On 2/11/22 03:36, Rémi Bernon wrote:
And use it for decoder transform input types.
Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=51931 Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=52391 Signed-off-by: Rémi Bernon rbernon@codeweavers.com
This uses a separate struct for encoded formats, as I believe it was suggested at some point (or I may be mistaken), but I'm not sure it's best. If we intend to support encoders or converters in the wg_transform the formats would need to be swapped, and in which case using the same struct may be cleaner.
In that case, maybe using a separate major type for encoded formats, or a separate major type for each encoded format would be better.
I'm not sure if I suggested such a thing, but in any case I don't think it makes sense to use a separate struct. Note also that we already express some compressed formats using the wg_format structure, namely Cinepak and MPEG-1 audio.
We don't use a separate major type for those, and that's essentially because they share almost all of the same other attributes (frame size, frame rate, channel count/mask, although not audio sample rate). On the other hand some compressed formats, like WMA, need extra data, so maybe it makes more sense to introduce a new major type for those. I don't have strong opinions either way. I also think it's fine to add a separate major type for some formats but not others, based on how different the data needs to be.