CIImage with alpha from sourceFrame of a video - swift

Question

I am trying to get CIImage with alpha, from a video that has an alpha.
The video is coded in HEVC with Alpha.
When I am doing this:
if let sourcePixel = request.sourceFrame(byTrackID: 3) {
let image = CIImage(cvPixelBuffer: sourcePixel
}
The alpha in the image is 1 (opaque).
If I test a pixel value from the relevant plane of sourcePixel, I got the correct alpha according to the value of the video.
if let sourcePixel = request.sourceFrame(byTrackID: 3) {
let results = CVPixelBufferLockBaseAddress(sourcePixel, .readOnly)
if let baseAddress = CVPixelBufferGetBaseAddressOfPlane(sourcePixel, 0) {
//Get width and height of buffer
let bytesPerRow = CVPixelBufferGetBytesPerRow(sourcePixel)
let buffer = baseAddress.assumingMemoryBound(to: UInt8.self)
var r = buffer[0]
var g = buffer[1]
var b = buffer[2]
var a = buffer[3]
print("value: ,(\(r),\(g),\(b),\(a))")
}
}
I tried to get the bitmap of the sourcePixel, and crate the CIImage from the bitmap.
But I didn't manage. I got image that looks like all the values are zero
if let sourcePixel = request.sourceFrame(byTrackID: layerInstruction.trackID) {
print (sourcePixel)
let results = CVPixelBufferLockBaseAddress(sourcePixel, .readOnly)
if let baseAddress = CVPixelBufferGetBaseAddressOfPlane(sourcePixel, 0) {
let width = CVPixelBufferGetWidth(sourcePixel)
let height = CVPixelBufferGetHeight(sourcePixel)
let bitmapData = Data(bytesNoCopy: baseAddress, count: width * height * 4, deallocator: .none)
let ciImage = CIImage(bitmapData: bitmapData, bytesPerRow: width * 4, size: CGSize(width: width, height: height), format: .ABGR8, colorSpace: CGColorSpaceCreateDeviceRGB())
}
}
Does using the bitmap is the right approach? if yes what am I doing wrong? If not what should I use?
And another related question, If I print the sourceFrame I got
<CVPixelBuffer 0x282458fa0 width=1080 height=1920 pixelFormat=420f iosurface=0x281778830 planes=2 poolName=22133:decode_1>.
<Plane 0 width=1080 height=1920 bytesPerRow=1088>.
<Plane 1 width=540 height=960 bytesPerRow=1088>
<attributes={
PixelFormatDescription = {
BitsPerComponent = 8;
ComponentRange = FullRange;
ContainsAlpha = 0;
ContainsGrayscale = 0;
ContainsRGB = 0;
ContainsYCbCr = 1;
FillExtendedPixelsCallback = {length = 24, bytes = 0x00000000000000001060248b010000000000000000000000};
IOSurfaceCoreAnimationCompatibility = 1;
IOSurfaceCoreAnimationCompatibilityHTPCOK = 1;
IOSurfaceOpenGLESFBOCompatibility = 1;
IOSurfaceOpenGLESTextureCompatibility = 1;
OpenGLESCompatibility = 1;
PixelFormat = 875704422;
Planes = (
{
BitsPerBlock = 8;
BlackBlock = {length = 1, bytes = 0x00};
},
{
BitsPerBlock = 16;
BlackBlock = {length = 2, bytes = 0x8080};
HorizontalSubsampling = 2;
VerticalSubsampling = 2;
}
);
};
} propagatedAttachments={
AlphaChannelMode = StraightAlpha;
CVFieldCount = 1;
CVImageBufferChromaLocationBottomField = Left;
CVImageBufferChromaLocationTopField = Left;
CVImageBufferColorPrimaries = "ITU_R_709_2";
CVImageBufferTransferFunction = "ITU_R_709_2";
CVImageBufferYCbCrMatrix = "ITU_R_709_2";
CVPixelAspectRatio = {
HorizontalSpacing = 1;
VerticalSpacing = 1;
};
} nonPropagatedAttachments={.
Why does the video have two planes?
why does it say that it contains no alpha?