Get ChannelLayout on OSX with non-deprecated methods.

Get ChannelLayout on OSX with non-deprecated methods.

BUG=139517
TEST=Compiles, runs on a 10.6.8 box.


Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=150722

[DaleCurtis, 2012-07-31 01:36:05]

PTAL guys. Not sure if this fixes the crash and have no 10.8 box to try it out
on. Pinging IRC to find a helper.

[scherkus (not reviewing), 2012-07-31 01:44:21]

http://codereview.chromium.org/10836022/diff/2001/media/audio/mac/audio_outpu...
File media/audio/mac/audio_output_mac.cc (right):

http://codereview.chromium.org/10836022/diff/2001/media/audio/mac/audio_outpu...
media/audio/mac/audio_output_mac.cc:118: property_address.mSelector =
kAudioDevicePropertyPreferredChannelLayout;
...but this change would run into the bug on my 10.6 machine!

If this is a fix for 10.8 then we should keep the old code and use
base::mac::IsOSMountainLionOrLater() w/ the new API

As you note it's a bit of a stab in the dark -- we'd need a 10.8 mac to test

[DaleCurtis, 2012-07-31 01:57:41]

On 2012/07/31 01:44:21, scherkus wrote:
>
http://codereview.chromium.org/10836022/diff/2001/media/audio/mac/audio_outpu...
> File media/audio/mac/audio_output_mac.cc (right):
> 
>
http://codereview.chromium.org/10836022/diff/2001/media/audio/mac/audio_outpu...
> media/audio/mac/audio_output_mac.cc:118: property_address.mSelector =
> kAudioDevicePropertyPreferredChannelLayout;
> ...but this change would run into the bug on my 10.6 machine!
> 
> If this is a fix for 10.8 then we should keep the old code and use
> base::mac::IsOSMountainLionOrLater() w/ the new API
> 
> As you note it's a bit of a stab in the dark -- we'd need a 10.8 mac to test

The original bug http://code.google.com/p/chromium/issues/detail?id=85100 seems
to fix something unrelated to the deprecated functions issue. Hopefully annacc
can shed some light on the original comment. 

FTR, this passes fine on your 10.6.8 box. I'll keep looking for a 10.8 box. It's
also looking like there may be some compounding issues on the reporters side.

[annacc, 2012-08-01 03:27:33]

On 2012/07/31 01:57:41, DaleCurtis wrote:
> On 2012/07/31 01:44:21, scherkus wrote:
> >
>
http://codereview.chromium.org/10836022/diff/2001/media/audio/mac/audio_outpu...
> > File media/audio/mac/audio_output_mac.cc (right):
> > 
> >
>
http://codereview.chromium.org/10836022/diff/2001/media/audio/mac/audio_outpu...
> > media/audio/mac/audio_output_mac.cc:118: property_address.mSelector =
> > kAudioDevicePropertyPreferredChannelLayout;
> > ...but this change would run into the bug on my 10.6 machine!
> > 
> > If this is a fix for 10.8 then we should keep the old code and use
> > base::mac::IsOSMountainLionOrLater() w/ the new API
> > 
> > As you note it's a bit of a stab in the dark -- we'd need a 10.8 mac to test
> 
> The original bug http://code.google.com/p/chromium/issues/detail?id=85100
seems
> to fix something unrelated to the deprecated functions issue. Hopefully annacc
> can shed some light on the original comment. 
> 
> FTR, this passes fine on your 10.6.8 box. I'll keep looking for a 10.8 box.
It's
> also looking like there may be some compounding issues on the reporters side.

I'm all for updating deprecated code, so I'd say go for it once you confirm
testing on 10.8.  I'm sorry I don't recall what about
AudioObjectGetPropertyDataSize() "doesn't work" :(

[scherkus (not reviewing), 2012-08-01 20:59:11]

q: now that we know it's a bad OS installation what do we want to do with this
CL?

[DaleCurtis, 2012-08-01 21:01:44]

On 2012/08/01 20:59:11, scherkus wrote:
> q: now that we know it's a bad OS installation what do we want to do with this
> CL?

I can't find any evidence online of a problem with these methods, so how about
letting it soak on dev channel / canary for a bit?  Easy enough to roll back.

[DaleCurtis, 2012-08-01 21:07:23]

On 2012/08/01 21:01:44, DaleCurtis wrote:
> On 2012/08/01 20:59:11, scherkus wrote:
> > q: now that we know it's a bad OS installation what do we want to do with
this
> > CL?
> 
> I can't find any evidence online of a problem with these methods, so how about
> letting it soak on dev channel / canary for a bit?  Easy enough to roll back.

Per chat discussion w/ scherkus, we'll back burner this until after M22 cut and
then land it for testing.

[DaleCurtis, 2012-08-09 00:00:43]

On 2012/08/01 21:07:23, DaleCurtis wrote:
> On 2012/08/01 21:01:44, DaleCurtis wrote:
> > On 2012/08/01 20:59:11, scherkus wrote:
> > > q: now that we know it's a bad OS installation what do we want to do with
> this
> > > CL?
> > 
> > I can't find any evidence online of a problem with these methods, so how
about
> > letting it soak on dev channel / canary for a bit?  Easy enough to roll
back.
> 
> Per chat discussion w/ scherkus, we'll back burner this until after M22 cut
and
> then land it for testing.

Branch has been cut. scherkus, annacc: LGTY?

[scherkus (not reviewing), 2012-08-09 01:10:51]

lgtm

[commit-bot: I haz the power, 2012-08-09 01:36:02]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/dalecurtis@chromium.org/10836022/4002

[commit-bot: I haz the power, 2012-08-09 04:04:28]

Change committed as 150722