On 7 July 2013 23:11, Camillo Bruni <[email protected]> wrote:
> I will change SUnit first produce proper error messages for the
> #shouldnt:raise: methods since I cannot properly debug them.
> Additionally at the point where the #:assert: happens the wrong exception is
> not in the context which means it does not get properly serialized by our
> test framework. All in all, with a little effort we could debug much easier.
I found this a big improvement:
TestCase >> shouldnt: aBlock raise: anExceptionalEvent
^ [ aBlock value ]
on: anExceptionalEvent
do: [:e | self fail: 'Block raised ', e className, ':
', e messageText].
TestCase >> fail: aString
^self assert: false description: aString.
frank
> BTW: There are so many strange methods on TAssertable that I do not understand
> - #shouldFix:
> - #executeShould:inScopeOf:
>
> and some that I honestly think they hurt more than they help:
> - #shouldnt:raise:whoseDescriptionIncludes:description: and friends?
>
> On 2013-07-07, at 23:47, "Dale K. Henrichs"
> <[email protected]> wrote:
>
>> Cami,
>>
>> I've been having trouble with getting any useful information from Pharo3.0
>> failures on travis-ci[1] for a week or so now ... Christophe will be looking
>> into this for me...
>>
>> Dale
>>
>> [1] https://github.com/dalehenrich/filetree/issues/80#issuecomment-20162125
>> ----- Original Message -----
>> | From: "Camillo Bruni" <[email protected]>
>> | To: [email protected]
>> | Sent: Sunday, July 7, 2013 2:00:10 PM
>> | Subject: [Pharo-dev] filetree jenkins job
>> |
>> | after I tried to get filetree with the tests installed and I didn't
>> | get it
>> | the first time I decided to create a jenkins job.
>> |
>> | https://ci.inria.fr/pharo-contribution/job/FileTree/
>> |
>> |
>>
>
>
