Foo.java

public class Foo {
    public static void main(String[] args) {
        String x = "This is weird: \u0022\u003B\u0069\u006E\u0074\u0020\u0079\u003D\u0031\u002F\u0030\u003B\u0078\u003D\u0022???";
        System.out.println("x = " + x);
    }
}

By this logic this https://gist.github.com/drXor/8100658 compiles.

Division by 0 x)
Now I just want to mess up with the compiler :)

:D

@drXor When people with no programming knowledge posted asking for someone to modify a plugin, I'd usually give it to them like that, just to see if they knew what was what.

Java supports non-ASCII for names of variables, classes and so forth. If you receive files from some developer with non-ASCII name in them which you don't want to corrupt, using native2ascii is one (dodgy, IMO) option you have to convert it to a plain ASCII file which compiles to the same result. Because Sun wanted to support this, Unicode escapes work anywhere, not just inside strings.

The fact that they also work for encoding basic stuff like double quotes is highly amusing, though.

BTW, if you run the file through native2ascii -reverse Foo.java Foo.java.native, you get the real code being compiled.