Re: [3/4] windowscodecs: Implement CreateBitmapFromMemory.

[Dmitry Timoshkov]

Vincent Povirk  wrote:
> -
>    hr = IWICBitmap_Lock(bitmap, NULL, WICBitmapLockWrite, &lock);
> -
>    if (SUCCEEDED(hr))
> -
>    {
> -
>        UINT buffersize;
> -
>        BYTE *buffer;
> -
> -
>        hr = IWICBitmapLock_GetDataPointer(lock, &buffersize, &buffer);
> -
> -
>        memcpy(buffer, pbBuffer, uiHeight * cbStride);
> -
> -
>        IWICBitmapLock_Release(lock);
> -
>    }

It's better to actually check return value of IWICBitmapLock_GetDataPointer
before memecpy(), or if that's not supposed to fail drop 'hr' assignment.

My approach to pass buffer directly to BitmapImpl_Create looks much simpler
than fiddling with IWICBitmap_Lock IMHO.

[3/4] windowscodecs: Implement CreateBitmapFromMemory.

[Vincent Povirk]

On Wed, Jan 2, 2013 at 11:42 PM, Dmitry Timoshkov  wrote:
> It's better to actually check return value of IWICBitmapLock_GetDataPointer
> before memecpy(), or if that's not supposed to fail drop 'hr' assignment.

Whoops, good catch. (I don't think it can fail, but I didn't mean to
make that assumption.)

I guess I prefer to keep interfaces as simple as possible even if it
means writing more code? I'd consider writing a helper function, but
none of the other methods would use it.

[3/4] windowscodecs: Implement CreateBitmapFromMemory.

[Dmitry Timoshkov]

Vincent Povirk  wrote:
> I guess I prefer to keep interfaces as simple as possible even if it
> means writing more code? I'd consider writing a helper function, but
> none of the other methods would use it.

Is there any hurry with CreateBitmapFromMemory implementation, or you
somehow prefer your own one? I'm asking because I was planning to send
my implementation (fixed to create a copy of the passed in buffer) after
new year vacations, but it looks like you need it right now.