>> It is still one of the todo items for the refactoring engine to make
>> Code Critics completely runnable on the refactoring code model. As the
>> Pharo code model is diverging more and more from the refactoring code
>> model this is more likely to happen anytime soon.
>
> Lukas what do you say that. Fixing the API of the RB code model should be 
> simple to do.
> Especially since the code model does not have to run on all the smalltalk 
> dialect so we are
> free to fix it.

That is not entirely true, Slime and Magritte Critics rules (at least
used to) run on other dialects too. Also there is external code that
depends on it, such as Helvetia, Opal, ST2JS, SmaCC, ...

The reason for moving Code Critics to use the refactoring model is
that there will be only a single interaction point with the base
system (Pharo). This will make maintenance simpler as only a single
place needs to be updated (BrowserEnvironment for read-reflection,
RefactoryChange for write-reflection).

I don't know why Code Critics does not use the refactoring code model?
It probably has historical reasons: Maybe it was a separate codebase
at some point? Maybe it was done for performance reasons?

> In addition the RB code model has always been a wrapper on top of other.

No, the refactoring code model is not a wrapper: it doesn't
necessarily wrap something.

Lukas

-- 
Lukas Renggli
www.lukas-renggli.ch

Reply via email to