Date: 2019-03-11 14:23 Sender: Martin MaechlerYour guess is correct: The bug is in R's isSymmetric(.) which is wrong here
[only for 1x1 matrices with asymmetric dimnames].
Hence a bug in R -- which I now have fixed in the R sources.
But "of course" it's still present in R 3.5.3 which was released today.
As this is a border case, I don't think we should establish a workaround (of the R bug) in Matrix. |