Thread: reference to self ... so to speak

I guess that's why I get an error when I'm creating the class

Fair enough. I do silly things all the time.

now I get an error here :

Code:

myVector[myClass.size-1]._reference_tomyClass = & myVector[myClass.size];

I think I did it right ... but maybe I missed something


***** EDIT

I'm anitiot ...

I missed the ()

thanks everyone for your help

Could you provide more code and the exact error message? It makes it easier to help diagnose the problem.

I will make a compromise with you, Elysia. I will always call a reference (&) a reference on the C++ board, and a pointer (*) a pointer on the C++ board. But on the C board I am allowed to call a pointer parameter (*) a reference.

By the way:

Originally Posted by urbandictionary

"anitiot isn't defined yet."

Well I do try to be respectful of other competent members. I suppose this may be an issue of newfangled teaching techniques that are floating around these days. I can agree with you to an extent, but in C and assembler a pointer is the only way to pass by reference. If anything, I would make stronger argumets for a reference to be renamed to something else. But there is no more appropriate moniker than "reference." So without introducing circular logic the point is moot. I do appreciate your concern over use of the appropriate descriptions though. It is beneficial to those who are learning. But a lot of books will also tend to say something is being passed by reference (when you would say "pointer").

If anything, I can re-negotiate the compromise to for C I will use the term "passing by address" which I think even you can conceed is accurate and unambiguous. But using such terms runs the risk of people not knowing what I mean.

>> I suppose this may be an issue of newfangled teaching techniques that are floating around these days. <<

I think this might just be Elysia's personal preference. I know I've never heard anybody else complain about referring to a pointer as referencing another object.

As long as people are aware that in C++ the term "reference" has a specific well-defined meaning beyond the normal programming term, then I don't think its wrong to use both in the proper context.

It's like the terms string or map. There is a C++ string and a C++ map, but that doesn't mean that a sequence of characters can't also be called a string and it doesn't mean you can't map one object to another with something other than a std::map.

>> No, you can only pass by pointer. Simple. <<

Not so simple. Pass-by-reference is a well defined programming term. You can't just stop using it because the word reference is used elsewhere.

Yeah we will have to agree to disagree. Though I guess I agree with you almost asymptotically. I mean there is only a very small amount of disagreement on the issue. I can certainly say its one of those things I will have to shrug off because it doesn't matter. I just hate needing to "choose my words carefully." However look at the silver lining, I will just say what I am going to say and by doing so you are guaranteed a boost in post count

Originally Posted by laserlight

"Pass by reference" is a parameter passing method in the literature. It can be implemented via pointers, C++ references, etc.

You are my official best friend of the day. Thank you laserlight.