Remove unnecessary condition for audioSourceProvider

Remove unnecessary condition for audioSourceProvider

Change audioSourceProvider to return reference so that it will never
be null. This allows to remove the unnecessary conditions.

Committed: https://crrev.com/969bf4fbb54129362114ad337ebe22eb5f200506
git-svn-id: svn://svn.chromium.org/blink/trunk@199518 bbb929c8-8fbe-4397-9dbb-9b2b20218538

[Srirama, 2015-07-27 14:06:18]

srirama.m@samsung.com changed reviewers:
+ philipj@opera.com

[Srirama, 2015-07-27 14:06:18]

PTAL

[philipj_slow, 2015-07-27 14:12:30]

There's also a redundant null check in modules/webaudio:
https://code.google.com/p/chromium/codesearch#chromium/src/third_party/WebKit...

Can you instead let this getter return a reference so that it's obviously never
null?

[Srirama, 2015-07-27 14:42:18]

On 2015/07/27 14:12:30, philipj wrote:
> There's also a redundant null check in modules/webaudio:
>
https://code.google.com/p/chromium/codesearch#chromium/src/third_party/WebKit...
> 
> Can you instead let this getter return a reference so that it's obviously
never
> null?

Done.

[philipj_slow, 2015-07-27 15:29:29]

philipj@opera.com changed reviewers:
+ rtoy@chromium.org

[philipj_slow, 2015-07-27 15:29:29]

HTMLMediaElement LGTM

rtoy@, can you please review modules/webaudio?

[Raymond Toy, 2015-07-27 16:32:12]

On 2015/07/27 15:29:29, philipj wrote:
> HTMLMediaElement LGTM
> 
> rtoy@, can you please review modules/webaudio?

lgtm for modules/webaudio.

This CL doesn't seem to show it, but how is m_audioSourceProvider guaranteed not
to be null now?

[Srirama, 2015-07-27 16:46:07]

On 2015/07/27 16:32:12, Raymond Toy wrote:
> On 2015/07/27 15:29:29, philipj wrote:
> > HTMLMediaElement LGTM
> > 
> > rtoy@, can you please review modules/webaudio?
> 
> lgtm for modules/webaudio.
> 
> This CL doesn't seem to show it, but how is m_audioSourceProvider guaranteed
not
> to be null now?

Thanks Raymond for the review.
Infact it is not null even before this CL as we are passing address of the
object (m_audioSourceProvider) which is gauranteed to be non-null. So I made it
passbyreference and removed the null checks everywhere.

[Raymond Toy, 2015-07-27 16:47:22]

On 2015/07/27 16:46:07, Srirama wrote:
> On 2015/07/27 16:32:12, Raymond Toy wrote:
> > On 2015/07/27 15:29:29, philipj wrote:
> > > HTMLMediaElement LGTM
> > > 
> > > rtoy@, can you please review modules/webaudio?
> > 
> > lgtm for modules/webaudio.
> > 
> > This CL doesn't seem to show it, but how is m_audioSourceProvider guaranteed
> not
> > to be null now?
> 
> Thanks Raymond for the review.
> Infact it is not null even before this CL as we are passing address of the
> object (m_audioSourceProvider) which is gauranteed to be non-null. So I made
it
> passbyreference and removed the null checks everywhere.

Hmm. I'm pretty sure there was some weird case where (at least on
modules/webaudio) that it was actually null.

[Raymond Toy, 2015-07-27 17:03:26]

On 2015/07/27 16:47:22, Raymond Toy wrote:
> On 2015/07/27 16:46:07, Srirama wrote:
> > On 2015/07/27 16:32:12, Raymond Toy wrote:
> > > On 2015/07/27 15:29:29, philipj wrote:
> > > > HTMLMediaElement LGTM
> > > > 
> > > > rtoy@, can you please review modules/webaudio?
> > > 
> > > lgtm for modules/webaudio.
> > > 
> > > This CL doesn't seem to show it, but how is m_audioSourceProvider
guaranteed
> > not
> > > to be null now?
> > 
> > Thanks Raymond for the review.
> > Infact it is not null even before this CL as we are passing address of the
> > object (m_audioSourceProvider) which is gauranteed to be non-null. So I made
> it
> > passbyreference and removed the null checks everywhere.
> 
> Hmm. I'm pretty sure there was some weird case where (at least on
> modules/webaudio) that it was actually null.

I might be mistaken. I can't find the actual case, so I'm fine if you want to
land this now.

[Srirama, 2015-07-27 17:07:55]

On 2015/07/27 17:03:26, Raymond Toy wrote:
> On 2015/07/27 16:47:22, Raymond Toy wrote:
> > On 2015/07/27 16:46:07, Srirama wrote:
> > > On 2015/07/27 16:32:12, Raymond Toy wrote:
> > > > On 2015/07/27 15:29:29, philipj wrote:
> > > > > HTMLMediaElement LGTM
> > > > > 
> > > > > rtoy@, can you please review modules/webaudio?
> > > > 
> > > > lgtm for modules/webaudio.
> > > > 
> > > > This CL doesn't seem to show it, but how is m_audioSourceProvider
> guaranteed
> > > not
> > > > to be null now?
> > > 
> > > Thanks Raymond for the review.
> > > Infact it is not null even before this CL as we are passing address of the
> > > object (m_audioSourceProvider) which is gauranteed to be non-null. So I
made
> > it
> > > passbyreference and removed the null checks everywhere.
> > 
> > Hmm. I'm pretty sure there was some weird case where (at least on
> > modules/webaudio) that it was actually null.
> 
> I might be mistaken. I can't find the actual case, so I'm fine if you want to
> land this now.

Just to clarify, before this CL https://codereview.chromium.org/1055503002/,
audioSourceProvider() in HTMLMediaElement used to return null if the m_player is
null. Here m_player is the intermediate layer(MediaPlayer) which is removed in
the above mentioned CL. So after the above CL it will never be null.

[Raymond Toy, 2015-07-27 17:16:51]

On 2015/07/27 17:07:55, Srirama wrote:
> On 2015/07/27 17:03:26, Raymond Toy wrote:
> > On 2015/07/27 16:47:22, Raymond Toy wrote:
> > > On 2015/07/27 16:46:07, Srirama wrote:
> > > > On 2015/07/27 16:32:12, Raymond Toy wrote:
> > > > > On 2015/07/27 15:29:29, philipj wrote:
> > > > > > HTMLMediaElement LGTM
> > > > > > 
> > > > > > rtoy@, can you please review modules/webaudio?
> > > > > 
> > > > > lgtm for modules/webaudio.
> > > > > 
> > > > > This CL doesn't seem to show it, but how is m_audioSourceProvider
> > guaranteed
> > > > not
> > > > > to be null now?
> > > > 
> > > > Thanks Raymond for the review.
> > > > Infact it is not null even before this CL as we are passing address of
the
> > > > object (m_audioSourceProvider) which is gauranteed to be non-null. So I
> made
> > > it
> > > > passbyreference and removed the null checks everywhere.
> > > 
> > > Hmm. I'm pretty sure there was some weird case where (at least on
> > > modules/webaudio) that it was actually null.
> > 
> > I might be mistaken. I can't find the actual case, so I'm fine if you want
to
> > land this now.
> 
> Just to clarify, before this CL https://codereview.chromium.org/1055503002/,
> audioSourceProvider() in HTMLMediaElement used to return null if the m_player
is
> null. Here m_player is the intermediate layer(MediaPlayer) which is removed in
> the above mentioned CL. So after the above CL it will never be null.

Thanks for clarifying this! I understand now why it's safe.

[Srirama, 2015-07-27 17:20:55]

The CQ bit was checked by srirama.m@samsung.com

[commit-bot: I haz the power, 2015-07-27 17:21:04]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1261643002/20001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1261643002/20001

[commit-bot: I haz the power, 2015-07-27 17:23:55]

Committed patchset #2 (id:20001) as
https://src.chromium.org/viewvc/blink?view=rev&revision=199518

[commit-bot: I haz the power, 2015-09-23 11:48:51]

Patchset 2 (id:??) landed as
https://crrev.com/969bf4fbb54129362114ad337ebe22eb5f200506