980811 - Modify the test_settings_sim_manager.py

Status: RESOLVED FIXED

(Firefox OS Graveyard :: Gaia::UI Tests, defect)

Description

[Askeing Yen[:askeing][:fyen]]

The "multi-sim" doesn't work with "type" option, ex: "--type=b2g-mutli-sime".
Have to modify it.

And it seems like we can add a check before tap the SIM Manager memu.
If there is no memu item, then skip the tests.
ref: https://github.com/mozilla-b2g/gaia/blob/master/tests/python/gaia-ui-tests/gaiatest/apps/settings/app.py#L138

Comment 1

[Askeing Yen[:askeing][:fyen]]

IMO, this test should be skip by default.
Only if user run with "--type=b2g+multisim" option, then gaiatest will run it.
Any suggestion, Zac?

Comment 2

[Dave Hunt [:davehunt] [he/him] ⌚BST]

(In reply to Askeing Yen[:askeing] from comment #1)
> IMO, this test should be skip by default.
> Only if user run with "--type=b2g+multisim" option, then gaiatest will run
> it.
> Any suggestion, Zac?

I think it's safer to run all tests by default, otherwise there's a risk of tests being overlooked. We could also have a skip in the test if the test is selected but the device does not support multiple sims, similar to our email tests [1]. Note that your type cannot contain a hyphen (-) as this is used as a prefix for excluding types.

[1] https://github.com/mozilla-b2g/gaia/blob/b3758a90b8888e9d95128846b2833b4d9444ef7f/tests/python/gaia-ui-tests/gaiatest/tests/functional/email/test_setup_basic_gmail.py#L14

Comment 3

[Askeing Yen[:askeing][:fyen]]

(In reply to Dave Hunt (:davehunt) from comment #2)
> (In reply to Askeing Yen[:askeing] from comment #1)
> > IMO, this test should be skip by default.
> > Only if user run with "--type=b2g+multisim" option, then gaiatest will run
> > it.
> > Any suggestion, Zac?
> 
> I think it's safer to run all tests by default, otherwise there's a risk of
> tests being overlooked. We could also have a skip in the test if the test is
> selected but the device does not support multiple sims, similar to our email
> tests [1]. Note that your type cannot contain a hyphen (-) as this is used
> as a prefix for excluding types.
> 
> [1]
> https://github.com/mozilla-b2g/gaia/blob/
> b3758a90b8888e9d95128846b2833b4d9444ef7f/tests/python/gaia-ui-tests/gaiatest/
> tests/functional/email/test_setup_basic_gmail.py#L14

agree :)
Then I'll suggest add a check before try to tap the SIM Manager memu item.

Comment 4

[Dave Hunt [:davehunt] [he/him] ⌚BST]

(In reply to Askeing Yen[:askeing] from comment #3)
> (In reply to Dave Hunt (:davehunt) from comment #2)
> > (In reply to Askeing Yen[:askeing] from comment #1)
> Then I'll suggest add a check before try to tap the SIM Manager memu item.

Is it not possible to detect this via a WebAPI?

Comment 5

[Zac C (:zac)]

We could just remove multisim altogether.

run-if = device == "taroko" (replacing taroko with whatever the code is)

however it would mean you have to always be running the dual sim device with two sims but other solutions noted here make this assumption too.

Comment 6

[Zac C (:zac)]

Attachment: github pr

Comment 7

[Eric Chang [:ericcc] [:echang]]

Thanks a lot, I checked on the Fugu, the device ID is "sp7710gaplus_gonk", I will submit another PR.
run-if = device == "sp7710gaplus_gonk"

Comment 8

[Eric Chang [:ericcc] [:echang]]

Attachment: PR 17012 change Fugu device ID

Comment 9

[Askeing Yen[:askeing][:fyen]]

Comment on attachment 8388346 [details] [review]
PR 17012 change Fugu device ID

I don't have fugu device.
But we can check the 'device' by 'adb shell getprop | grep ro.product.device'.
r+ if the device name works correctly.

Comment 10

[Zac C (:zac)]

(In reply to Eric Chang [:ericcc] [:echang] from comment #7)
> Thanks a lot, I checked on the Fugu, the device ID is "sp7710gaplus_gonk", I
> will submit another PR.
> run-if = device == "sp7710gaplus_gonk"

Thanks for checking that for us!

Comment 11

[Zac C (:zac)]

Comment on attachment 8388346 [details] [review]
PR 17012 change Fugu device ID

r+

Comment 12

[Zac C (:zac)]

Merged:
https://github.com/mozilla-b2g/gaia/commit/11b44326c3a9a513c3d930ac7c1afac52d7c8b52

Comment 13

[Eric Chang [:ericcc] [:echang]]

(In reply to Dave Hunt (:davehunt) from comment #4)
> (In reply to Askeing Yen[:askeing] from comment #3)
> > (In reply to Dave Hunt (:davehunt) from comment #2)
> > > (In reply to Askeing Yen[:askeing] from comment #1)
> > Then I'll suggest add a check before try to tap the SIM Manager memu item.
> 
> Is it not possible to detect this via a WebAPI?
Good idea, I checked that on https://wiki.mozilla.org/WebAPI/WebIccManager/Multi-SIM
navigator.mozMobileConnections[0] we can check the length of this array to see the number of the slots of a device, I tried that on one of my packaged app. And navigator.mozIccManager.iccIds.length, this is the number of sims inserted.