Thats what I meant. Params are (theoretically) stored on the stack in
C I don't know about in Cocoa.

8 Jun 2005, at 16:13, Clark Cox wrote:

6/8/05, Theodore H. Smith <delete (AT) elfdata (DOT) comwrote:
>
>>
>Am I right in thinking that self is a local (stack based) variable?
>>
Yes and no. Self is actually a hidden function parameter. For
instance, the method:

Do not post admin requests to the list. They will be ignored.
Cocoa-dev mailing list (Cocoa-dev (AT) lists (DOT) apple.com)
Help/Unsubscribe/Update your Subscription:
%40developershed.com

This email sent to bsdarchive (AT) developershed (DOT) com

6/8/05, Theodore H. Smith <delete (AT) elfdata (DOT) comwrote:

Am I right in thinking that self is a local (stack based) variable?

for example the code:
-(id) init {
self = [self initSpecial];
return self;
}

should alter everything in exactly the same way as:
-(id) init {
return [self initSpecial];
}

The short answer is Yes, kind of.

self is the hidden first parameter to the method and will be stored in
a register on PPC. As with any parameter or other local variable
assigning to it will have no effect outside the scope of the method
and the two pieces of code are therefore equivalent.

Do not post admin requests to the list. They will be ignored.
Cocoa-dev mailing list (Cocoa-dev (AT) lists (DOT) apple.com)
Help/Unsubscribe/Update your Subscription:
%40developershed.com

This email sent to bsdarchive (AT) developershed (DOT) com

Jun 8, 2005, at 10:59 AM, Theodore H. Smith wrote:
Am I right in thinking that self is a local (stack based) variable?

It is an implicit argument to every method, so yes it is local (i.e.,
its scope is the method) and yes it is stack-based. Technically you
can treat it like any other method argument, including setting it to
whatever value you want. You should almost never do this outside of
an init method, because it can make your code confusing and error-
prone to have "self" not be the original receiver of the method.
Some people feel init methods are a special case because of their
unique semantics -- hence this idiom:
- (id)init {
if ((self = [super init]))
{
//
}

return self;
}

for example the code:

-(id) init {
self = [self initSpecial];
return self;
}

should alter everything in exactly the same way as:

-(id) init {
return [self initSpecial];
}

Yes, those two code examples have identical effects.

Do not post admin requests to the list. They will be ignored.
Cocoa-dev mailing list (Cocoa-dev (AT) lists (DOT) apple.com)
Help/Unsubscribe/Update your Subscription:
%40developershed.com

This email sent to bsdarchive (AT) developershed (DOT) com

Theodore,

8.6.2005, at 16:59, Theodore H. Smith wrote:

Am I right in thinking that self is a local

Completely.

(stack based) variable?

Depends on how the compiler feels at the moment :) -- it might keep
it in the register it comes in (for it actually is the first
argument, $r3).

At least, for today, whilst we still run on a decent architecture

for example the code:

-(id) init {
self = [self initSpecial];
return self;
}
should alter everything in exactly the same way as:
-(id) init {
return [self initSpecial];
}

Exactly. Note also that the former is *needed* in case of more
complex initialization:
-init {
self=[self initSpecial]; // also may need to check for nil
depending on the following code
[self whatever];
return self;
}

ada
CSoftware: ocs (AT) ocs (DOT) cz http://www.ocs.cz
private ondra (AT) ocs (DOT) cz http://www.ocs.cz/oc

Do not post admin requests to the list. They will be ignored.
Cocoa-dev mailing list (Cocoa-dev (AT) lists (DOT) apple.com)
Help/Unsubscribe/Update your Subscription:
%40developershed.com

This email sent to bsdarchive (AT) developershed (DOT) com

Jun 8, 2005, at 8:13 AM, Clark Cox wrote:

>
>for example the code:
>>
>-(id) init {
>self = [self initSpecial];
>return self;
>}
>>
>should alter everything in exactly the same way as:
>>
>-(id) init {
>return [self initSpecial];
>}
>>
>
Yes.

Maybe.

If initSpecial calls [super init], some supers discard the original
object and return a new one.

I am fairly certain; I haven't tested it so YMMV, but the
NSDictionary returned by initWAndKeys is not the one you
alloc'ed.
-Mark
Duck, duck, duck, duck, GSE!

Do not post admin requests to the list. They will be ignored.
Cocoa-dev mailing list (Cocoa-dev (AT) lists (DOT) apple.com)
Help/Unsubscribe/Update your Subscription:
%40developershed.com

This email sent to bsdarchive (AT) developershed (DOT) com

Jun 8, 2005, at 11:39 AM, Mark Lively wrote:
Jun 8, 2005, at 8:13 AM, Clark Cox wrote:
for example the code:

-(id) init {
self = [self initSpecial];
return self;
}

should alter everything in exactly the same way as:

-(id) init {
return [self initSpecial];
}


>>
>Yes.
>>
>
Maybe.

If initSpecial calls [super init], some supers discard the original
object and return a new one.

I think you're confusing this with a different init pitfall. The
above two methods are in fact identical. In the first, *whatever* -
initSpecial returns (even if it's not the original receiver) will be
assigned to the local variable self, and that value will be
returned. In the second, *whatever* -initSpecial returns will be
returned immediately.

You might be thinking of the mistake some people make when they
separate the call to +alloc from the call to -init.

Do not post admin requests to the list. They will be ignored.
Cocoa-dev mailing list (Cocoa-dev (AT) lists (DOT) apple.com)
Help/Unsubscribe/Update your Subscription:
%40developershed.com

This email sent to bsdarchive (AT) developershed (DOT) com