Walter Bright wrote:
> Lucas Goss wrote:
>> Walter Bright wrote:
>>>
>>> Sometimes it is valuable to be able to say "you can't do this operation with this type". I don't see why this ability should be restricted to C++.
>>
>> Can someone come up with an example or explain why this would be valuable? To make sure we're on the same page, we're talking about:
>>
>> ----
>> class Base {
>>   public:
>>     virtual int Number() = 0;
>> };
>> ...
>> class Child : public Base {
>>   private:
>>     int Number() { return 1; }
>> };
> 
> I think it's a bug that the compiler allows this. It should be illegal code, because it breaks encapsulation.

Um, maybe I didn't make it clear enough, or maybe I'm not understanding. The above is C++ code that went with this code:

(1)
Child c;
int num = c.Number(); //error: cannot access private member

(2)
Child* cp = new Child();
num = cp->Number(); //error: cannot access private member

(3)
Base* bp = new Child();
num = bp->Number(); // num = 1

So the compiler does error with the first two attempts, but the third compiles and runs fine. I was just trying to find an example of 'poisoning' a base class that you mentioned. To me it doesn't sound like very good practice (maybe that's why C# and I think Java disallow such practice). I just wanted to make sure I'm understanding correctly.

Lucas

Lucas Goss wrote:
> So the compiler does error with the first two attempts, but the third compiles and runs fine.

Yes, it is C++ and is perfectly legal C++. A more interesting question is should it be legal? I posed the question in comp.lang.c++.moderated, and got a couple of interesting answers.

Walter Bright wrote:
> Lucas Goss wrote:
> 
>> So the compiler does error with the first two attempts, but the third compiles and runs fine.
> 
> 
> Yes, it is C++ and is perfectly legal C++. A more interesting question is should it be legal? I posed the question in comp.lang.c++.moderated, and got a couple of interesting answers.

in Java you cannot override a method with a more protected access.  I consider it a safeguard, so that I do not get caught in a lattice of protection attributes and inheritance.

-DavidM

Lucas Goss wrote:
> Bruno Medeiros wrote:
>>
>> Good. That should be documented in the spec then, that isn't mentioned anywhere, as far as we know.
>>
> 
> Uh, that was C++ code, sorry I didn't make it clear.
> 
> Lucas

Yes, but what Walter said (about the spec) was referring to D, even if your post was a C++ code.

-- 
Bruno Medeiros - CS/E student
http://www.prowiki.org/wiki4d/wiki.cgi?BrunoMedeiros#D

Walter Bright wrote:
> Lucas Goss wrote:
>> So the compiler does error with the first two attempts, but the third compiles and runs fine.
> 
> Yes, it is C++ and is perfectly legal C++. A more interesting question is should it be legal? I posed the question in comp.lang.c++.moderated, and got a couple of interesting answers.

Could you post the link to that discussion? (I don't think I can easily find it myself)

( Also, if you don't know already, C# adopts an invariant behavior: it disallows to change the protection in any way. Java has a covariant behavior, as mentioned by David. )

-- 
Bruno Medeiros - CS/E student
http://www.prowiki.org/wiki4d/wiki.cgi?BrunoMedeiros#D

Walter Bright wrote:
> Lucas Goss wrote:
>> So the compiler does error with the first two attempts, but the third compiles and runs fine.
> 
> Yes, it is C++ and is perfectly legal C++. A more interesting question is should it be legal? I posed the question in comp.lang.c++.moderated, and got a couple of interesting answers.

That's what I'm trying to discover, is there a good reason it should be legal. I found the thread here (for those interested).
http://groups.google.com/group/comp.lang.c++.moderated/browse_thread/thread/38a8bf39637180de/#

Lucas

Bruno Medeiros wrote:
> Walter Bright wrote:
> 
>> Lucas Goss wrote:
>>
>>> So the compiler does error with the first two attempts, but the third compiles and runs fine.
>>
>>
>> Yes, it is C++ and is perfectly legal C++. A more interesting question is should it be legal? I posed the question in comp.lang.c++.moderated, and got a couple of interesting answers.
> 
> 
> Could you post the link to that discussion? (I don't think I can easily find it myself)
> 
> ( Also, if you don't know already, C# adopts an invariant behavior: it disallows to change the protection in any way. Java has a covariant behavior, as mentioned by David. )
> 

Not to mention you can always throw an Exception in the disallowed operations, so I don't see this as a great usage of protection attributes.

-DavidM

Bruno Medeiros wrote:
> Could you post the link to that discussion? (I don't think I can easily find it myself)

It's entitled:

	Restricting access should be illegal?

posted 7/15/2006 3:38 AM on comp.lang.c++.moderated.

It's an ongoing discussion, so it'd be premature to summarize here.