[vlc-devel] Re: How is CD-Text (or CDDB, Musicbrainz) supposed to work with the new infrastructure?

[Derk-Jan Hartman]

Clément Stenac wrote:
>>I'm not really all that enthusiastic about working on something
>>then finding it's been changed so either the work I spent was
>>superflous (as in the OGT and CVD subtitles)
> 
> I don't think one can say this work was superflous as anyway, some of it
> will get reused in the new subtitle filters.

I don't think it was superflous at all. I think it was the basic idea 
the whole new subtitle rendering proces was based on.
However for a complete and 'final' solution there is always more to 
consider and learn. For instance you were drawing to png, but that was 
adding additional dependancies, and was also complicating the work for 
blending into a stream..

A filter system was needed, and a lossless and easy to draw chroma 
format (YUVA) was used instead allowing for greater flexibility in a 
final system.
Combined with the picture filter system that is (kind of) in 
developement, multiple actions will be possible, like scaling, 
overlayíng etc etc.

However the basic idea is still very much based i think on your original 
suggestions and code. A common format with a centralized position in the 
core. The png route though was just adding a bit too much overhead...

In my experience the first implementation of something is destined to 
fail, but also is a path for the future.

DJ

-- 
This is the vlc-devel mailing-list, see http://www.videolan.org/vlc/
To unsubscribe, please read http://developers.videolan.org/lists.html
If you are in trouble, please contact <postmaster at videolan.org>