> I know that you have plan to code refactoring about simtoolkit
> code.(bugzilla : 875679)
> is it possible to fix this problem, when you proceed code refactoring??

Sure, we've it in our radar ;)

Comment 18

[Fernando R. Sela (no CC, needinfo please) [:frsela]]

880352 landed in master, can you retest this issue again?
Thanks in advance

Comment 19

[jongsoo.oh]

With patch, leo works correctly. Thank you.

Comment 20

[Fernando R. Sela (no CC, needinfo please) [:frsela]]

\o/ solved with 880352

Comment 21

[John Ford [:jhford] CET/CEST Berlin Time]

fixed on branch