Re: [PATCH v2 1/5] winegstreamer: Require media source's IMFByteStream to be seekable.

[Zebediah Figura]

On 11/2/20 1:34 PM, Derek Lesho wrote:
> If there is ever a need to support an un-seekable bytestream, we'll have to use GstAppSrc.

Well, no, not really, you can push the buffers yourself. More saliently,
I'm not sure that GstAppSrc provides a lot of benefit, either. In order
for downstream parser elements to use pull mode, you need to use
"random-access" scheduling, but that requires that you're able to seek
the upstream source, so you'd basically need to use two different code
paths anyway.
> Signed-off-by: Derek Lesho
>
> dlls/winegstreamer/media_source.c | 7 +++++++
>  1 file changed, 7 insertions(+)
>
> diff --git a/dlls/winegstreamer/media_source.c b/dlls/winegstreamer/media_source.c
> index dbc656a24da..df5dedd484d 100644
> --- a/dlls/winegstreamer/media_source.c
> +++ b/dlls/winegstreamer/media_source.c
> @@ -1232,6 +1232,7 @@ static HRESULT media_source_constructor(IMFByteStream *bytestream, struct media_
>  struct media_source *object = heap_alloc_zero(sizeof(*object));
>  IMFStreamDescriptor **descriptors = NULL;
> - DWORD bytestream_caps;
> unsigned int i;
> HRESULT hr;
> int ret;
>
> @@ -1239,6 +1240,12 @@ static HRESULT media_source_constructor(IMFByteStream *bytestream, struct media_
>      if (!object)
>          return E_OUTOFMEMORY;
> - if (FAILED(hr = IMFByteStream_GetCapabilities(bytestream, &bytestream_caps)))
> -
>    return hr;
> -
> - if (!(bytestream_caps & MFBYTESTREAM_IS_SEEKABLE))
> -
>    return MF_E_BYTESTREAM_NOT_SEEKABLE;
> - object->IMFMediaSource_iface.lpVtbl = &IMFMediaSource_vtbl;
>    object->async_commands_callback.lpVtbl = &source_async_commands_callback_vtbl;
>    object->ref = 1;