[Webkit-unassigned] [Bug 35333] [GTK] video playback position query flood when mouse over the video element

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=35333





--- Comment #6 from Philippe Normand <pnormand at igalia.com>  2010-03-09 02:29:32 PST ---
(In reply to comment #5)
> (From update of attachment 49802 [details])
>  1572     float maxTime = maxTimeSeekable();
> 15721573     // FIXME real ranges support
> 1573      if (!maxTimeSeekable())
>  1574     if (!maxTime)
> 15741575         return TimeRanges::create();
> 1575      return TimeRanges::create(minTimeSeekable(), maxTimeSeekable());
>  1576     return TimeRanges::create(minTimeSeekable(), maxTime);
> 
> Unrelated micro-optimization? Is this really a problem? I'd put this in a
> separate patch, and get Eric Carlson to look at it.
> 

I can remove it, no worries.

> 5864 // backend can live at runtime.
> 5965 typedef struct PlatformMedia {
> 6066     QTMovie* qtMovie;
>  67     GstElement* gstPipeline;
> 6168 } PlatformMedia;
> 
> This looks weird to me. I don't think our struct should have a QTMovie in it.
> Maybe we need a #if/#else here, instead? Or event better, just write an
> implementation for platformMedia that returns the playbin! See:
> 

The comment above the declaration:

// Structure that will hold every native
// types supported by the current media player.
// We have to do that has multiple media players
// backend can live at runtime.

So as I understood it we need to stuff inside this structure. Or should we
instead conditionally define it depending on the platform? Now I am confused!
:)

> 77        virtual PlatformMedia platformMedia() const { return NoPlatformMedia;
> }
> 
> 676      return paintMediaButton(paintInfo.context, r, mediaElement->canPlay()
> ? m_playButton.get() : m_pauseButton.get(), m_panelColor, m_mediaIconSize);
>  676     return paintMediaButton(paintInfo.context, r, mediaElement->paused() ?
> m_playButton.get() : m_pauseButton.get(), m_panelColor, m_mediaIconSize);
> 
> This seems totally unrelated?

It is not! :) canPlay() triggers a call to endedPlayback() which queries the
duration. This query is useless, just checking paused() is enough in this
context and this what MediaControlPlayButtonElement::updateDisplayType does,
FWIW.

-- 
Configure bugmail: https://bugs.webkit.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.