Re: Getting MDR 4.0 working with TCK

Brian Topping <[email protected]> Tue, 26 Oct 2004 06:12:51 -0400
Newsgroups gmane.comp.java.netbeans.modules.mdr.devel
Message-ID <[email protected]>
This is a multi-part message in MIME format.
--------------030608080806090609070107
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit

Yes, looking at it, it is a bit strange.

/Test: 
modelsem/AttributeTests/InstanceAttributes/GeneratedInterfaces/TD.html#Child_1-1_setShared

Based on the results of the preceeding sections, the script determined 
that the result of the test is:  Failed.  Value returned bu second owner 
com.test.simplemodel.Owner2 using method getChildAttr is not equal to 
the initial attribute value (operation not completed due to caught 
javax.jmi.reflect.CompositionViolationException)

Executing method getChildAttr on an instance of 
com.tst.simplemodel.Owner2$Impl with 0 arguments
Before attribute of second instance is set, initial value of  attribute 
is Child$Impl  ID: .:0000000000000782  MID: .:000000000000071D  OPCKG: 
.:000000000000075B
Executing method setChildAttr on an instance of 
com.tst.simplemodel.Owner1$Impl with 1 arguments
  arg[0] => id=.:0000000000000784 : SimpleModel.Child
Executing method setChildAttr on an instance of 
com.tst.simplemodel.Owner2$Impl with 1 arguments
  arg[0] => id=.:0000000000000784 : SimpleModel.Child
Expecting the method to throw exception 
javax.jmi.reflect.CompositionViolationException
Method setChildAttr threw a 
javax.jmi.reflect.CompositionViolationException with message: null
Expected exception javax.jmi.reflect.CompositionViolationException caught
Executing method getChildAttr on an instance of 
com.tst.simplemodel.Owner2$Impl with 0 arguments
The attribute of a com.tst.simplemodel.Owner2 has the value Child$Impl  
ID: .:0000000000000784  MID: .:000000000000071D  OPCKG: 
.:000000000000075B accessed using method getChildAttr/

As best I can tell, this is the correct behavior, that a CVE is supposed 
to be thrown.  The three tests that failed are all the same test, 
failing in 
modelsem/AttributeTests/InstanceAttributes/GeneratedInterfaces, 
modelsem/AttributeTests/InstanceAttributes/ReflectiveInterfaces/Metaobject-Designated 
and 
modelsem/AttributeTests/InstanceAttributes/ReflectiveInterfaces/String-Designated.  
The debug.log does not appear to be unhappy with what happened, but I 
have not spent much time understanding all the code yet.

I got the TCK from the JMI team several months ago, but presumably it's 
not a work in progress, right? :-)

thanks,

-b


Martin Matula wrote:

> That's strange. All tests work for me. You may need to clean the TCK 
> workdir and MDR storage files before you re-run the tests.
> Martin
>
> Brian Topping wrote:
>
>> Hi Martin,
>>
>> Those instructions worked well.  My problem ended up being a bunch of 
>> objects in the btree store that were left over from when I didn't 
>> have the classpath set up correctly.  There are about three tests 
>> that aren't working but I'm not going to worry about that for now.
>>
>> Thanks!
>>
>> Brian
>
>
>
>
>
>


--------------030608080806090609070107
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit

<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
  <meta content="text/html;charset=ISO-8859-1" http-equiv="Content-Type">
</head>
<body bgcolor="#ffffff" text="#000000">
Yes, looking at it, it is a bit strange. <br>
<br>
<i>Test:
modelsem/AttributeTests/InstanceAttributes/GeneratedInterfaces/TD.html#Child_1-1_setShared<br>
<br>
Based on the results of the preceeding sections, the script determined
that the result of the test is:&nbsp; Failed.&nbsp; Value returned bu second
owner com.test.simplemodel.Owner2 using method getChildAttr is not
equal to the initial attribute value (operation not completed due to
caught javax.jmi.reflect.CompositionViolationException)<br>
<br>
Executing method getChildAttr on an instance of
com.tst.simplemodel.Owner2$Impl with 0 arguments<br>
Before attribute of second instance is set, initial value of&nbsp; attribute
is Child$Impl&nbsp; ID: .:0000000000000782&nbsp; MID: .:000000000000071D&nbsp; OPCKG:
.:000000000000075B<br>
Executing method setChildAttr on an instance of
com.tst.simplemodel.Owner1$Impl with 1 arguments<br>
&nbsp; arg[0] =&gt; id=.:0000000000000784 : SimpleModel.Child<br>
Executing method setChildAttr on an instance of
com.tst.simplemodel.Owner2$Impl with 1 arguments<br>
&nbsp; arg[0] =&gt; id=.:0000000000000784 : SimpleModel.Child<br>
Expecting the method to throw exception
javax.jmi.reflect.CompositionViolationException<br>
Method setChildAttr threw a
javax.jmi.reflect.CompositionViolationException with message: null<br>
Expected exception javax.jmi.reflect.CompositionViolationException
caught<br>
Executing method getChildAttr on an instance of
com.tst.simplemodel.Owner2$Impl with 0 arguments<br>
The attribute of a com.tst.simplemodel.Owner2 has the value Child$Impl&nbsp;
ID: .:0000000000000784&nbsp; MID: .:000000000000071D&nbsp; OPCKG:
.:000000000000075B accessed using method getChildAttr</i><br>
<br>
As best I can tell, this is the correct behavior, that a CVE is
supposed to be thrown.&nbsp; The three tests that failed are all the same
test, failing in
modelsem/AttributeTests/InstanceAttributes/GeneratedInterfaces,
modelsem/AttributeTests/InstanceAttributes/ReflectiveInterfaces/Metaobject-Designated
and
modelsem/AttributeTests/InstanceAttributes/ReflectiveInterfaces/String-Designated.&nbsp;
The debug.log does not appear to be unhappy with what happened, but I
have not spent much time understanding all the code yet.<br>
<br>
I got the TCK from the JMI team several months ago, but presumably it's
not a work in progress, right? :-)<br>
<br>
thanks,<br>
<br>
-b<br>
<br>
<br>
Martin Matula wrote:
<blockquote cite="[email protected]" type="cite">That's
strange. All tests work for me. You may need to clean the TCK workdir
and MDR storage files before you re-run the tests.
  <br>
Martin
  <br>
  <br>
Brian Topping wrote:
  <br>
  <br>
  <blockquote type="cite">Hi Martin,
    <br>
    <br>
Those instructions worked well.&nbsp; My problem ended up being a bunch of
objects in the btree store that were left over from when I didn't have
the classpath set up correctly.&nbsp; There are about three tests that
aren't working but I'm not going to worry about that for now.
    <br>
    <br>
Thanks!
    <br>
    <br>
Brian
    <br>
  </blockquote>
  <br>
  <br>
  <br>
  <br>
  <br>
</blockquote>
<br>
</body>
</html>

--------------030608080806090609070107--