CVS update & new ideas

Hello!,

    There are a lot up updates in the CVS.  The aim is to get to a working milestone 0
prototype: read a YUV file, intra predict, encode, decode, write to another YUV file.
A lot of things remain to be done though ( specially debugging).

    We must discuss how the classes should be created.  We will end-up with a lot
of specific implementations ( for Transform, Quantize, SAD...) that must be choosed
and created at runtime, and used all over the codec.  Somehow there must be a central
factory class managing this, but may be there is a better way to do it, any ideas?

    Also we must start thinking in CAVLC and bit stream generation/parsing, any more
ideas?

    /AITOR