1119640 - [FFOS2.0][Woodduck][STK]Setup Event executed not ok:Location status Event.

Status: RESOLVED FIXED

(Firefox OS Graveyard :: General, defect, P1)

Description

[pengfei.huang]

Dear Sean Lee & Josh,
  Our VAL guy(another tester) find a strange performance on test version. There are two same envelops(EVENT DOWNLAOD) when event(MT call, Call connected, Call disconnected) received. The wrong envelop come out when we click confirm button of DISPLAY TEXT dialog.  
  Besides, he add a test about LOCATION STATUS. When network(mcc or mnc) change, receive three envelops(EVENT DOWNLOAD). The location event belong to EVENT DOWNLOAD.
  Need your help again. Many thanks. :)

Comment 1

[pengfei.huang]

Hi all,
  This bug was fix at bug1088611, see comment41.Thanks.

Comment 2

[Sean Lee [:seanlee][:weilonge]]

Hello Pengfei,

I will start the review process.
Thanks for your help.

Comment 3

[Sean Lee [:seanlee][:weilonge]]

see also: bug 1088611 comment 41

Comment 4

[Sean Lee [:seanlee][:weilonge]]

Attachment: PR for master

Hello Fernando,

Here is my PR for this bug. Please help to review it. Thank you.

Comment 5

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

Comment on attachment 8546420 [details] [review]
PR for master

LGTM. Thank you for taking this :)

Comment 6

[Sean Lee [:seanlee][:weilonge]]

Thank you! Fernando!

try-server: https://treeherder.mozilla.org/ui/#/jobs?repo=gaia-try&revision=9ee93634c34e

landed on master: https://github.com/mozilla-b2g/gaia/commit/8b9d7d5ea0a4e9c18db00932dd33b360e15ad7cb

landed on v2.0m : https://github.com/mozilla-b2g/gaia/commit/471e5fa57abaec153c7b0627212b23351ddf48e0

Comment 7

[Kai-Zhen Li [:seinlin][:kli]]

Per comment 6.

Comment 8

[Sean Lee [:seanlee][:weilonge]]

Comment on attachment 8546420 [details] [review]
PR for master

[Approval Request Comment]
[Bug caused by] (feature/regressing bug #):
[User impact] if declined:
The behavior of following STK setup events are incorrect.
STK_EVENT_TYPE_LOCATION_STATUS
STK_EVENT_TYPE_IDLE_SCREEN_AVAILABLE
STK_EVENT_TYPE_BROWSER_TERMINATION

Telecom carrier will not accept this if we do not fix it.

[Testing completed]:
Partner tested in their LAB.

[Risk to taking this patch] (and alternatives if risky):
The code change is only for STK and very minor.

[String changes made]:
None

Comment 9

[GitHub Autolander Bot]

Attachment: [PullReq] weilonge:seanlee/STK/v2.1/Bug1119640 to mozilla-b2g:v2.1

Comment 10

[GitHub Autolander Bot]

Attachment: [PullReq] weilonge:seanlee/STK/v2.2/Bug1119640 to mozilla-b2g:v2.2

Comment 11

[Ryan VanderMeulen [:RyanVM]]

v2.2: https://github.com/mozilla-b2g/gaia/commit/c7018d2c766f56945b28fd50e0a211d74b382e7a
v2.1: https://github.com/mozilla-b2g/gaia/commit/88084bc7ef5ba6627dd09c774ef2f7fa96cbed71

Comment 12

[jinghua.xing]

Sean：

As the protocol says "As a result of sending this command to the SIM, the ME shall remove the idle screen available event from its current event list. This is in order for the ME to report this event only once after the event has been requested bythe SIM." 

So the event listener for IDLE_SCREEN_AVAILABLE should be removed once the event happened. But as your modify,the listener can only be removed when we receive another EVENT LIST command.

What do you think about this?

Comment 13

[jinghua.xing]

And besides, I think your patch is want to remove the event listener when receive new EVENT LIST command. But when I test the case STK_EVENT_TYPE_IDLE_SCREEN_AVAILABLE in v2.1, I found the patch you merged seems do not work. And if we receive two EVENT LIST command of STK_EVENT_TYPE_IDLE_SCREEN_AVAILABLE, when the event happens, we still receive two same envelops(EVENT DOWNLAOD).

What do you think about this?

Comment 14

[Sean Lee [:seanlee][:weilonge]]

I create a new bug 1132363 to discuss the implementation of IDLE_SCREEN_AVAILABLE.

Comment 15

[Sean Lee [:seanlee][:weilonge]]

Hi Jinghua,

Sorry to interrupt this discussion.
Let's move to bug 1132363 and focus on the implementation of IDLE_SCREEN_AVAILABLE .