将C uint8_t指针转换为C＃字节数组

Question

I have exported the following C function to be invoked in a DLL file.

uint8_t* _stdcall GetCurrentImage();

Now I want to call this function in C# to get a bitmap Image. How can I do this?

Thanks in advance!

Answer 1

You will need to know the exact number of bytes returned and the bitmap dimensions (height, width and encoding). Then you can declare it in C# as:

[DllImport("yourlib.dll")]
private static extern IntPtr GetCurrentImage();

The IntPtr you get from that can be used with Marshal.Copy , to get the raw bytes:

byte[]  buffer = new byte[length];
IntPtr beginPtr = GetCurrentImage();
Marshal.Copy(beginPtr, buffer,0,length);

Finally, declare a Bitmap with the dimensions of your image and the PixelFormat used (if it is a non-standard pixelformat, you might have to do some conversion yourself). Then you can copy the data into the raw bitmap bytes, by using LockBits to get a BitmapData instance that points to the raw bitmap data.

Answer 2

If you want to keep on working in an unsafe context even in C#, you can also simplify everything by bypassing Marshal and IntPtrs as follows:

[DllImport("Library.dll")]
internal static unsafe extern Byte* GetCurrentImage();

Keep in mind that's always good practice to insert all of your imports in a static class called NativeMethods and declare them as internal.

Answer 3

The common strategy for things like this is to pass in a buffer and have the function fill the buffer. Otherwise you run into trouble with disposing of the memory allocated; in other words, you will have a memory leak.

So you should have C functions like this...

uint32_t _stdcall GetCurrentImageSize();
void _stdcall GetCurrentImage(uint8_t* buffer, int offset, int bufferSize);

And then in C# you would do something like...

var buffer = new byte [ GetCurrentImageSize() ];
fixed (byte* b = &buffer[0])
    GetCurrentImage(b, 0, buffer.Length);

// 'buffer' now contains the image