Re: [Dev-C++] Off:Classes Access Restrictions

[B. <ama...@uo...>]

Yes, perfect both methods and the solutions suggested. Thank you. The cla=
ss side, that
is, when the access to the fuction members of the class is made, it is we=
ll defined.

Now I'm thinking in a way to discover whose program is accessing them (th=
e classes).

In a Unix world I could check the gid and discover which kind of user pro=
gram is
requesting it.
And in a Win environment?

Are you understanding now? How can, in a run-time mode, I can check this =
access

Regards
Andr=E9
=3D=3D=3D
Daniel Glenfield wrote:

> I'd considered derived classes as a solution, but I think perhaps the
> exercise (it sounds like one) involves creating a portable class - one =
that
> can be plugged in to a project.
>
> The variables specified could have been anything, so we're not sure on
> whether our solutions are going to work or not. The functions are not
> specified so they could do anyhting. I thought setting access via the
> constructors would work better. After all, if the object was to derive =
the
> class into child classes the exercise would have said so surely?
>
> >From: "Abhijit Shylanath" <mu...@gm...>
> >To: "Daniel Glenfield"
> ><sp...@ho...>,<ama...@uo...>,<dev...@li...=
rceforge.net>
> >Subject: Re: [Dev-C++] Off:Classes Access Restrictions
> >Date: Tue, 29 Oct 2002 18:26:02 +0530
> >
> >Greets, everyone.
> >
> > > So a superdummy could be (DummyAccess =3D true, NerdAccess =3D true=
)
> > > and the function will not care what the object is named, it will ju=
st
> >see
> > > the value of the variable for NerdAccess. If it were a string, the
> >function
> > > would have to be rewritten to accomodate "superdummy"
> >
> >I came in late (I'd disabled my mail delivery for sometime), so I may =
not
> >be
> >talking about what you're talking about. But according to what Daniel =
is
> >talking about, someone wants a class, some parts of which are accessib=
le to
> >some dudes, other parts to other dudes. His method makes sense. Also, =
it
> >should be more efficient than using strings (less time to compare or
> >check).
> >You could also modify the method to use a flag, like
> >
> >     Constructor(int access_mode);
> >
> >which could be a combination of
> >
> >     #define FOO_MODE_DUMMY    0x01
> >     #define FOO_MODE_NERD     0x02
> >
> >like so:
> >
> >     Constructor(FOO_MODE_DUMMY | FOO_MODE_NERD);
> >
> >Within the methods, checking would be like so:
> >
> >     if(access_mode & FOO_MODE_NERD)
> >     {
> >         // Do nerdy stuff
> >     }
> >
> >But this method is run-time checking, and you wouldn't know what to do=
 if a
> >dummy tries to access a nerd-only-function. Error-message? Drop to OS?
> >Wouldn't it be better if there were compile-time errors? Yes, so we ha=
ve
> >the
> >new and improved C++-style DERIVED CLASSES! Yes, with one simple wave =
of
> >the
> >hand... never mind... how's this?
> >
> >class FooParent
> >{
> >     public:
> >         FooParent();
> >         ~FooParent();
> >
> >         virtual int WhichFoo() =3D 0;
> >
> >     private:
> >         int    num_stuff;
> >};
> >
> >class FooDummy : public FooParent
> >{
> >     /* blah */
> >};
> >
> >class FooNerd : public FooNerd
> >{
> >     /* different blah */
> >};
> >
> >Was this what was asked?
> >
> >___________________________________________________________
> >
> >Abhijit Shylanath
> >
> >E-mail:   mu...@gm... || ibr...@bi...
> >Web-site: http://mudeth.tripod.com/
>
> _________________________________________________________________
> Surf the Web without missing calls! Get MSN Broadband.
> http://resourcecenter.msn.com/access/plans/freeactivation.asp