www.digitalmars.com         C & C++   DMDScript  

digitalmars.D.learn - abstract base class and class members

reply David <d dav1d.de> writes:
Is this intended behaviour? http://ideone.com/xrvvL

shouldn't the 2nd writeln print the same as the first, well at least the 
same content of i?
Mar 04 2012
parent reply Timon Gehr <timon.gehr gmx.ch> writes:
On 03/04/2012 06:16 PM, David wrote:

 Is this intended behaviour? http://ideone.com/xrvvL

 shouldn't the 2nd writeln print the same as the first, well at least the
 same content of i?


This is intended behaviour. You have two distinct definitions of i. If 
you want to set i to 2 in the derived class, do so in the class constructor.
Mar 04 2012
parent reply "H. S. Teoh" <hsteoh quickfur.ath.cx> writes:
On Sun, Mar 04, 2012 at 06:22:47PM +0100, Timon Gehr wrote:

 On 03/04/2012 06:16 PM, David wrote:

Is this intended behaviour? http://ideone.com/xrvvL

shouldn't the 2nd writeln print the same as the first, well at least the
same content of i?


 
 This is intended behaviour. You have two distinct definitions of i.
 If you want to set i to 2 in the derived class, do so in the class
 constructor.


Yeah, only member functions can be overridden in the derived class (and
even then, D requires you to explicitly state that with the 'override'
keyword).

Makes one wonder, though... from an OO perspective, does it make sense
to have overridable non-function members? What semantics would (should)
that have?


T

-- 
"Life is all a great joke, but only the brave ever get the point." -- Kenneth
Rexroth
Mar 04 2012
next sibling parent reply David <d dav1d.de> writes:
Am 04.03.2012 19:22, schrieb H. S. Teoh:

 On Sun, Mar 04, 2012 at 06:22:47PM +0100, Timon Gehr wrote:

 On 03/04/2012 06:16 PM, David wrote:

 Is this intended behaviour? http://ideone.com/xrvvL

 shouldn't the 2nd writeln print the same as the first, well at least the
 same content of i?


 This is intended behaviour. You have two distinct definitions of i.
 If you want to set i to 2 in the derived class, do so in the class
 constructor.


 Yeah, only member functions can be overridden in the derived class (and
 even then, D requires you to explicitly state that with the 'override'
 keyword).

 Makes one wonder, though... from an OO perspective, does it make sense
 to have overridable non-function members? What semantics would (should)
 that have?


 T


Thanks for your answers.

Maybe we should also allow override for fields.
Mar 04 2012
parent reply Jonathan M Davis <jmdavisProg gmx.com> writes:
On Sunday, March 04, 2012 19:36:20 David wrote:

 Am 04.03.2012 19:22, schrieb H. S. Teoh:

 On Sun, Mar 04, 2012 at 06:22:47PM +0100, Timon Gehr wrote:

 On 03/04/2012 06:16 PM, David wrote:

 Is this intended behaviour? http://ideone.com/xrvvL
 
 shouldn't the 2nd writeln print the same as the first, well at least the
 same content of i?


 
 This is intended behaviour. You have two distinct definitions of i.
 If you want to set i to 2 in the derived class, do so in the class
 constructor.


 
 Yeah, only member functions can be overridden in the derived class (and
 even then, D requires you to explicitly state that with the 'override'
 keyword).
 
 Makes one wonder, though... from an OO perspective, does it make sense
 to have overridable non-function members? What semantics would (should)
 that have?
 
 
 T


 
 Thanks for your answers.
 
 Maybe we should also allow override for fields.


Variables can't be polymorphic, and it wouldn't make sense for them to be. 
Overidding is done to change behavior.

By the way, I wouldn't rely on much that ideone says about D at this point. 
It's still on version 2.042 of dmd, whereas the latest release is 2.058.

- Jonathan M Davis
Mar 04 2012
next sibling parent David <d dav1d.de> writes:
Am 04.03.2012 21:24, schrieb Jonathan M Davis:

 On Sunday, March 04, 2012 19:36:20 David wrote:

 Am 04.03.2012 19:22, schrieb H. S. Teoh:

 On Sun, Mar 04, 2012 at 06:22:47PM +0100, Timon Gehr wrote:

 On 03/04/2012 06:16 PM, David wrote:

 Is this intended behaviour? http://ideone.com/xrvvL

 shouldn't the 2nd writeln print the same as the first, well at least the
 same content of i?


 This is intended behaviour. You have two distinct definitions of i.
 If you want to set i to 2 in the derived class, do so in the class
 constructor.


 Yeah, only member functions can be overridden in the derived class (and
 even then, D requires you to explicitly state that with the 'override'
 keyword).

 Makes one wonder, though... from an OO perspective, does it make sense
 to have overridable non-function members? What semantics would (should)
 that have?


 T


 Thanks for your answers.

 Maybe we should also allow override for fields.


 Variables can't be polymorphic, and it wouldn't make sense for them to be.
 Overidding is done to change behavior.

 By the way, I wouldn't rely on much that ideone says about D at this point.
 It's still on version 2.042 of dmd, whereas the latest release is 2.058.

 - Jonathan M Davis


Sometimes you need constants which vary from the parents, which sould be 
statically accessable and which you can use to identify the class, I 
don't like the idea of having an additional functoion ( property) for 
doing this.

I used ideone just to show you, I discovered this on my PC with dmd 2.058
Mar 04 2012
prev sibling parent reply "Jesse Phillips" <jessekphillips+D gmail.com> writes:
On Sunday, 4 March 2012 at 20:25:40 UTC, Jonathan M Davis wrote:


 By the way, I wouldn't rely on much that ideone says about D at 
 this point.
 It's still on version 2.042 of dmd, whereas the latest release 
 is 2.058.

 - Jonathan M Davis


Then ask for the latest version, and as David pointed out, it is 
reliable to show behavior when you already have observed it.
Mar 05 2012
parent "Jonathan M Davis" <jmdavisProg gmx.com> writes:
On Monday, March 05, 2012 11:32:39 Jesse Phillips wrote:

 On Sunday, 4 March 2012 at 20:25:40 UTC, Jonathan M Davis wrote:

 By the way, I wouldn't rely on much that ideone says about D at
 this point.
 It's still on version 2.042 of dmd, whereas the latest release
 is 2.058.
 
 - Jonathan M Davis


 
 Then ask for the latest version, and as David pointed out, it is
 reliable to show behavior when you already have observed it.


I was just pointing out that relying on ideone's behavior is a bad idea, since 
it's very behind on its version of dmd. I didn't say that what it showed was 
necessarily wrong.

- Jonathan M Davis
Mar 05 2012
prev sibling parent Timon Gehr <timon.gehr gmx.ch> writes:
On 03/04/2012 07:22 PM, H. S. Teoh wrote:

 Makes one wonder, though... from an OO perspective, does it make sense
 to have overridable non-function members? What semantics would (should)
 that have?


Yes, it does make sense. const/immutable/(final) fields can be 
covariant. OTOH, it is not evident that it would be extremely useful.
Mar 04 2012