Re: [cdt-dev] cdt.ui.tests

[Peter Graves <pgraves@xxxxxxx>]

I will comment on any of the failures that I can inline below...

Camelon, John wrote:

> I select cdt.ui.tests and then go to the Run Menu and say Run As ->
> Junit
> plugin test. I am running on Windows XP w/JDK 1.3.1.
>
> On my development view, A bunch of UI stuff comes up from time to time
> to
> ask me about how different C++ UI stuff lays out and I for the most part
> just click "Yes" for that stuff. :-)  
I believe these are remnants of when Seb pulled over the jdt ui test 
suite to the cdt side. I have never really looked into these to see what 
they are trying to test, and likely we will want to prune them out, or 
at least not run them.

> I get the same 18 failures/3
> errors in
> what I get out of CVS vs. what I have in my development snapshot, so I
> am
> unsure as to when or how these tests got busted.  
>
> I get the following errors:
> - BinaryTests::testGetCpu() - NPE @ line 256 of BinaryTests.java
This used to pass for me. I will take a look and see if I can figure out 
what is happening.

> - AllBuildTests::testExtensions() - NPE @ line 159 of
> OptionReference.references()
> - AllBuildTests::testProject() - ArrayOutOfBoundsException @ line 179
> of
> AllBuildTests::checkRootProjects()
These are both form Douglas Shaefer so he would have to comment on them 
(I am not seeing these failures though)

> I get the following failures:
> - ElementDeltaTests::testElementDeltas() 
I am not seeing this failure. Alain originated this test case, and Doug 
S. has made a couple updated to it, so they may have a comment.

> - BinaryTests::testGetChildren()
> - BinaryTests::testGetNeededSharedLibs()
These 2 tests used to pass. The getChilden gets the list of children of 
a binary file, so if the parsing of binary files changed it could have 
caused this failures.  The GetNeededSharedLibs should not be failing, 
unless the binary.getNeededSharedLibs() method is not working anymore.

> - BinaryTests::testIsBinRead()
This test is based on the fact that an IBinary that was pointing to an 
executable used to always return true.  If this is not the case anymore 
then the test should be updated.

> - ArchiveTests::testGetBinaries()
I am not seeing this failure.


> - TranslationUnitTests::testGetElementAtLine() (comes up as 2 failures)
>  
>
> - TranslationUnitTests::testGetInclude() (comes up as 2 failures)
> - BinaryTests::testGetData()
>
> - BinaryTests::testGetText()
>  
>
> - TranslationUnitTests::testGetIncludes() (comes up as 2 failures)
>  
These tests are all related to open prs (look at the start of the 
failure message and it will list the pr number).

> - HelloWorld::test1()
This is just a sample test that came over with the origninal import.

> - PartitionTokenScannerTest::testTestCaseCRLF() (comes up as 2
> failures)
> - PartitionTokenScannerTest::testTestCase2CRLF() (comes up as 2
> failures)
>  
These are also fall overs from the original import Sebastien did, and 
looking at them they are both parsing java code, so I am guessing these 
could be pruned out of the tree.

> The tests that come up twice are likely due to the fact that they are
> also
> contained in another TestCase's suite() method.  
> Please let me know if these tests require additional setup or a
> different
> configuration.  I am particularly interested as to whether or not I have
> broken something in the CModel.
>
> JohnC
>
>  
>
> > -----Original Message-----
> > From: Alain Magloire [mailto:alain@xxxxxxx]
> > Sent: Monday, April 21, 2003 12:14 PM
> > To: cdt-dev@xxxxxxxxxxx
> > Subject: Re: [cdt-dev] cdt.ui.tests
> >
> >
> >    
> >
> > > When I run the whole UI Tests folder as a JUnit Plugin 
> > >      
> > test, I get 18
> >    
> >
> > > failures and 3 errors out of 383 runs (not bad).  I am 
> > >      
> > interested in running
> >    
> >
> > > the whole suite as a regression test for every patch that I 
> > >      
> > submit, and I am
> >    
> >
> > > wondering how new these failures are and whether or not the 
> > >      
> > tests should be
> >    
> >
> > > cleaned up or if new defects should be issued.  
> > >
> > > Can anybody provide some history on the matter? 
> > >
> > >      
> > Which one fails?
> > Peter Graves, did the initial framework, but he is not here 
> > today to sched some light.
> >
> >
> > _______________________________________________
> > cdt-dev mailing list
> > cdt-dev@xxxxxxxxxxx
> > http://dev.eclipse.org/mailman/listinfo/cdt-dev
> >
> >    
> _______________________________________________
> cdt-dev mailing list
> cdt-dev@xxxxxxxxxxx
> http://dev.eclipse.org/mailman/listinfo/cdt-dev
>