Im trying to load an image in a coremediaIO extension. Ive successfully loaded the image and draw it in 3 different ways. Once from the AppGroup, once from the extension bundle and another time as a base64 string. No problems each time the image loads and renders. However, when I call the image to be drawn in a pixelbuffer it loads the extension into the dock. Id love to be able to draw an image and the extension not appear in the dock but I cannot figure this out. Can render text in this timer loop no problem. Any suggestions other than sending the image across the sink stream?
Essentially i want to render the app logo whenever the host app is not sending signal to the extension.
func startStreaming() {
guard let _ = _bufferPool else {
return
}
_streamingCounter += 1
_timer = DispatchSource.makeTimerSource(flags: .strict, queue: _timerQueue)
_timer!.schedule(deadline: .now(), repeating: 1.0/Double(kFrameRate), leeway: .seconds(0))
_timer!.setEventHandler {
if self.sinkStarted {
return
}
var err: OSStatus = 0
var pixelBuffer: CVPixelBuffer?
err = CVPixelBufferPoolCreatePixelBufferWithAuxAttributes(kCFAllocatorDefault, self._bufferPool, self._bufferAuxAttributes, &pixelBuffer)
if err != 0 {
logger.debug("out of pixel buffers \(err)")
}
if let pixelBuffer = pixelBuffer {
CVPixelBufferLockBaseAddress(pixelBuffer, [])
let pixelData = CVPixelBufferGetBaseAddress(pixelBuffer)
let width = CVPixelBufferGetWidth(pixelBuffer)
let height = CVPixelBufferGetHeight(pixelBuffer)
let rgbColorSpace = CGColorSpaceCreateDeviceRGB()
if let context = CGContext(data: pixelData,
width: width,
height: height,
bitsPerComponent: 8,
bytesPerRow: CVPixelBufferGetBytesPerRow(pixelBuffer),
space: rgbColorSpace,
bitmapInfo: CGBitmapInfo.byteOrder32Little.rawValue | CGImageAlphaInfo.premultipliedFirst.rawValue) {
let graphicsContext = NSGraphicsContext(cgContext: context, flipped: false)
NSGraphicsContext.saveGraphicsState()
NSGraphicsContext.current = graphicsContext
let cgContext = graphicsContext.cgContext
let dstRect = CGRect(x: 0, y: 0, width: width, height: height)
cgContext.clear(dstRect)
cgContext.setFillColor(NSColor.black.cgColor)
cgContext.fill(dstRect)
let imageWidth = 400 // You can adjust the width as needed
let imageHeight = 400 // You can adjust the height as needed
let imageOrigin = CGPoint(x: (width - imageWidth) / 2, y: (height - imageHeight) / 2) // Center the image
if let decodedData = Data(base64Encoded: imageBaseString, options: .ignoreUnknownCharacters),
let image = NSImage(data: decodedData), let cgImage = image.cgImage(forProposedRect: nil, context: nil, hints: nil) {
cgContext.draw(cgImage, in: CGRect(origin: imageOrigin, size: NSSize(width: imageWidth, height: imageHeight)))
}
NSGraphicsContext.restoreGraphicsState()
}
CVPixelBufferUnlockBaseAddress(pixelBuffer, [])
}
if let pixelBuffer = pixelBuffer {
var sbuf: CMSampleBuffer!
var timingInfo = CMSampleTimingInfo()
timingInfo.presentationTimeStamp = CMClockGetTime(CMClockGetHostTimeClock())
err = CMSampleBufferCreateForImageBuffer(allocator: kCFAllocatorDefault, imageBuffer: pixelBuffer, dataReady: true, makeDataReadyCallback: nil, refcon: nil, formatDescription: self._videoDescription, sampleTiming: &timingInfo, sampleBufferOut: &sbuf)
if err == 0 {
self._streamSource.stream.send(sbuf, discontinuity: [], hostTimeInNanoseconds: UInt64(timingInfo.presentationTimeStamp.seconds * Double(NSEC_PER_SEC)))
}
}
}
_timer!.setCancelHandler {}
_timer!.resume()
}
Culprit is here:
if let decodedData = Data(base64Encoded: imageBaseString, options: .ignoreUnknownCharacters),
let image = NSImage(data: decodedData), let cgImage = image.cgImage(forProposedRect: nil, context: nil, hints: nil) {
cgContext.draw(cgImage, in: CGRect(origin: imageOrigin, size: NSSize(width: imageWidth, height: imageHeight)))
}
Doesnt matter how I load the image, the dock extension will trigger as soon as the draw loop is triggered or on extension init if I load the image then. Is this possible?
