Looking at the coverage of the file: why is the "ideal by ideal" function for the absolute case in there three times? (And only covered once.)
But that is independent of this pull request, which should be done from my side.
Originally posted by @joschmitt in #2188 (comment)