Yes, it's ok to use unicode in the source, and the output will work.
Not very important feature, this just came up when i want to scan a source that coded in BLIde.
Not very important feature, this just came up when i want to scan a source that coded in BLIde.