[MPlayer-dev-eng] mp-g2 pre14

[Fabian Franz]

Am Montag, 21. April 2003 22:23 schrieb Arpi:
> Hi,
>
> > > Which will be called next, vd can also prepend filters before the
> > > first.
> > >
> > > I don't see, why this won't be possible...
> >
> > Because this does not accomplish the goal at all. It assumes input and
> > output frames are in one-to-one correspondence, and it probably
> > wouldn't work at all with dr/slices.
>
> yes, but i've mixed the 2 ideas. what about this code:

nice, that was waht I thought after the reply, but you were faster :-)

>
>
> mp_image_t* vf_process_image(vf_start){
>     vf_dunno_t* vf=vf_start;
>     mp_image_t* mpi=NULL;
>
>     // find which filter will produce the image for us:
>     // (normally this loop goes up to vf_vd, ie to the codec,
>     // except if any of the filters above us have pending queued images)
>     while((vf=vf->prev)){
> 	if(vf->process_image)
> 	    if((mpi=vf->process_image(vf,NULL))) break;
>     }
>     if(!mpi || !vf) return NULL; // error (codec failed?)
>
>     // now vf points to the top filter which produced an image: mpi, and
>     // vf_start points to the bottom filter, where we requested the image
>
>     // drive our 'mpi' through the filters from vf to vf_start, including
>     // the newcomers (auto-inserted filters)
>     while((vf=vf->next) && vf!=vf_start){
> 	if(vf->process_image)
> 	    if(!(mpi=vf->process_image(vf,mpi))) // filter image
> 		return NULL; // some filter inbetween failed :(((((
>     }
>
>     // done.
>     return mpi;
> }
>
> example filter:
>
> static mp_image_t* process_image(vf_dunno_t* vf, mp_image_t* mpi){
>     mp_image_t* dmpi;
>     if(!mpi){
> 	if(we have pending images?){
> 	    dmpi=vf_get_image(...)
> 	    render
> 	    return dmpi;
> 	}
> 	return NULL; // no thanks
>     }
>     dmpi=vf_get_image(...)
>     filter mpi into dmpi
>     while(we need one more image?){
> 	mpi=vf_process_image(vf);
> 	if(!mpi) sorry, no more...
> 	filter mpi into dmpi
>     }
>     return dmpi;
> }

This filter does always the same afais.

So you could also encapsulate it in some function like:

static mp_image_t* process_image(vf_dunno_t* vf, mp_image_t* mpi, void* 
render, void* filter)

That would have this concept, which would allow auto-inserting, but also allow 
filters to consist just out of one render and one filter-function!

simplicity and functionality.

Hm, would this work ? Or did I overlook again something ?

cu

Fabian

>
> it should work (in theory, point me posisble problems please).
> it allows any filter to insert new filters bellow itself while in
> process_image(), and the new filters will be immediately used as soon
> as it returns with mpi;
>
>
> A'rpi / Astral & ESP-team
>
> --
> Developer of MPlayer, the Movie Player for Linux - http://www.MPlayerHQ.hu
>
> _______________________________________________
> MPlayer-dev-eng mailing list
> MPlayer-dev-eng at mplayerhq.hu
> http://mplayerhq.hu/mailman/listinfo/mplayer-dev-eng