Hi,
> I wonder whether the function should not return the ref > pointer, regardless whether it is set to NULL. In that case it overwrites > the actual value of the pointer. Maybe this is more like expected behaviour. Hmm not sure what to expect here, requiring the pointer to be set to NULL is a clear trigger that something is wrong if an already used reference is passed. Using clear error handling normally would make this easy to find. Also a response to your question regarding error messages for the return codes, if this is in place it can be used to detect such problems. So yes it has to be added, feel free to add an issue. I will see if I can add the basic code for this, so that only the list of messages has to be extended. -- Met vriendelijke groet, Alexander Broekhuis
