On Sep 2, 2011, at 8:34 PM, laurent laffont wrote:

> 
> On Fri, Sep 2, 2011 at 8:57 AM, Stéphane Ducasse <[email protected]> 
> wrote:
> could be interesting. But it would just hides our poor comment quality.
> I prefer to see that I'm poor than pretend I'm rich :)
> 
> I agree. But sometimes there's so few comments in a package that's I don't 
> bother search for one. It should be easy to see classes that have comments in 
> the browser (icon ? color ?). 

indeed.

Stef
> 
> Laurent.
>  
> 
> For the RBRule what would be great (and this was on my overly long todo) was 
> a improve the description.
> May be we should have a short and long one. Because something I do not 
> understand (well I do most of the times) but
> I would love to get a nice description because we really want to push the 
> rules usage.
> 
> We got a postdoc with chili (yesterday - hello and welcome angel :) ) and we 
> want him to help building better infrastructure around rules
>        - book chapter on rule engine
>        - storing metadata because it does not scale to add pragmas in method 
> (we probably need a package manifesto).
>        - analyzing the rules and their application to pharo and pharo projects
>        - get rules automatically run with jenkins
> 
> Stef
> 
> On Sep 2, 2011, at 8:27 AM, laurent laffont wrote:
> 
> > On Thu, Sep 1, 2011 at 10:08 PM, Lukas Renggli <[email protected]> wrote:
> > >> Today:  RBSubclassResponsibilityNotDefinedRule
> > >
> > > This rules identifies when several methods in a set of subclasses are 
> > > defined but not in the superclass.
> > > It identifies that an abstract method is missing in the superclass.
> >
> >    Checks that all subclassResponsibility methods are defined in all
> > leaf classes.
> >
> > Note that each of the 120 subclasses of RBLintRule has a method called
> > #rationale that describes its intent. There are even tests that verify
> > this. Of course the string is not a class comment, but something to be
> > displayed in the code critics browser. Still I note a certain degree
> > of overlap ...
> >
> >
> > I wonder, we may just comment RBLintRule.  Then wouldn't it be nice that 
> > class browser, when displaying a class without comment, goes up in the 
> > hierarchy up to Object (excluded) to get the first comment found ?
> >
> >
> > Laurent.
> >
> >
> >
> > Lukas
> >
> > --
> > Lukas Renggli
> > www.lukas-renggli.ch
> >
> >
> 
> 
> 


Reply via email to