983054 - [Dialer] Cannot make a outgoing call from suggestion bar

Status: RESOLVED FIXED

(Firefox OS Graveyard :: Gaia::Dialer, defect)

Description

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

Buri
Gaia      e61dc0019d9d6135d88ba15153c37f73a952567e
Gecko     https://hg.mozilla.org/mozilla-central/rev/a10d0ba50f64
BuildID   20140312160210
Version   30.0a1
ro.build.version.incremental=eng.archermind.20131114.105818
ro.build.date=Thu Nov 14 10:58:33 CST 2013

### STR
1. Create a contact
2. Launch Dialer
3. Enter the first 4~5 digit of full cell phone number
4. Tap the contact on suggestion bar

### Expect
1. make a outgoing call

### Actual
1. no outgoing call

Comment 1

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

Attachment: video.3gp

Comment 2

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

last success

Buri
Gaia      a351fe62c11737c722ad33aaff438f6ccd00bd4a
Gecko     https://hg.mozilla.org/mozilla-central/rev/41d962d23e81
BuildID   20140311040203
Version   30.0a1
ro.build.version.incremental=eng.archermind.20131114.105818
ro.build.date=Thu Nov 14 10:58:33 CST 2013

Comment 3

[Robert Chira [:RobertC]]

I xfailed the test for now.
From what I noticed after tapping on the suggested contact (which completes the phone number) you can call the number only by tapping on the call button.

Comment 4

[Zac C (:zac)]

This looks suspiciously like new expected behaviour.

I am ni? Rik to confirm.

Comment 5

[Anthony Ricaud (:rik)]

This is a regression of bug 946866. Although it's a regression, it's not that bad, people can still call.

Carrie: Look, another case we forgot! What should happen in this case? We can talk about it during our work week, it's not urgent.

Comment 6

[Anthony Ricaud (:rik)]

The regression is because we changed KeypadManager.makeCall to KeypadManager.fetchLastCalled. suggestion_bar.js calls KeypadManager.makeCall() and fails.

Comment 7

[Jason Smith [:jsmith]]

(In reply to Anthony Ricaud (:rik) from comment #5)
> This is a regression of bug 946866. Although it's a regression, it's not
> that bad, people can still call.
> 
> Carrie: Look, another case we forgot! What should happen in this case? We
> can talk about it during our work week, it's not urgent.

I would call this bad because it caused a test to go down, which never a good thing. I wish we knew about this earlier, as we should have backed out immediately.

Comment 8

[Anthony Ricaud (:rik)]

I meant it's not bad in terms of how much user impact it has. I would have backed out if it had user impact. But we are late on DSDS implementation and a lot of code landed since bug 946866. We will fix this for 1.4 (most likely next week once we have UX feedback), I've added it to the meta bug for DSDS.

I agree that we should have thought about this when planning the DSDS work.

Comment 9

[Doug Sherk (:drs) (inactive)]

Can we just do a really dumb CallButton.getCardIndex() call to fix this for now, at least in the non-always ask case?

Comment 10

[Anthony Ricaud (:rik)]

We have a lot of work on our hands and I prefer to not spend it on fixing bugs twice.

Comment 11

[Carrie Wang [New account: [e-mail]]]

Hi Anthony, 
sure thing. keep ni? here. Will come bacl to this issue next week. Thanks!

Comment 12

[Anthony Ricaud (:rik)]

Discussed this with Carrie in person.

We're gonna go with this behaviour:
- One SIM in device: Autocomplete and call directly (same behaviour that the regression is highlighting)
- Multi SIMs in device: just autocomplete, do not call. The user will use the call button to choose how to call.

Comment 13

[Carrie Wang [New account: [e-mail]]]

Yes, you got it. Thanks! ;-]

Comment 14

[Doug Sherk (:drs) (inactive)]

We ran into an additional issue when implementing this. In the multi-SIM case, if you tap on a suggested phone number, it will fill the active phone number with that contact's phone number, but the suggestion remains and doesn't do anything if you tap on it again. Carrie and I talked in person and we decided to just hide the suggestion bar if there's an exact match and we're multi-SIM.

Comment 15

[Doug Sherk (:drs) (inactive)]

Attachment: Link to Github pull-request: https://github.com/mozilla-b2g/gaia/pull/17287

Comment 16

[Joe Cheng [:jcheng] (please needinfo)]

triage: 1.4+ regression

Comment 17

[Anthony Ricaud (:rik)]

Comment on attachment 8392824 [details] [review]
Link to Github pull-request: https://github.com/mozilla-b2g/gaia/pull/17287

Looks like we're missing a test to check that we call hideOverlay().

Also consider cleaning the files for JSHint and removing requireApp calls. Not mandatory but helpful to clean files as we touch them.

Comment 18

[Doug Sherk (:drs) (inactive)]

Comment on attachment 8392824 [details] [review]
Link to Github pull-request: https://github.com/mozilla-b2g/gaia/pull/17287

Updated PR with code review fixes.

Comment 19

[Doug Sherk (:drs) (inactive)]

https://github.com/mozilla-b2g/gaia/commit/1f74f96af078f3e69d208b45a770bed1be6798a1

Comment 20

[Ryan VanderMeulen [:RyanVM]]

v1.4: 35bf1832f7e8cb96e137d5b35b39d234c9e8cde5