Re: Keyed decoding of geometry

[Fred Kiefer]

Gregory John Casamento wrote:
> --- Fred Kiefer <address@hidden> wrote:
>
> > Gregory John Casamento wrote:
> >
> > > Does Apple use a separate method for this?   If not, why should we?
> >
> > To save me some typing.
> > Or if this is not sufficent for you (Which would be sad to hear), than 
> > yes, Apple does it the same way. At least they want me to spend my time 
> > with better things than typing the same lines over and over again.
>
>
> Fred, my reply was partially facetious.   Also, I don't understand your
> attitude/indignation, as I was asking a simple question which had *NOT* been
> addressed in your previous emails.  
>
> Apple, to my knowledge, does not split this out into two methods, the
> aforementioned initWithCoder: and initWithKeyedCoder:, as YOU proposed.   They
> test the NSCoder instance to see if it's capable of keyed coding in the
> initWithCoder: method.   This keeps all coding in one method, *where it
> belongs*.

Hi Gregory,

I did run into one of the most obvious traps of top posting with my 
reply. As you did put your answer before the original mail I was not 
sure which of the following paragraphs you did refer to and took the 
most obvious, eg the last one. Which was wrong as I can now tell from 
your more explicit mail. Richard did judge better and saw what you where 
aiming at.
Perhaps you could in the future make clear what you are refering to (ond 
stop top posting!) and I will try to ask back before making a guess? 
This surely would help our communication.

As for the original question, as I now understand it: Should we have a 
initWithKeyedCoder: method. I had offered a solution long before your 
mail. Is this something you could live with? The idea was start off with 
that method, to keep changes to the existing initWithCoder: method 
minimal and to decide later, if we keep this method or merge the code back.

Sorry for the misunderstanding

Fred