[Webkit-unassigned] [Bug 43590] Implementing video frame sharing between WebKit and Chromium

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=43590





--- Comment #5 from James Robinson <jamesr at chromium.org>  2010-08-09 17:36:22 PST ---
I don't know much about the video stack so I just have some more general questions:

Why does the VideoFrameProvider have both a getCurrentFrame() and a putCurrentFrame()?  I'm having trouble figuring out which way data is going here.  I would guess from the name VideoFrameProvider that it's an object that gives you something, not that expects to get.

Can getCurrentFrame() return a pointer instead of taking a **?

When the texture rect is resized, is the old texture being deleted somehow?

Declaring static const ints or size_ts in a .h and not providing storage for them in a .cpp is an error and won't work across all compilers we use.  Either make these enums (ugly) or leave the values out of the .h and define them in a .cpp.  People reading the header file shouldn't depend on particular values of the constants anyway so hiding them in a .cpp seems better to me.

There aren't many uses of size_t in the WebKit API except for things that are numbers of bytes.  Can the plane-related constants be unsigneds or ints?

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