928166 - [B2G][Twitter]Twitter encounters a crash when attempting to add a photo to a tweet via Camera or Gallery.

Status: RESOLVED DUPLICATE of bug 923310

(Core :: DOM: Core & HTML, defect)

Description

[Peter Bylenga [:PBylenga]]

Attachment: Logcat of repro steps and crash.

Description:
A crash occurs when user attempts to add a photo to a new tweet either from a photo just taken with the camera or Gallery.  If user elects to use the camera to take a new picture the crash occurs when user confirms the photo taken by tapping on [Select] button.  If user elects gallery option the crash occurs after user confirms the Crop Edit after selecting an image to attach.

Repro Steps:
1) Updated Buri to Build ID: 20131017004001
2) Install Twitter via everything.me
3) Login to Twitter
4) Create a New Tweet in Twitter
5) Select Camera Option in lower left and take a photo with Camera
6) Choose to Select the photo just taken and observe

Actual:
Twitter app encounters a Crash.

Expected:
No crash is encountered and user is able to add a photo to a tweet.

Environmental Variables
Device: Buri v1.2 Mozilla RiL
Build ID: 20131017004001
Gecko: http://hg.mozilla.org/releases/mozilla-aurora/rev/7ca5d1e81d37
Gaia: dbcc171eae6b4d9d168a48291d5ea54c7580561a
Platform Version: 26.0a2
Firmware Version: US_20130912

Notes:
Repro frequency: 7/7, 100%
See attached: logcat of repro steps + crash


Crash Reports:
Crash ID: bp-5a74dd49-edbe-4dee-859b-f2b212131017

Comment 1

[Peter Bylenga [:PBylenga]]

Environmental Variables
Device: Leo v 1.1.0 COM RIL
Build ID: 20131008041201
Gecko: http://hg.mozilla.org/releases/mozilla-b2g18/rev/c630289d6388
Gaia: e5adcf3c23b6709a51087539bc4cdb67683ca4a0
Platform Version: 18.1
RIL Version: 01.01.00.019.246

Comment 2

[Jason Smith [:jsmith]]

Looking at the changeset the crash stack points to, I'm wondering if this is related to the changes that landed in bug 894840. We can confirm that if the regression window shows that the last good build was on 9/3/2013 & first bad was 9/4/2013.

Comment 3

[Natalya Kot [:nkot]]

i guess you are right, it got broken on that day

regression range:

-last working-
BuildID: 20130903040204
Gaia: c6d58088f207829d96e7bb33ddacf990e90752fb
Gecko: d54e0cce6c17
Version: 26.0a1

-first broken-
BuildID: 20130904040205
Gaia: b6c5bf1d24230bfed4a8f680e625fa2175001f82
Gecko: 7ff96bd19c1c
Version: 26.0a1

Comment 4

[Jonathan Watt [:jwatt]]

According to comment 3, that mean the regression happened in this range:

https://hg.mozilla.org/integration/mozilla-inbound/pushloghtml?fromchange=d54e0cce6c17&tochange=7ff96bd19c1c

Comment 5

[Jonathan Watt [:jwatt]]

Of the five changesets that I pushed in that range, the only one that seems capable of causing this bug is:

https://hg.mozilla.org/integration/mozilla-inbound/rev/be3663c7c734

Was this fixed by bug 923310? I suspect it was.

Comment 6

[Jason Smith [:jsmith]]

(In reply to Jonathan Watt [:jwatt] from comment #5)
> Of the five changesets that I pushed in that range, the only one that seems
> capable of causing this bug is:
> 
> https://hg.mozilla.org/integration/mozilla-inbound/rev/be3663c7c734
> 
> Was this fixed by bug 923310? I suspect it was.

Looking at the line of code baku mentions, it looks like that's right. I'll dupe & switch bug 923310 to koi+.