On Mon, Sep 21, 2009 at 1:44 PM, Alain Plantec <beranger22@gmail.com> wrote:
Hi Mariano,

strange, I have the same problem when I try
to load it cdirectly with the FileBrowser.
but, with MC it is loaded without any problem.


Excellent! I loaded it with MC now and worked perfect. I love this work. I just played a couple of minutes but seems an excellent work. We will finally have nice preference/settings browsers!!! wiiii
Alain

Mariano Martinez Peck a �crit :
> Alain: I just file in the changeset and I have a sintax error. I do it
> in latest Pharo.
>
> It doesn't let me copy paste, so I attach and screenshot.
>
> Best
>
> Mariano
>
> On Mon, Sep 21, 2009 at 11:36 AM, Alain Plantec <alain.plantec@free.fr
> <mailto:alain.plantec@free.fr>> wrote:
>
> � � just forgot to indicate the entry point.
> � � In order to open the setting browser:
> � � SystemSettingBrowser open.
>
> � � Cheers
> � � Alain
>
> � � Alain Plantec a �crit :
> � � > Hi all,
> � � > I've changed the new setting implementation.
> � � > The new proposition is attached.
> � � > I think that this �implementation is in sync with Lukas remarks;
> � � > and yes, I think it is much better now, thanks!
> � � > I also removed PragmaHandler and SettingHandler classes as Stephane
> � � > suggested.
> � � > It temporarily includes several (stupid) examples for testing
> � � purpose.
> � � > Comments have not yet been changed.
> � � > I will do it asap.
> � � > Remarks, comments etc are welcome.
> � � > Cheers
> � � > Alain
> � � >
> � � > Lukas Renggli a �crit :
> � � >>> - A setting with value is an instance of SettingValue or of
> � � one of its
> � � >>> subclasses. Such a setting holds a useful global system
> � � parameter and
> � � >>> is made persistent: it is stored in a class variable which is
> � � >>> initialized by the method that declares the setting.
> � � >>>
> � � >>
> � � >> I think the problem with this proposal is that it mixes the
> � � model (the
> � � >> value of the setting) and the meta-model (the model of the
> � � setting).
> � � >> This is problematic because then all code depends on the settings
> � � >> package and when reading/writing settings application code must be
> � � >> aware of the indirection introduced through the setting.
> � � >>
> � � >> Why isn't a setting simply a variable local to the package, and the
> � � >> setting-description a separate object (method) that can also be
> � � >> packaged separately?
> � � >> � Lukas
> � � >>
> � � >>
> � � >
> � � >
> � � ------------------------------------------------------------------------
> � � >
> � � > _______________________________________________
> � � > Pharo-project mailing list
> � � > Pharo-project@lists.gforge.inria.fr
> � � <mailto:Pharo-project@lists.gforge.inria.fr>
> � � > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project
>
>
> � � _______________________________________________
> � � Pharo-project mailing list
> � � Pharo-project@lists.gforge.inria.fr
> � � <mailto:Pharo-project@lists.gforge.inria.fr>
> � � http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project
>
>
>
> ------------------------------------------------------------------------
>


_______________________________________________
Pharo-project mailing list
Pharo-project@lists.gforge.inria.fr
http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project