Expected failures/errors on SUnit

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
3 messages Options
Reply | Threaded
Open this post in threaded view
|

Expected failures/errors on SUnit

hernan.wilkinson
There is a tool we are writing that needs to stop running the tests of a test suite if there is an error or a failure (as a performance improvement)
To my surprise, using failuresCount and errorsCount made the tool run slower than faster and the reason is because of the expected failures and errors... asking for the failuresCount and the errorCounts goes through all the failures and error selecting only those that should not pass... anyway, it is a slow implementation. 
I'm not sure if this is the right list to ask this for, but does somebody know why is it implemented that way? could this be changed? who is the responsible for maintaining/changing SUnit?

Thanks,
Hernan.

_______________________________________________
Pharo-project mailing list
[hidden email]
http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project
Reply | Threaded
Open this post in threaded view
|

Re: Expected failures/errors on SUnit

Lukas Renggli
> know why is it implemented that way? could this be changed? who is the
> responsible for maintaining/changing SUnit?

Joseph Pelrine is the maintainer of SUnit
(http://sunit.sourceforge.net). The last official cross platform SUnit
3.1 release dates back to 2003. The implementation of SUnit in
Squeak/Pharo has independently changed over the years and has no
official maintainer, as far as I know.

Lukas

--
Lukas Renggli
http://www.lukas-renggli.ch

_______________________________________________
Pharo-project mailing list
[hidden email]
http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project
Reply | Threaded
Open this post in threaded view
|

Re: Expected failures/errors on SUnit

Stéphane Ducasse
In reply to this post by hernan.wilkinson
Please propose fixes and enhancements.
Have a look at the Test project extensions to SUnit made by keith.
I need somebody to have a look and tell us if this is ok to merge.

Stef

On Aug 10, 2009, at 3:29 AM, Hernan Wilkinson wrote:

> There is a tool we are writing that needs to stop running the tests  
> of a test suite if there is an error or a failure (as a performance  
> improvement)
> To my surprise, using failuresCount and errorsCount made the tool  
> run slower than faster and the reason is because of the expected  
> failures and errors... asking for the failuresCount and the  
> errorCounts goes through all the failures and error selecting only  
> those that should not pass... anyway, it is a slow implementation.
> I'm not sure if this is the right list to ask this for, but does  
> somebody know why is it implemented that way? could this be changed?  
> who is the responsible for maintaining/changing SUnit?
>
> Thanks,
> Hernan.
> _______________________________________________
> Pharo-project mailing list
> [hidden email]
> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project


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