On 2/7/22 10:09, Alexandre Julliard wrote:
Andrew Eikum aeikum@codeweavers.com writes:
On Mon, Feb 07, 2022 at 04:57:31PM +0100, Alexandre Julliard wrote:
Nikolay Sivov nsivov@codeweavers.com writes:
On 2/7/22 17:49, Andrew Eikum wrote:
@@ -853,7 +854,7 @@ static HRESULT WINAPI video_mixer_transform_SetInputType(IMFTransform *iface, DW unsigned int count; GUID *guids;
- TRACE("%p, %lu, %p, %#lx.\n", iface, id, media_type, flags);
- TRACE("%p, %lu, %p %s, %#lx.\n", iface, id, media_type, wine_dbgstr_mfmediatype(media_type), flags);
I haven't tried it, but I agree that it looks useful. One minor thing is that on null argument you get two nulls printed, but that's not a big deal.
It may be useful, but it looks awfully specific for a global header. If we want that sort of thing it would have to be made more generally useful.
Yeah, I agree. This was the least-bad solution I could come up with.
Is there a better idea? Can some kind of PARENTSRC arrangement be made to work?
With the non-recursive makefile it should now be possible to include a header directly from some other module's directory. That's not something we'd want to encourage, but for some specific cases it may be acceptable.
Perhaps it wouldn't be bad to extend strmbase, or create a new static library.