On 9/29/10, Alexander Lazarević <laza at blobworks.com> wrote:
> Lukas Renggli and Yanni Chiu did all the work to build a very flexible
> framework and to provide good documentation. For me it was just a
> matter of seeing how the bits and pieces work together and write some
> Squeak specific scripts.

This is called 'Installation' and 'System Integration' and 'System
Integration Testing'   :-)

I mean it is real work.....
Of course you build up your work on other's work.

Many thanks - this is an important contribution and I think will help
us to save time and get more convenience and reliability.

And a good reuse of work done by the Pharo people.

> See:
>
> Yannis Hudson Server at http://hudson.jooshr.org
> Lukas Hudson Server at http://hudson.lukas-renggli.ch
> Sources and Doku at http://github.com/renggli/builder
> (MagLev seems to be using it too http://github.com/MagLev/builder)
>
> When I get knack out of github I will try to fork and upload the few
> changes I made
>
OK

BTW which server OS did you use in your VirtualMachine onto which you
installed Hudson? E.g. CentOS? Which OS does the Squeak server use?

--Hannes

> Alex
>
> 2010/9/29 Enrico Spinielli <enrico.spinielli at googlemail.com>:
>> Very well done!
>> Bravo!
>> Bye
>> Enrico
>>
>> On Wed, Sep 29, 2010 at 06:08, Alexander Lazarević <laza at blobworks.com>
>> wrote:
>>> Ok, there you go. Testresults are in. Haven't looked at the lint stuff
>>> yet. But I guess it would be ok if this moves from my laptop to some
>>> squeak server. This is if there are plans to have a 4.2 at all ...
>>>
>>> Alex
>>>
>>> 2010/9/28 Bert Freudenberg <bert at freudenbergs.de>:
>>>>
>>>> On 28.09.2010, at 18:03, Alexander Lazarević wrote:
>>>>
>>>>> At least some progress :)
>>>>>
>>>>> Build #5 was triggered by my commit to the repository, but it failed
>>>>> because of a data timeout.
>>>>
>>>> Well, nice progress anyway :)
>>>>
>>>>> It is pretty slick to trigger a build only if there is a commit, but
>>>>> the URL-Trigger on the squeak source repository acts immediately if
>>>>> there is a change and so the build might start while there are some
>>>>> commits still in progress. I guess a scheduled daily build would be a
>>>>> little more robust choice.
>>>>
>>>> Daily would be sufficient I think. Or start a build when there was no
>>>> commit within, say, 30 minutes.
>>>>
>>>>> Next thing I will have a look at, as time permits, is how to run the
>>>>> tests.
>>>>
>>>> Way to go!
>>>>
>>>> - Bert -
>>>>
>>>>
>>>>
>>>>
>>>
>>>
>>>
>>>
>>
>>
>>
>> --
>> Enrico Spinielli
>> "Do Androids dream of electric sheep?"— Philip K. Dick
>> "Hear and forget; see and remember;do and understand."—Mitchel Resnick
>>
>>
>
>