Compilation (still) depends on order of imports

[Liam Miller-Cushon]

I think I found a bug related to JEP 216. With javac9-dev @ r2850, the
following program fails to compile unless the order of imports is reversed.

It looks like the current implementation isn't lazy enough to handle the
dependency between the two on-demand imports if 'I' is inherited into 'C'.
If 'I' is declared directly in 'C' then it works.

{{{
package P;

import static P.Outer.Nested.*;
import static P.Q.C.*;

public class Outer {
  public static class Nested implements I {
  }
}

package P.Q;

public class C extends D {
}

package P.Q;

public class D {
  public interface I {
  }
}
}}}

$ javac P/Outer.java P/Q/D.java P/Q/C.java
P/Outer.java:7: error: cannot find symbol
  public static class Nested implements I {
                                        ^
  symbol:   class I
  location: class Outer
1 error
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.openjdk.java.net/pipermail/compiler-dev/attachments/20150315/05932dc3/attachment.html>