Use MessageLoopProxy instead of MessageLoop in base::Timer.

Use ThreadTaskRunnerHandle in base::Timer.

Now ThreadTaskRunnerHandle is used instead of MessageLoop::current(). This allows to use Timer classes on threads that don't have MessageLoop but have a different task runner.


Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=136667

[willchan no longer on Chromium, 2012-04-24 23:18:35]

Can you explain this a bit further? The description doesn't make sense to
me. If there's no MessageLoop, MessageLoopProxy::current() will also return
NULL...right?

On Tue, Apr 24, 2012 at 4:01 PM, <sergeyu@chromium.org> wrote:

> Reviewers: willchan,
>
> Description:
> Use MessageLoopProxy instead of MessageLoop in base::Timer.
>
> This allows to use Timer classes on threads that don't have MessageLoop,
> but have MessageLoopProxy.
>
>
> Please review this at
http://codereview.chromium.**org/10213012/<http://codereview.chromium.org/102...
>
> SVN Base:
svn://svn.chromium.org/chrome/**trunk/src<http://svn.chromium.org/chrome/trunk/src>
>
> Affected files:
>  M base/timer.cc
>
>
> Index: base/timer.cc
> diff --git a/base/timer.cc b/base/timer.cc
> index e51660c89794ef9d379bbcdf58f0a7**03ec674345..**
> 3ae04d08c73d62987800ae353c93ec**6c3148d6fa 100644
> --- a/base/timer.cc
> +++ b/base/timer.cc
> @@ -5,7 +5,7 @@
>  #include "base/timer.h"
>
>  #include "base/logging.h"
> -#include "base/message_loop.h"
> +#include "base/message_loop_proxy.h"
>
>  namespace base {
>
> @@ -129,7 +129,8 @@ void Timer::PostNewScheduledTask(**TimeDelta delay) {
>   DCHECK(scheduled_task_ == NULL);
>   is_running_ = true;
>   scheduled_task_ = new BaseTimerTaskInternal(this);
> -  MessageLoop::current()->**PostDelayedTask(posted_from_,
> +  base::MessageLoopProxy::**current()->PostDelayedTask(
> +      posted_from_,
>       base::Bind(&**BaseTimerTaskInternal::Run,
> base::Owned(scheduled_task_)),
>       delay);
>   scheduled_run_time_ = desired_run_time_ = TimeTicks::Now() + delay;
>
>
>

[willchan no longer on Chromium, 2012-04-24 23:19:51]

NM, I see your other changelist now :)

On 2012/04/24 23:18:35, willchan wrote:
> Can you explain this a bit further? The description doesn't make sense to
> me. If there's no MessageLoop, MessageLoopProxy::current() will also return
> NULL...right?
> 
> On Tue, Apr 24, 2012 at 4:01 PM, <mailto:sergeyu@chromium.org> wrote:
> 
> > Reviewers: willchan,
> >
> > Description:
> > Use MessageLoopProxy instead of MessageLoop in base::Timer.
> >
> > This allows to use Timer classes on threads that don't have MessageLoop,
> > but have MessageLoopProxy.
> >
> >
> > Please review this at
>
http://codereview.chromium.**org/10213012/%3Chttp://codereview.chromium.org/1...>
> >
> > SVN Base:
>
svn://svn.chromium.org/chrome/**trunk/src<http://svn.chromium.org/chrome/trunk/src>
> >
> > Affected files:
> >  M base/timer.cc
> >
> >
> > Index: base/timer.cc
> > diff --git a/base/timer.cc b/base/timer.cc
> > index e51660c89794ef9d379bbcdf58f0a7**03ec674345..**
> > 3ae04d08c73d62987800ae353c93ec**6c3148d6fa 100644
> > --- a/base/timer.cc
> > +++ b/base/timer.cc
> > @@ -5,7 +5,7 @@
> >  #include "base/timer.h"
> >
> >  #include "base/logging.h"
> > -#include "base/message_loop.h"
> > +#include "base/message_loop_proxy.h"
> >
> >  namespace base {
> >
> > @@ -129,7 +129,8 @@ void Timer::PostNewScheduledTask(**TimeDelta delay) {
> >   DCHECK(scheduled_task_ == NULL);
> >   is_running_ = true;
> >   scheduled_task_ = new BaseTimerTaskInternal(this);
> > -  MessageLoop::current()->**PostDelayedTask(posted_from_,
> > +  base::MessageLoopProxy::**current()->PostDelayedTask(
> > +      posted_from_,
> >       base::Bind(&**BaseTimerTaskInternal::Run,
> > base::Owned(scheduled_task_)),
> >       delay);
> >   scheduled_run_time_ = desired_run_time_ = TimeTicks::Now() + delay;
> >
> >
> >

[Sergey Ulanov, 2012-05-09 19:08:27]

Updated this CL to use ThreadTaskRunnerHandle.

[willchan no longer on Chromium, 2012-05-11 18:51:07]

lgtm

[commit-bot: I haz the power, 2012-05-11 18:52:13]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/sergeyu@chromium.org/10213012/4002

[commit-bot: I haz the power, 2012-05-11 19:04:52]

Try job failure for 10213012-4002 (retry) on android for steps "compile, build"
(clobber build).
It's a second try, previously, steps "compile, build" failed.
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=android&nu...

[commit-bot: I haz the power, 2012-05-11 19:28:39]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/sergeyu@chromium.org/10213012/7003

[commit-bot: I haz the power, 2012-05-11 21:16:14]

Change committed as 136667