Of course, you are right. I seem to have carried the belief that freeing a null pointer was as bad as freeing one twice. And now I know. Seems to be a common m

Of course, you are right. I seem to have carried the belief that freeing a null pointer was as bad as freeing one twice. And now I know. Seems to be a common misconception [1], thanks for pointing out, that makes the GGP code correct in all counts and my switch redundant. [1] https://news.ycombinator.com/item?id=8844031 https://news.ycombinator.com/item?id=8844031