Issue 5507 in pharo: test RGClassDefinitionTest >> testWithComment failure / TimeStamp comparison

Status: FixReviewNeeded
Owner: [hidden email]

New issue 5507 by [hidden email]: test RGClassDefinitionTest >>  
testWithComment failure / TimeStamp comparison
http://code.google.com/p/pharo/issues/detail?id=5507

The test RGClassDefinitionTest >> testWithComment is failing. The reason  
lies in comparison of a DateAndTime and TimeStamp. TimeStamp is subclass of  
DateAndTime but = method in DateAndTime compares firstly species that is  
different (TimeStamp for TimeStamp instances and DateAndTime for  
DateAndTime instances). TimeStamp deoesn't add any instance variable.

This fix changes species of TimeStamp to DateAndTime.

Attachments:
        TimeStamp-species.st  163 bytes


_______________________________________________
Pharo-bugtracker mailing list
[hidden email]
http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-bugtracker

Comment #1 on issue 5507 by [hidden email]: test  
RGClassDefinitionTest >> testWithComment failure / TimeStamp comparison
http://code.google.com/p/pharo/issues/detail?id=5507

RGCommentDefinitionTest fails because of the same reason.


_______________________________________________
Pharo-bugtracker mailing list
[hidden email]
http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-bugtracker

Updates:
        Labels: Type-Bug Milestone-1.4

Comment #2 on issue 5507 by [hidden email]: test  
RGClassDefinitionTest >> testWithComment failure / TimeStamp comparison
http://code.google.com/p/pharo/issues/detail?id=5507

(No comment was entered for this change.)


_______________________________________________
Pharo-bugtracker mailing list
[hidden email]
http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-bugtracker

Updates:
        Status: FixToInclude

Comment #3 on issue 5507 by [hidden email]: test  
RGClassDefinitionTest >> testWithComment failure / TimeStamp comparison
http://code.google.com/p/pharo/issues/detail?id=5507

(No comment was entered for this change.)


_______________________________________________
Pharo-bugtracker mailing list
[hidden email]
http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-bugtracker

Updates:
        Status: Integrated

Comment #4 on issue 5507 by [hidden email]: test  
RGClassDefinitionTest >> testWithComment failure / TimeStamp comparison
http://code.google.com/p/pharo/issues/detail?id=5507

in 14400


_______________________________________________
Pharo-bugtracker mailing list
[hidden email]
http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-bugtracker

Comment #5 on issue 5507 by [hidden email]: test RGClassDefinitionTest  
>> testWithComment failure / TimeStamp comparison
http://code.google.com/p/pharo/issues/detail?id=5507

I think that fix is not needed..  And it is weird because it does not fail  
in my pharo1.4 image.

Stef proposed an enhancement in Ring where he changed the timeStamp value  
of a method and comment from a DataAndTime to a TimeStamp object.  But it  
seems the tests were not changed.

I would just update both tests and not change the species of TimeStamp.

Attachments:
        Ring-timeStamp.cs  1.7 KB


_______________________________________________
Pharo-bugtracker mailing list
[hidden email]
http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-bugtracker

Updates:
        Status: FixReviewNeeded

Comment #6 on issue 5507 by [hidden email]: test  
RGClassDefinitionTest >> testWithComment failure / TimeStamp comparison
http://code.google.com/p/pharo/issues/detail?id=5507

(No comment was entered for this change.)


_______________________________________________
Pharo-bugtracker mailing list
[hidden email]
http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-bugtracker

Comment #7 on issue 5507 by [hidden email]: test  
RGClassDefinitionTest >> testWithComment failure / TimeStamp comparison
http://code.google.com/p/pharo/issues/detail?id=5507

Veronica the problems is that we do not understand the tests and why they  
are failing.



_______________________________________________
Pharo-bugtracker mailing list
[hidden email]
http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-bugtracker

Comment #8 on issue 5507 by [hidden email]: test RGClassDefinitionTest  
>> testWithComment failure / TimeStamp comparison
http://code.google.com/p/pharo/issues/detail?id=5507

when do they fail?
I just loaded a new 1.4 image, and I ran both test without changes, without  
the new species in TimeStamp, and after changing the tests (as the file I  
attached before)... they are always green!!

What I see is that the = method in DateAndTime was recently modified...  
maybe the problem is coming from that side


_______________________________________________
Pharo-bugtracker mailing list
[hidden email]
http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-bugtracker

Comment #9 on issue 5507 by [hidden email]: test  
RGClassDefinitionTest >> testWithComment failure / TimeStamp comparison
http://code.google.com/p/pharo/issues/detail?id=5507

They are not failing anymore after I added the TimeStamp-species.st  (I  
think I removed the #species before, it looked to me that it was the same  
as the superclass but I was wrong).


_______________________________________________
Pharo-bugtracker mailing list
[hidden email]
http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-bugtracker

Comment #10 on issue 5507 by [hidden email]: test  
RGClassDefinitionTest >> testWithComment failure / TimeStamp comparison
http://code.google.com/p/pharo/issues/detail?id=5507

Yes, as I noted in issue 5502, the species method in DateAndTime is still  
needed, as per Issue 2872.

I suggest reverting to use the fix from Issue 2872


_______________________________________________
Pharo-bugtracker mailing list
[hidden email]
http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-bugtracker

Comment #11 on issue 5507 by [hidden email]: test  
RGClassDefinitionTest >> testWithComment failure / TimeStamp comparison
http://code.google.com/p/pharo/issues/detail?id=5507

Which, by the way, changes DateAndTime species rather than Timestamp  
species.


_______________________________________________
Pharo-bugtracker mailing list
[hidden email]
http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-bugtracker

Comment #12 on issue 5507 by [hidden email]: test  
RGClassDefinitionTest >> testWithComment failure / TimeStamp comparison
http://code.google.com/p/pharo/issues/detail?id=5507

The reason for that being mostly semantic, in that other potential  
subclasses would have an equivalent issue with the definition of  
DateAndTime >> #=



_______________________________________________
Pharo-bugtracker mailing list
[hidden email]
http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-bugtracker

Comment #13 on issue 5507 by [hidden email]: test  
RGClassDefinitionTest >> testWithComment failure / TimeStamp comparison
http://code.google.com/p/pharo/issues/detail?id=5507

Now I am completely confused. (In general, this is why code in addition to  
verbal descriptions is so valuable...)


_______________________________________________
Pharo-bugtracker mailing list
[hidden email]
http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-bugtracker

Updates:
        Status: Workneeded

Comment #14 on issue 5507 by [hidden email]: test  
RGClassDefinitionTest >> testWithComment failure / TimeStamp comparison
http://code.google.com/p/pharo/issues/detail?id=5507

(No comment was entered for this change.)


_______________________________________________
Pharo-bugtracker mailing list
[hidden email]
http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-bugtracker

Updates:
        Labels: -Milestone-1.4 Milestone-1.5

Comment #15 on issue 5507 by [hidden email]: test  
RGClassDefinitionTest >> testWithComment failure / TimeStamp comparison
http://code.google.com/p/pharo/issues/detail?id=5507

(No comment was entered for this change.)


_______________________________________________
Pharo-bugtracker mailing list
[hidden email]
http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-bugtracker

Updates:
        Labels: -Milestone-2.0 Milestone-3.0

Comment #18 on issue 5507 by [hidden email]: test  
RGClassDefinitionTest >> testWithComment failure / TimeStamp comparison
http://code.google.com/p/pharo/issues/detail?id=5507

I think we will do this in 3.0


_______________________________________________
Pharo-bugtracker mailing list
[hidden email]
http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-bugtracker