[Ffmpeg-devel] volunteer for patching pal8 in libswscale
Karl H. Beckers
karl.h.beckers
Thu Mar 8 10:35:44 CET 2007
Hi all,
have recently submitted two patches:
1)
khb at ubuntu:~/Quellen/xvidcap/trunk/ffmpeg/libswscale$ diff -Naurw swscale.c.orig swscale.c
--- swscale.c.orig 2007-02-28 09:21:36.000000000 +0100
+++ swscale.c 2007-02-28 09:22:22.000000000 +0100
@@ -116,7 +116,7 @@
|| (x)==PIX_FMT_NV12 || (x)==PIX_FMT_NV21\
|| (x)==PIX_FMT_GRAY16BE || (x)==PIX_FMT_GRAY16LE\
|| (x)==PIX_FMT_GRAY8 || (x)==PIX_FMT_YUV410P)
-#define isPacked(x) ((x)==PIX_FMT_YUYV422 || (x)==PIX_FMT_UYVY422 ||isRGB(x) || isBGR(x))
+#define isPacked(x) ((x)==PIX_FMT_PAL8 || (x)==PIX_FMT_YUYV422 || (x)==PIX_FMT_UYVY422 ||isRGB(x) || isBGR(x))
#define RGB2YUV_SHIFT 16
#define BY ((int)( 0.098*(1<<RGB2YUV_SHIFT)+0.5))
this one Michael approved and said I need to pester some of the folks he
delegated patching to (though he used not exactly the same words ;) ).
But who would that be?
It's not like I found a list of people to turn to anywhere. Is there a
volunteer?
2)
--- rgb2rgb.c.orig 2007-03-05 10:53:26.000000000 +0100
+++ rgb2rgb.c 2007-03-05 11:44:46.000000000 +0100
@@ -358,12 +358,13 @@
dst[3]= palette[ src[i]*4+2 ];
dst[2]= palette[ src[i]*4+1 ];
dst[1]= palette[ src[i]*4+0 ];
+ dst[0]= 0;
#else
//FIXME slow?
dst[0]= palette[ src[i]*4+2 ];
dst[1]= palette[ src[i]*4+1 ];
dst[2]= palette[ src[i]*4+0 ];
- //dst[3]= 0; /* do we need this cleansing? */
+ dst[3]= 0;
#endif
dst+= 4;
}
@@ -378,12 +379,13 @@
dst[3]= palette[ src[i]*4+0 ];
dst[2]= palette[ src[i]*4+1 ];
dst[1]= palette[ src[i]*4+2 ];
+ dst[0]= 0;
#else
//FIXME slow?
dst[0]= palette[ src[i]*4+0 ];
dst[1]= palette[ src[i]*4+1 ];
dst[2]= palette[ src[i]*4+2 ];
- //dst[3]= 0; /* do we need this cleansing? */
+ dst[3]= 0;
#endif
dst+= 4;
this one has gone quite unnoticed and has not been approved, yet.
But the major thing, as you can see, is that I can answer the question
in the comment with "yes". We need to make sure the alpha value is set
to a defined correct value, because I have found at least one case where
not doing so yields transparent images (when encoding to png, e.g.).
TIA,
Karl.
More information about the ffmpeg-devel
mailing list