As per device certification requirements we need to support the *#06# MMI code to get the device IMEI code.

Is this a P1 blockers or is it a nice-to-have?

(In reply to Andrew Overholt [:overholt] from comment #1) > Is this a P1 blockers or is it a nice-to-have? Blocker because device certification requirement.

I consider this task as LOE:S once bug 793186 lands.

Not much to explain. This patch allows the retrieval of the device's IMEI via the *#06# MMI code.

Sorry for the spam Marshall, I didn't qref when added the 'error' test.

Comment on attachment 669437 [details] [diff] [review] Part 1: RIL Review of attachment 669437 [details] [diff] [review]: ----------------------------------------------------------------- Looks good! missed one comment.. ::: dom/system/gonk/RILContentHelper.js @@ +716,5 @@ > case "RIL:SendMMI:Return:OK": > case "RIL:CancelMMI:Return:OK": > request = this.takeRequest(msg.json.requestId); > if (request) { > + Services.DOMRequest.fireSuccess(request, msg.json.result); does this change the semantics for any of the other Send/CancelMMI results? (are we testing these DOM results anywhere?)

Comment on attachment 669441 [details] [diff] [review] Part 2: Test Review of attachment 669441 [details] [diff] [review]: ----------------------------------------------------------------- It might be good to test the actual IMEI of the emulator at some point, but these tests are a good start. Maybe a file a follow up?

Thanks for the review Marshall! :) (In reply to Marshall Culpepper [:marshall_law] from comment #7) > Comment on attachment 669437 [details] [diff] [review] > Part 1: RIL > > Review of attachment 669437 [details] [diff] [review]: > ----------------------------------------------------------------- > > Looks good! missed one comment.. > > ::: dom/system/gonk/RILContentHelper.js > @@ +716,5 @@ > > case "RIL:SendMMI:Return:OK": > > case "RIL:CancelMMI:Return:OK": > > request = this.takeRequest(msg.json.requestId); > > if (request) { > > + Services.DOMRequest.fireSuccess(request, msg.json.result); > > does this change the semantics for any of the other Send/CancelMMI results? > (are we testing these DOM results anywhere?) No, sending 'msg.json' instead of 'msg.json.result' was actually a bug. And no, I am afraid that we are not testing these DOM results anywhere yet. (In reply to Marshall Culpepper [:marshall_law] from comment #8) > Comment on attachment 669441 [details] [diff] [review] > Part 2: Test > > Review of attachment 669441 [details] [diff] [review]: > ----------------------------------------------------------------- > > It might be good to test the actual IMEI of the emulator at some point, but > these tests are a good start. Maybe a file a follow up? Indeed, I have a patch ready with marionette tests (where we also check the DOMRequest.result), but I couldn't run it cause the emulator is not working for me :\. I'll file a follow-up bug for the marionette tests.

https://hg.mozilla.org/integration/mozilla-inbound/rev/d8cf2a31d1b9 https://hg.mozilla.org/integration/mozilla-inbound/rev/5bff19e795d6

https://hg.mozilla.org/mozilla-central/rev/d8cf2a31d1b9 https://hg.mozilla.org/mozilla-central/rev/5bff19e795d6

https://hg.mozilla.org/releases/mozilla-aurora/rev/ae52f5c31f5c https://hg.mozilla.org/releases/mozilla-aurora/rev/8ab4d2cd6120