junit test suite

[Matthieu Bollot-4]

Hi devs,

We (at nereide) are starting to make a continuous integration server,
the purpose is to have seleniums and junit tests ran daily to react
quickly in a near future.
This night, from 2 failures, we now have 29 failures and 6 errors (I'm
obviously talking about ofbiz trunk)

Is it something well-known ? Does a lot of people take care about it ?

Cheers,
Matthieu.


PS : if you are not aware with junit tests, all you need to do is ant
run-tests, and open runtime/logs/test-results/html/index.html, or see
our latest results here :
http://selenium.neogia.org/ofbizAddOn/ofbiz/runtime/logs/test-results/html/index.html

[Scott Gray-2]

Hi Matthieu,

I'm not sure if you've been watching the commit logs but I'm actively  
working my way through the junit tests and the moment and getting them  
all back to passing again.

I haven't used selenium yet but my results for the junit tests have 0  
errors and 12 failures.

Getting some continuous integrations setup is a great idea, hopefully  
we can get something setup on the apache infrastructure in the not too  
distant future and then reach an agreement among the committers (and  
contributors) that the tests should never be allowed to fail (or at  
least not for very long).

Regards
Scott

HotWax Media
http://www.hotwaxmedia.com

On 18/11/2009, at 2:18 AM, Matthieu Bollot wrote:

> Hi devs,
>
> We (at nereide) are starting to make a continuous integration server,
> the purpose is to have seleniums and junit tests ran daily to react
> quickly in a near future.
> This night, from 2 failures, we now have 29 failures and 6 errors (I'm
> obviously talking about ofbiz trunk)
>
> Is it something well-known ? Does a lot of people take care about it ?
>
> Cheers,
> Matthieu.
>
>
> PS : if you are not aware with junit tests, all you need to do is ant
> run-tests, and open runtime/logs/test-results/html/index.html, or see
> our latest results here :
> http://selenium.neogia.org/ofbizAddOn/ofbiz/runtime/logs/test-results/html/index.html
>

[Tim Ruppert]

+1 - as well as hopefully we can have a policy around fixing bugs -  
that when you fix them - that we put a test in to catch what's being  
fixed!

Cheers,
Ruppert
--
Tim Ruppert
HotWax Media
http://www.hotwaxmedia.com

o:801.649.6594
f:801.649.6595

On Nov 17, 2009, at 1:50 PM, Scott Gray wrote:

> Hi Matthieu,
>
> I'm not sure if you've been watching the commit logs but I'm  
> actively working my way through the junit tests and the moment and  
> getting them all back to passing again.
>
> I haven't used selenium yet but my results for the junit tests have  
> 0 errors and 12 failures.
>
> Getting some continuous integrations setup is a great idea,  
> hopefully we can get something setup on the apache infrastructure in  
> the not too distant future and then reach an agreement among the  
> committers (and contributors) that the tests should never be allowed  
> to fail (or at least not for very long).
>
> Regards
> Scott
>
> HotWax Media
> http://www.hotwaxmedia.com
>
> On 18/11/2009, at 2:18 AM, Matthieu Bollot wrote:
>
>> Hi devs,
>>
>> We (at nereide) are starting to make a continuous integration server,
>> the purpose is to have seleniums and junit tests ran daily to react
>> quickly in a near future.
>> This night, from 2 failures, we now have 29 failures and 6 errors  
>> (I'm
>> obviously talking about ofbiz trunk)
>>
>> Is it something well-known ? Does a lot of people take care about  
>> it ?
>>
>> Cheers,
>> Matthieu.
>>
>>
>> PS : if you are not aware with junit tests, all you need to do is ant
>> run-tests, and open runtime/logs/test-results/html/index.html, or see
>> our latest results here :
>> http://selenium.neogia.org/ofbizAddOn/ofbiz/runtime/logs/test-results/html/index.html
>>
>

[Jacques Le Roux]

From: "Scott Gray" <[hidden email]>

> Hi Matthieu,
>
> I'm not sure if you've been watching the commit logs but I'm actively  
> working my way through the junit tests and the moment and getting them  
> all back to passing again.
>
> I haven't used selenium yet but my results for the junit tests have 0  
> errors and 12 failures.
>
> Getting some continuous integrations setup is a great idea, hopefully  
> we can get something setup on the apache infrastructure in the not too  
> distant future and then reach an agreement among the committers (and  
> contributors) that the tests should never be allowed to fail (or at  
> least not for very long).

+1

Thanks for your work on this Scott.
Knowing that the Neogia team will check it is also good...

Jacques
 

> Regards
> Scott
>
> HotWax Media
> http://www.hotwaxmedia.com
>
> On 18/11/2009, at 2:18 AM, Matthieu Bollot wrote:
>
>> Hi devs,
>>
>> We (at nereide) are starting to make a continuous integration server,
>> the purpose is to have seleniums and junit tests ran daily to react
>> quickly in a near future.
>> This night, from 2 failures, we now have 29 failures and 6 errors (I'm
>> obviously talking about ofbiz trunk)
>>
>> Is it something well-known ? Does a lot of people take care about it ?
>>
>> Cheers,
>> Matthieu.
>>
>>
>> PS : if you are not aware with junit tests, all you need to do is ant
>> run-tests, and open runtime/logs/test-results/html/index.html, or see
>> our latest results here :
>> http://selenium.neogia.org/ofbizAddOn/ofbiz/runtime/logs/test-results/html/index.html
>>
>
>

[Matthieu Bollot-4]

Le mercredi 18 novembre 2009 à 09:50 +1300, Scott Gray a écrit :
> Hi Matthieu,
>
> I'm not sure if you've been watching the commit logs but I'm actively  
> working my way through the junit tests and the moment and getting them  
> all back to passing again.
I read them quickly, I know "sometimes" people are working on them but didn't know if it was something important for ofbiz dev or not.

> I haven't used selenium yet but my results for the junit tests have 0  
> errors and 12 failures.
Seleniums are fine, don't worry at the moment ;)

> Getting some continuous integrations setup is a great idea, hopefully  
> we can get something setup on the apache infrastructure in the not too  
> distant future and then reach an agreement among the committers (and  
> contributors) that the tests should never be allowed to fail (or at  
> least not for very long).

What you both said is good to know, and of course we will try to help as
much as we can.

Cheers,
Matthieu.