Skip to main content
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [List Home]
Re: [aspectj-dev] AspectJ 1.1.1 is on its way... AspectJ 1.1.1rc1 is here 
Thanks for the sources.

I checked in fixes, so you shouldn't see these problems again.
(And I'm inclined to hack JUnit to make the running order of tests random...) 

- AspectJBuildManagerTest.java
All "tests" put in one because some depended for builds on others.
Tested functionality is the same as before when passing.  Mik can
review and pull out independent tests at his leisure.

- TestingDrivers...
Now cloning the test specifications before reuse, since option
processing dirties them.

Wes


George Harley1 wrote:
Hi Wes, Sure, I will send the modified sources to you off-line together with some other observations. 

Best regards,
George
________________________________________
George C. Harley





Wes Isberg <wes@xxxxxxxxxxxxxx>
Sent by: aspectj-dev-admin@xxxxxxxxxxx
28/08/2003 14:58
Please respond to aspectj-dev
To: aspectj-dev@xxxxxxxxxxx cc: Subject: Re: [aspectj-dev] AspectJ 1.1.1 is on its way... AspectJ 1.1.1rc1 is here
> Maybe the fact that we have unit test cases whose success relies on being 
 > run in a particular sequence is something we might want to address post
 > 1.1.1 ?

That would clearly be a test bug, but I'm not sure that's why the
TestingDriversModulesTests are failing.  I'll look into those.
Can you send me the source file with the ordering that works
for IBM JDK 1.3.1 for that test?

Thanks -
Wes


George Harley1 wrote:



Hi,
I spent a bit of time today doing some install and verification testing
of
this release candidate on my Linux server. Everything seemed fine. On Windows I ran the JUnit and harness tests against the rc on the
latest
level IBM 131 and 141 JVMs. On the IBM 141 everything passed 100% while
on
the IBM 131 there were a grand total of three JUnit test failures. These 



were :-

AjdeModuleTests (1 failure)
-----------------------------------------
testCompileOutput : failure : expected class testdata\AspectJBuildManagerTest\bin\figures\Main.class junit.framework.AssertionFailedError: expected class testdata\AspectJBuildManagerTest\bin\figures\Main.class at 

org.aspectj.ajde.internal.AspectJBuildManagerTest.testCompileOutput(AspectJBuildManagerTest.java:59)



TestingDriversModuleTests (2 failures)
----------------------------------------------------------
testTitleListFileSelectionPluralFailOnly : failure : skips expected:<15> 



but was:<16>
junit.framework.AssertionFailedError: skips expected:<15> but was:<16>
at 

org.aspectj.testing.drivers.HarnessSelectionTest.checkSelection(HarnessSelectionTest.java:346)
at 

org.aspectj.testing.drivers.HarnessSelectionTest.testTitleListFileSelectionPluralFailOnly(HarnessSelectionTest.java:280)
testTitleListFileSelectionPlural : failure : skips expected:<15> but was:<16> 
junit.framework.AssertionFailedError: skips expected:<15> but was:<16>
at 

org.aspectj.testing.drivers.HarnessSelectionTest.checkSelection(HarnessSelectionTest.java:346)
at 

org.aspectj.testing.drivers.HarnessSelectionTest.testTitleListFileSelectionPlural(HarnessSelectionTest.java:269)
As previously noticed, differences between the IBM 131 reflection implementation and those found in the Sun 131, Sun 141 and IBM 141 JVMs affects the order in which the JUnit tests get run and it looks like
this
is again responsible for these failing cases (I understand that the IBM reflection code has changed a lot between 131 and 141). By changing the order in which the IBM 131 JVM runs the tests (just moving the methods around the source files) the above "failing" tests can be made to pass. Maybe the fact that we have unit test cases whose success relies on
being
run in a particular sequence is something we might want to address post 1.1.1 ? 

Best regards,
George
________________________________________
George C. Harley





Adrian Colyer/UK/IBM@IBMGB
Sent by: aspectj-dev-admin@xxxxxxxxxxx
26/08/2003 09:14
Please respond to aspectj-dev


       To:     ajdt-dev@xxxxxxxxxxx, <aspectj-dev@xxxxxxxxxxx>
cc: Subject: [aspectj-dev] AspectJ 1.1.1 is on its way... 

AspectJ 1.1.1rc1 is here
We're planning to put out an AspectJ 1.1.1 release soon. To help us get the 
quality of the release as high as possible, we have built a release
candidate, 1.1.1rc1.

We appreciate your help in squashing any bugs in this release before
1.1.1.  If you find any bugs, please submit a reproducible test case to
https://dev.eclipse.org (-> Bug Reports) (product AspectJ, component
Compiler for compiler bugs).

You can access the download at
 http://download.eclipse,org/technology/ajdt/aspectj-1.1.1rc1.jar.

There are the following known issues with this release:
* the changes file has not been updated - this will be done by 1.1.1
* there are 'half-a-dozen or so very minor broken docs links' that can 

be


repaired in time for 1.1.1

In addition:

A companion AJDT release (1.1.4) is available from the AJDT developer
update
site:

 http://download.eclipse.org/technology/ajdt/dev/update

There is the following known issue with this release of AJDT:
* Building projects that use linked source folders, and contain non-java
resource files causes an IOException when trying to copy the resources into the output folder. We will fix this ASAP and post an AJDT refresh ahead 

of


1.1.1.

Thanks,
Adrian, on behalf of The AspectJ Team



_______________________________________________
aspectj-dev mailing list
aspectj-dev@xxxxxxxxxxx
http://dev.eclipse.org/mailman/listinfo/aspectj-dev






_______________________________________________
aspectj-dev mailing list
aspectj-dev@xxxxxxxxxxx
http://dev.eclipse.org/mailman/listinfo/aspectj-dev

Back to the top