[system-settings] shouldn't hardcode the modem number

[Sebastien Bacher]

We currently get the imei information from the device 0, that happens to work but we should probalby be smarter than that and deal with the cases where the modem is not the first device in the list

[Tony Espy (awe)]

@Seb

Can you please explain what you mean? The IMEI is available directly from ofono as the Serial property. It's also available via the Android property 'ro.serialno'. See bug #1210502 for more details.

[Sebastien Bacher (seb128)]

> Can you please explain what you mean? The IMEI is available directly from ofono as the Serial property.

Hey Tony,

The qtsystems api takes a device number, as an input parameter, and checks the properties on this device. We currently hardcode "0" (the first modem available), I'm not sure if some devices have > 1 modem and if we need to be smarted about which one we should query for the information...

I think the current approach should be good enough in most cases, but still it seems like an assumption that might get wrong at some point and bite us back

[Tony Espy (awe)]

Ah, OK. I'd assumed that most of the settings plugins were using DBus to directly access services ( especially ofono ).

I think in the 14.04 cycle we really need to do a closer look at qtsystems and how it implements certain features.

I guess for now, the assumption that a single modem exists is probably OK for 13.10, given that we allocate time to investigate / review our current approach(es).

[Sebastien Bacher (seb128)]

setting back as triaged, we could be smarter about handling > 1 modem in a device in the futur, but that's not a priority for v1

[Pat McGowan (pat-mcgowan)]

Now that we have qofono this is fairly easy to switch over. (for imei, it does not have the serial number) The design will need updates to accommodate 2 modems

[Sebastien Bacher (seb128)]

Jonas said he would look at this while adding support for dual sim, thanks!

[Launchpad Janitor (janitor)]

This bug was fixed in the package ubuntu-system-settings - 0.3+15.04.20141210.1-0ubuntu1

---------------
ubuntu-system-settings (0.3+15.04.20141210.1-0ubuntu1) vivid; urgency=low

  [ jonas-drange ]
  * [cellular] Use a repeater to lay out two separate ItemSelectors for
    modems, making sure it works well between connMan power cycles (LP:
    #1377455)
  * [about] properly display IMEI values for no devices ("None"), one
    device and multiple devices. Remove test now redundant. (LP:
    #1205294)

  [ Iain Lane ]
  * Use changed_properties instead of invalidated_properties. timedated
    no longer sends the latter. We also now get the new value for free
    along with the change, so save it instead of doing another roundtrip
    over the bus to ask for it again. (LP: #1400353)

  [ Mathieu Trudel-Lapierre ]
  * Re-enable all device types.

  [ CI Train Bot ]
  * Resync trunk

  [ Ken VanDine ]
  * [brightness] ensure properties are not use undefined (LP: #1385467)

  [ Chris Gagnon ]
  * wait for selectTimeZoneField in autopilot test (LP: #1396335)
 -- Ubuntu daily release <email address hidden> Wed, 10 Dec 2014 16:01:14 +0000

[Launchpad Janitor (janitor)]

This bug was fixed in the package ubuntu-system-settings - 0.3+15.04.20150204~rtm-0ubuntu1

---------------
ubuntu-system-settings (0.3+15.04.20150204~rtm-0ubuntu1) 14.09; urgency=medium

  [ Ken VanDine ]
  * Revoke any cached authorizations before trying to change password
    mode. This makes sure that we authenticate the user again even if we
    just did so (for UX consistency).
  * [about] properly display IMEI values for no devices ("None"), one
    device and multiple devices. Remove test now redundant. (LP:
    #1205294)
  * [security/privacy] add connectivity binding and let user unlock a
    given SIM (LP: #1378883)
 -- Ubuntu daily release <email address hidden> Wed, 04 Feb 2015 17:38:10 +0000