> > Hi,
> >
> > I'm having a look at using the refreshCommand of the Property interface.
> > I'm not sure if it still exists or if it will remain in the current RTT
> > development. Anyway here is a piece of code from Proprty.h
> >
> >        virtual CommandInterface* refreshCommand( const PropertyBase*
> > other)
> >        {
> >            if ( !_value )
> >                return 0;
> >            // refresh is just an update of the datasource.
> >            DataSourceBase::shared_ptr sourcebase =
> > other->getDataSource();
> >            return _value->updateCommand( sourcebase.get() );
> >        }
> >
> > Here you see the refreshCommand is forwarded to the update command.
> > Witch eventually updates the property. Is this intentional? Update and
> > refresh have a different meaning. I'm not interested in update(), since
> > it also updates the description (which leads to memory allocation during
> > the string copy).
>
> it is forwarded to updateCommand of the DataSourceBase class, which is
> different from updateCommand in the PropertyBase class. The former updates
> the value (there is no description in a DataSource), the latter the
> value and the description.
>
Ok, I overlooked the DataSource. But to be of any use I prefer the use of the Properties refreshCommand because I override the Property::refresh() method.

> So the code is correct. I'm not sure yet what the future is of these
> family
> of functions. That will be discussed in a few months, when we're through
> the
> data flow refactoring.
>
I think they are valuable in some use cases. We already use a command based update of properties.

Sander.