Re: [ANNOUNCE] Tk::Gripes and Tk::SignalSlot

--- Nick Ing-Simmons <nick@xxxxxxxxxxxxxxx> wrote:

> Well :
>    A. There is nothing stopping that callback
> calling others.

Correct, but that adds unnecessary coupling between
otherwise unrelated pieces of code. It's not a real
problem, but can lead to some complicated code. For
Tcl/Tk, which is procedural, this is not an issue. But
for Perl/Tk, which has an object-oriented syntax, I
believe signals and slots can result in cleaner code
most of the time.

>    B. It isn't true
>         bind associates a callback with a tag 
>         when an event occurs Tk calls the callbacks
> - in sequence 
>         - for all the bindtags associate with the
> widget via bindtags.
> 
> So if you are writing a mega-widget you should bind
> to the class name
> of that implements it. And leave the specific
> instance name tag for 
> end-user use.
> 
> There is even a mechanism for a derived class to
> inherit bindings 
> from its base class(es)!

True. I wasn't thinking about bindtags. While very
powerful, bindtags can be confusing, and a pain to
work with. I haven't seen a similar concept in any
other toolkit. Perhaps I need to modify the article,
but I still believe that signals and slots are a very
convenient alternative to bindings.

--Ala


__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 
-++**==--++**==--++**==--++**==--++**==--++**==--++**==
This message was posted through the Stanford campus mailing list
server.  If you wish to unsubscribe from this mailing list, send the
message body of "unsubscribe ptk" to majordomo@xxxxxxxxxxxxxxxxxx