Hi Stephan,

I agree for the documentation improvements. Many methods returning a
modifiable list where already documented so it wasn't a pain to make sure
all the remaining ones were documented as well.

This is done in SVN trunk.

Best regards,
Jerome  

> -----Message d'origine-----
> De : [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] 
> Envoyé : samedi 2 février 2008 18:31
> À : [email protected]
> Objet : Re: Adding Cookies to a Client Request
> 
> Hi,
> 
> perhaps we could add a comment to the method setCookies and the other 
> Collection setters that the developer should use the getter 
> to get the 
> Collection and add the elements to the collection directly. I 
> know, that 
> this is a lot of work because there are a lot of classes with 
> a lot of 
> Collections. But this could be a process: Only if someone is anyway 
> working on a class, than he add the comments to this class.
> So no one must go into all classes and add the comments. 
> Perhaps a start 
> could be made with the Request and the Response class and the other 
> classes are processed later.
> 
> best regards
>    Stephan
> 
> Jerome Louvel schrieb:
> > Hi all,
> >
> > I'm generally open to compromises but if we have a 
> "addCookie" method,
> > newcomers will definitely ask a "getCookie" one, and as 
> several cookies can
> > have the same name, we'll open the Pandora box.
> >
> > This is the whole purpose of the org.restlet.util.Series<E> 
> class, to allow
> > getting the first cookie with a given name, or all cookie 
> having the same
> > name, etc.
> >
> > For now, I prefer imposing an additional learning step to 
> new users rather
> > than giving the illusion of simplicity when the data 
> structure is truly
> > complex. We also need to care for consistency across the 
> whole API, which
> > has a high priority for me.
> >
> > Let's see if this problem comes back again and again in the 
> future. We can
> > also reconsider the whole approach for Restlet 2.0 if you 
> want, maybe there
> > are radically different ways to solve this problem.
> >
> > Best regards,
> > Jerome

Reply via email to