Re: JDK 9 RFR of 8044698: Fix finally lint warnings in javax.swing

[Phil Race]

Looks ok to me ..

-phil.

On 6/3/2014 10:53 AM, Joe Darcy wrote:

Hello,

Please review this change to fix

    8044698: Fix finally lint warnings in javax.swing
    http://cr.openjdk.java.net/~darcy/8044698.0/

The "finally" javac lint warning issues a warning when a finally block always completes abruptly. This happens if an exception is thrown in the finally block, but by the definitions in the JLS, also happens for a return statement in the block. (A return statement in a finally will make pending exceptions be ignored as well as an another in-progress return statement.)

For the method in question here, the fix is simple. Since the entire code of the method is in a single try block and the return statement is the last statement in the finally clause, the return statement is just moved outside of the finally block:

--- old/src/share/classes/javax/swing/JEditorPane.java 2014-06-03 10:40:24.000000000 -0700 +++ new/src/share/classes/javax/swing/JEditorPane.java 2014-06-03 10:40:24.000000000 -0700

@@ -679,8 +679,8 @@
                         }
                     });
                 }
-                return (pageLoaded ? page : old);
             }
+            return (pageLoaded ? page : old);
         }

         /**

The other return statements in the try block are consistent with the last return statement's handling of pageLoaded. In other words, the other return statements return old when pageLoaded is false. Therefore, the value computed by "return (pageLoaded ? page : old);" is consistent with computed values of the other return statements.

Thanks,

-Joe