[Webkit-unassigned] [Bug 154235] [GTK][GStreamer] ClearKey EME v1 decryption support

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=154235

--- Comment #38 from Philippe Normand <pnormand at igalia.com> ---
Comment on attachment 272316
  --> https://bugs.webkit.org/attachment.cgi?id=272316
patch

View in context: https://bugs.webkit.org/attachment.cgi?id=272316&action=review

>>>> Source/WebCore/platform/graphics/gstreamer/WebKitClearKeyDecryptorGStreamer.h:43
>>>> +    WebKitMediaCommonEncryptionDecrypt parent;
>>> 
>>> So the base parent class is for encryption and decryption? Do I understand correctly that this is the ClearKey implementation for the base class for decryption?
>> 
>> CommonEncryption (cenc) is what is specified in MPEG ISO BMF. CommonEncryptionDecrypt is the base class for decryption.
> 
> What's ClearKey then?

It's described in the spec :) https://www.w3.org/TR/encrypted-media/ section 9.1

>>>> Source/WebCore/platform/graphics/gstreamer/WebKitCommonEncryptionDecryptorGStreamer.cpp:56
>>>> +    return cencClass->setupCipher(self);
>>> 
>>> Is it expected that derived classes implement this method? Then add an assert to ensure cencClass is mot null, otherwise add a null check.
>> 
>> We provide a default implementation for this vfunc, so I don't think this code needs to be changed.
> 
> Can't there be more implementations? Why are we using two classes then? This is not about what the known implementation does, but what the base class wants to enforce. Think of it as whether this is a pure virtual method or not.

Yes there can be multiple implementations. At least the PlayReady key system can also rely on common-encryption.

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.webkit.org/pipermail/webkit-unassigned/attachments/20160226/d7d04dcd/attachment.html>