Re: [OpenJDK 2D-Dev] JDK 9 RFR of JDK-8044694: Fix finally lint warnings in sun.print

[Phil Race]

Looks good.

-phil.

On 6/3/2014 11:42 AM, Joe Darcy wrote:

On 06/03/2014 11:34 AM, Phil Race wrote:

Both of these came in via an IBM contributed fix :-
http://mail.openjdk.java.net/pipermail/2d-dev/2011-October/002206.html

He [it appears] wanted to be very sure the throw was executed but it seems like it could just be moved after the finally block which might be better

than suppressing the warning.

That approach would be fine with me. New webrev uploaded to

    http://cr.openjdk.java.net/~darcy/8044694.1

Patch below.

Thanks,

-Joe

diff -r 6d4b3a9ca33d src/share/classes/sun/print/PSPrinterJob.java

--- a/src/share/classes/sun/print/PSPrinterJob.java Tue Jun 03 09:22:59 2014 -0700 +++ b/src/share/classes/sun/print/PSPrinterJob.java Tue Jun 03 11:41:38 2014 -0700

@@ -708,8 +708,8 @@
                     }
                 } finally {
                     pw.flush();
-                    throw new IOException(sw.toString());
                 }
+                throw new IOException(sw.toString());
             }
         }

diff -r 6d4b3a9ca33d src/solaris/classes/sun/print/UnixPrintJob.java

--- a/src/solaris/classes/sun/print/UnixPrintJob.java Tue Jun 03 09:22:59 2014 -0700 +++ b/src/solaris/classes/sun/print/UnixPrintJob.java Tue Jun 03 11:41:38 2014 -0700

@@ -992,8 +992,8 @@
                     }
                 } finally {
                     pw.flush();
-                    throw new IOException(sw.toString());
                 }
+                throw new IOException(sw.toString());
             }
         }

-phil.

On 6/3/2014 10:11 AM, Joe Darcy wrote:

Hello,

Please review a fix for

    JDK-8044694: Fix finally lint warnings in sun.print
    http://cr.openjdk.java.net/~darcy/8044694.0/

One of javac's lint warnings is for a finally block that cannot complete normally. The printing code has two instances with a finally block like

                  } finally {
                      pw.flush();
                      throw new IOException(sw.toString());
                  }

This cannot complete normally, since an exception is unconditionally thrown, but it very much seems to be the intended behavior. The fix is to add @SuppressWarnings("finally") annotations:

--- old/src/share/classes/sun/print/PSPrinterJob.java 2014-06-03 10:07:16.000000000 -0700 +++ new/src/share/classes/sun/print/PSPrinterJob.java 2014-06-03 10:07:16.000000000 -0700

@@ -687,6 +687,7 @@

// Inner class to run "privileged" to invoke the system print command

+ @SuppressWarnings("finally") // Exception always thrown in finally block private class PrinterSpooler implements java.security.PrivilegedAction {

         PrinterException pex;

--- old/src/solaris/classes/sun/print/UnixPrintJob.java 2014-06-03 10:07:16.000000000 -0700 +++ new/src/solaris/classes/sun/print/UnixPrintJob.java 2014-06-03 10:07:16.000000000 -0700

@@ -971,6 +971,7 @@

// Inner class to run "privileged" to invoke the system print command

+ @SuppressWarnings("finally") // Exception always thrown in finally block private class PrinterSpooler implements java.security.PrivilegedAction {

         PrintException pex;

Thanks,

-Joe