On Tue, Jan 10, 2023 at 12:11:12AM +1100, Mark Jamsek wrote:
> I've made all suggested changes except for moving 'path' into function
> scope because (1) with your other brilliant suggestion to allocate
> 'change' in get_diffstat() (I can't believe I didn't think of that!), we
> now only have one free(path) on failure; and (2) I actually have a diff
> in my tree making some minor changes to diff_blobs() doing much the same
> as you mentioned regarding failure leaks. I didn't want to mix that in
> with this diff though. So I might hoist path up to function scope in
> that along with the other cleanup.
> 
> Thanks, op!

Looks fine to me, ok.