Change TraceDataEndPoint::ReceiveTraceChunk() to accept unique_ptr<string>

Change TraceDataEndPoint::ReceiveTraceChunk() to accept unique_ptr<string>

Committed: https://crrev.com/e9307936cf37cc30c5a11e37a9085c9d447b84a9
Cr-Commit-Position: refs/heads/master@{#405913}

[caseq, 2016-07-14 21:50:48]

caseq@chromium.org changed reviewers:
+ oysteine@chromium.org

[caseq, 2016-07-15 01:23:30]

Changed to use unique_ptr<string> instead of string&& -- looks a bit more messy
now, but I still think it's the way to go considering how often we send it
across the threads.

[oystein (OOO til 10th of July), 2016-07-15 16:56:27]

Nice, lgtm w/ one comment/question!

https://codereview.chromium.org/2143033004/diff/20001/content/browser/tracing...
File content/browser/tracing/tracing_controller_impl_data_sinks.cc (right):

https://codereview.chromium.org/2143033004/diff/20001/content/browser/tracing...
content/browser/tracing/tracing_controller_impl_data_sinks.cc:241:
stream_->next_in = reinterpret_cast<unsigned char*>(&*chunk->begin());
Just curious: What's the reason for &*chunk->begin() instead of *chunk->data()?

[caseq, 2016-07-15 22:14:20]

On 2016/07/15 16:56:27, oystein wrote:
> Nice, lgtm w/ one comment/question!
> 
>
https://codereview.chromium.org/2143033004/diff/20001/content/browser/tracing...
> File content/browser/tracing/tracing_controller_impl_data_sinks.cc (right):
> 
>
https://codereview.chromium.org/2143033004/diff/20001/content/browser/tracing...
> content/browser/tracing/tracing_controller_impl_data_sinks.cc:241:
> stream_->next_in = reinterpret_cast<unsigned char*>(&*chunk->begin());
> Just curious: What's the reason for &*chunk->begin() instead of
*chunk->data()?

string.data() is const and we need non-const char*, so we would also need a
const_cast<> in addition to reinterpret_cast<> there (it was previously done
with a C-style cast that allows both case in one shot, but I think it's
forbidden by the style guide).

[caseq, 2016-07-15 22:28:40]

Description was changed from

==========
Change TraceDataEndPoint::ReceiveTraceChunk() to accept rvalue ref
==========

to

==========
Change TraceDataEndPoint::ReceiveTraceChunk() to accept unique_ptr<string>
==========

[caseq, 2016-07-16 00:11:06]

The CQ bit was checked by caseq@chromium.org

[commit-bot: I haz the power, 2016-07-16 00:11:24]

CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-07-16 00:55:52]

Committed patchset #2 (id:20001)

[commit-bot: I haz the power, 2016-07-16 00:55:58]

CQ bit was unchecked.

[commit-bot: I haz the power, 2016-07-16 00:57:37]

Description was changed from

==========
Change TraceDataEndPoint::ReceiveTraceChunk() to accept unique_ptr<string>
==========

to

==========
Change TraceDataEndPoint::ReceiveTraceChunk() to accept unique_ptr<string>

Committed: https://crrev.com/e9307936cf37cc30c5a11e37a9085c9d447b84a9
Cr-Commit-Position: refs/heads/master@{#405913}
==========

[commit-bot: I haz the power, 2016-07-16 00:57:38]

Patchset 2 (id:??) landed as
https://crrev.com/e9307936cf37cc30c5a11e37a9085c9d447b84a9
Cr-Commit-Position: refs/heads/master@{#405913}