>> The point is that we should be able to have good tools inside the default 
>> system so that we can get faster.
>> We are fedup to do refactoring of core manually.
> 
> You have a build server that builds PharoCore and Pharo images
> immediately and all the time. So, why are you not improving the system
> in Pharo and immediately let your servers build experimental PharoCore
> and Pharo images right after you commit? If tests pass the files get
> propagated and generate the official builds.


The point is that when I work with core right now we do not have RB loaded. So 
a pain ;)
Now if we work on Pharo (RB loaded) and change RB (we will have to do that for 
RPackage for example).
How do I publish, reload everything? I want to close the loop because it should 
be possible and that jenkins does all the job for me. 

        Right now scriptLoader does not deal with external packages and 
dependencies
        So we could do it but we will lose all the effort in making these 
external packages loadable outside
        of the "pharo" image.
        So this is why I proposed. 
        
>       seed + MetacelloSpec + modification => seed' + MetacelloSpec'
> 
>       and that we are able to build hudson that takes
>               aSeed + a spec
>                       => run all the tests + run all the quality check
> 
> Right now we have
>       seed + list of package + modification => seed' + list of package'
> 
>       and we get in trouble when we modify seed parts that influence external 
> package

        This way: pharo hardcore like us can work with the tools at full speed.
        And other people can make sure that they can load the tools we use and 
modified in the different 
        version of the images using Metacello and the dependencies management 
(the scary part of packages for me).

> As long as Pharo packages are clean without any overrides there should
> be no problem (this is the case with my builds).

Yes more or less because dependencies between versions will be evil and kill us 
if we do not pay attention. 

> If you remember, we did together some refactorings while I was visiting you 
> in Lille last year.

Yes but I got stuck :( 
I should invest throwing away scriptLoader :)
Mariano, Guillermo I would love to have a Metacello conf for Pharo (old 
pharo-core).
And that the system generate it for me. 

Ok I will invest in that....
Argh. Now I spent 1/3 of my time in meetings (as lab representent reagrh....).

Stef (I will survive).


> 
> Lukas
> 
> -- 
> Lukas Renggli
> www.lukas-renggli.ch
> 


Reply via email to