Opened this issue 4 years ago · 0 comments
The current implementation is very naive. Making it go faster would be nice.