>> where are you drawing the line between an
>> "extension", and just a "visitable" object? What "visitable" objects are
>> you thinking other than "extensions"?
>
> I think on viewer components registered by extensions at NodeView's
> content pane. They should not be referenced by their class name. That's
> why many elements of the same class can belong to the same
> VisitableCollection. (If want we have multiple views for one map, we are
> likely to have multiple viewers per node and extension property).
Didn't quite get the kind of answer I was looking for, but lets see. For  
the rest of the class diagrams I could compare things with existing code  
base where things weren't being too clear. Here I can't do that. So lets  
see what happens as the details get filled in over time. Things should get  
clearer for me.

>
>> Are you planning to go into a refactoring phase before this new design 
>> is implemented? [...]
>
> I would like to make as small steps as possible. I do not believe in
> final design, I believe in iterations. And I am no sure whether we
> should introduce additional complexity right now. IMHO after the
> refactoring is done, further changes can be introduced easier than now.
>
Na no final designs are coming up in open-source projects thats for sure.  
But the thing I was worried about, that during refactoring phase some  
other features do get added along as well ... so changes like these might  
as well be implemented as early on as possible before more code gets to  
depend on such things. but if things are purely/mostly refactoring, yes  
things can actually become easier.

> I mean methods getSelectedXXX() and setSelectedXXX(...). (I have
> intentionally used other name for stimulating a discussion about what
> name is better: getActiveXXX() or getSelectedXXX() .
Honestly, to me, either one is good. After an hour of reading through the  
API, I'd be used to either one. Probably not the answer you were looking  
for, but sorry :(

>
>> Right :-) Actually I realized this part when I was done writing the  
>> first
>> half, so I went on to add the 2nd half saying that such calls in  
>> MindMaps
>> should probably just wrap the ones provided by ModeController.
>>
>> [...]
>
> Agreed.
>
Thanks.

Reasamp

-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
Freemind-developer mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/freemind-developer

Reply via email to