public inbox for mauve-discuss@sourceware.org
 help / color / mirror / Atom feed
* Possible Mauve bug.
@ 2005-09-03 11:12 Meskauskas Audrius
  0 siblings, 0 replies; 3+ messages in thread
From: Meskauskas Audrius @ 2005-09-03 11:12 UTC (permalink / raw)
  To: mauve-discuss

[-- Attachment #1: Type: text/plain, Size: 987 bytes --]

The attached test case specifically fails when running as part of the 
Mauve under Sun's jre 1.4.2 but succeeds if executed as standalone 
program under the mentioned Sun's jre. GNU Classpath implementation 
passes this test regardless if it is executed standalone or as part of 
the Mauve.

The test case is derived from the prepared Mauve tests for the 
org.omg.PortableInterceptor. I am not sure if I can commit the test set 
that totally fails under Sun's jre due, I think, Mauve related reason. 
This set runs perfectly under JUnit.

As I cannot look into the Sun's sources, it is very difficult for me to 
find the exact reason, why the test fails. All I can do is guess looking 
at my own code that is working. The CORBA implementation must call 
Class.forName(String) somewhere; probably that call fails?

Probably the problem could be easier solved by somebody who knows Mauve 
on much better level "I wrote it" (I did not participate in Mauve 
development).

Regards

Audrius


[-- Attachment #2: mt.java --]
[-- Type: text/x-java, Size: 1427 bytes --]

// Tags: JDK1.4

package gnu.testlet.org.omg.PortableInterceptor.Interceptor;

import gnu.testlet.TestHarness;
import gnu.testlet.Testlet;

import org.omg.CORBA.LocalObject;
import org.omg.CORBA.ORB;
import org.omg.PortableInterceptor.ORBInitInfo;
import org.omg.PortableInterceptor.ORBInitializer;

import java.util.Properties;

public class mt extends LocalObject implements Testlet, ORBInitializer
{
  static boolean was_pre_init;
  static boolean was_post_init;

  public void test(TestHarness harness)
  {
    try
      {
        Properties initialisers = new Properties();
        initialisers.setProperty(
          "org.omg.PortableInterceptor.ORBInitializerClass." +
          getClass().getName(),
          ""
        );

        // Create and initialize the ORB
        ORB orb = org.omg.CORBA.ORB.init(new String[ 0 ], initialisers);
      }
    catch (Exception e)
      {
        e.printStackTrace();
      }
    if (!was_pre_init || !was_post_init)
      System.out.println("mt:TEST FAILED");
    else
      System.out.println("mt:Test passed");
  }

  public void pre_init(ORBInitInfo parm1)
  {
    System.out.println("mt:Pre init called");
    was_pre_init = true;
  }

  public void post_init(ORBInitInfo parm1)
  {
    System.out.println("mt:Post init called");
    was_post_init = true;
  }

  public static void main(String[] args)
  {
    // Run as standalone application.
    new mt().test(null);
  }
}

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: Possible Mauve bug.
  2005-09-03 12:08 Jeroen Frijters
@ 2005-09-03 14:52 ` Meskauskas Audrius
  0 siblings, 0 replies; 3+ messages in thread
From: Meskauskas Audrius @ 2005-09-03 14:52 UTC (permalink / raw)
  To: mauve-discuss

Then probably some very version - specific bug in 1.4.2. Thanks, I commit.

Jeroen Frijters wrote:

>Meskauskas Audrius wrote:
>  
>
>>The attached test case specifically fails when running as part of the 
>>Mauve under Sun's jre 1.4.2 but succeeds if executed as standalone 
>>program under the mentioned Sun's jre. GNU Classpath implementation 
>>passes this test regardless if it is executed standalone or 
>>as part of the Mauve.
>>    
>>
>
>FWIW, on JRE 1.4.1 and JDK 1.5 (both on Windows) this tests run fine
>under Mauve.
>
>Here's the output:
>C:\mauve>\j2re1.4.1\bin\java gnu.testlet.SimpleTestHarness
>gnu.testlet.org.omg.PortableInterceptor.Interceptor.mt
>mt:Pre init called
>mt:Post init called
>mt:Test passed
>^Z
>0 of 0 tests failed
>
>C:\mauve>\jdk1.5\bin\java gnu.testlet.SimpleTestHarness
>gnu.testlet.org.omg.PortableInterceptor.Interceptor.mt
>mt:Pre init called
>mt:Post init called
>mt:Test passed
>^Z
>0 of 0 tests failed
>
>BTW, thanks for the incredible work you're doing on the CORBA
>implementation.
>
>Regards,
>Jeroen
>
>
>  
>

^ permalink raw reply	[flat|nested] 3+ messages in thread

* RE: Possible Mauve bug.
@ 2005-09-03 12:08 Jeroen Frijters
  2005-09-03 14:52 ` Meskauskas Audrius
  0 siblings, 1 reply; 3+ messages in thread
From: Jeroen Frijters @ 2005-09-03 12:08 UTC (permalink / raw)
  To: audriusa, mauve-discuss

Meskauskas Audrius wrote:
> The attached test case specifically fails when running as part of the 
> Mauve under Sun's jre 1.4.2 but succeeds if executed as standalone 
> program under the mentioned Sun's jre. GNU Classpath implementation 
> passes this test regardless if it is executed standalone or 
> as part of the Mauve.

FWIW, on JRE 1.4.1 and JDK 1.5 (both on Windows) this tests run fine
under Mauve.

Here's the output:
C:\mauve>\j2re1.4.1\bin\java gnu.testlet.SimpleTestHarness
gnu.testlet.org.omg.PortableInterceptor.Interceptor.mt
mt:Pre init called
mt:Post init called
mt:Test passed
^Z
0 of 0 tests failed

C:\mauve>\jdk1.5\bin\java gnu.testlet.SimpleTestHarness
gnu.testlet.org.omg.PortableInterceptor.Interceptor.mt
mt:Pre init called
mt:Post init called
mt:Test passed
^Z
0 of 0 tests failed

BTW, thanks for the incredible work you're doing on the CORBA
implementation.

Regards,
Jeroen

^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2005-09-03 14:52 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2005-09-03 11:12 Possible Mauve bug Meskauskas Audrius
2005-09-03 12:08 Jeroen Frijters
2005-09-03 14:52 ` Meskauskas Audrius

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).