Apparently, the ptBR locale uses "." for comma separators AND for thousand separators. This makes it much more complicated to detect ANY number properly, even if the current implementation seems to work for higher AK levels.

I'm not really sure what to do about this right now, as there are also issues with trailing zeroes and any workaround I came up with didn't really solve the problem correctly. On the other hand, I also don't want to rewrite the entire scanning algorithm just for this.