[vorbis-dev] HTTP streaming / mime type

Michael Smith msmith at labyrinth.net.au
Sat Jul 29 02:15:56 PDT 2000



> > In short - I wish. Both xmms and winamp shove all responsibility for
> > fetching the data to the actual input plugin - so each plugin has its own
> > (probably broken in one way or another) implementation. Other players
> > (freeamp comes to mind, though I haven't looked at the guts of that) do
> > this sanely - there's a seperate layer to fetch data, then one to decode.
> 
> Yes, I discovered this when I took a look at the code. :(
> 
> I have talked to the developers about it since, and they do intend to fix
> it soon, but have no concrete timeline. The current plan, so far as I
> understand, is to cut and paste the streaming code from the mp3 plugin to
> the plugin support library. Then (here it gets muddy for me) hack in some
> method of figuring out which plugin to ask to open the stream.
> 
> The also claim the streaming code stinks and needs a total rewrite.

Yeah, I've heard this too. As long as we get SOME way of doing this, then we'll
cope. I just don't want to write an entire NEW streaming layer, at least not
without a really good reason.

The same problem exists with the winamp plugin, but I suspect that ISN'T going
to change in the near future. I'm DEFINATELY not writing that much icky-windows
code without some help.

> 
> > As a result of this, writing the streaming code means doing most of it
from
> > scratch (or duplicating large chunks of, say, the mp3 plugin code). 
> 
> So the second is in fact the official plan, only without the duplication.
> 
> I'm trying to sell them on adding another layer or two of codecs, but
> don't expect to succeed. :)

It'd be nice - keep trying ;)
Regardless, it's good to know they DO recognise the problem and intend to do
something about it.

Michael

--- >8 ----
List archives:  http://www.xiph.org/archives/
Ogg project homepage: http://www.xiph.org/ogg/



More information about the Vorbis-dev mailing list