Thread: #define

I dont know if it is because it is [] etc...
Is it possible to #define things like this or point to them in any way.

hmm... I have never seen such a macro before. MSVC8 reports "error C2008: '[' : unexpected in macro definition". The MinGW port of g++ 3.4.5 reports "warning: ISO C requires whitespace after the macro name". The Comeau online compiler does not complain at all.

Anyway, macros considered harmful in C++, so you might as well avoid them if you can.

Originally Posted by Coding

If it doesn´t work with a macro like this. Is there any other way to do this. Use pointers or any other technique.

Yes - use pointers or iterators

As in you want to map Number[n] to number[d+n]? One way would be to use a function, or even a function object. It depends on what exactly are you trying to use this in.

EDIT:
On second thought, if you use a function you need to pass d in anyway (or use a global variable, which can be a Bad Thing), and a function object might be unnecessarily complicated. If that is the case, then a pointer/iterator would be better.

I can take a simple example with a code:
I will read a textfile line by line, these lines consist of these numbers:

1
2
3

Code:

std::vector<int> Number(500);

int d = 500;
int NumberX = 0;

while ( getline(File, NumberX, '\n') )
{
      d = (d - 1);

    
      File.get();

      Number[d] = NumberX;    // For the first line NumberX[d] will have value 1
 
}

So if you are reading this textfile the Number[d] for the first line will have value: 1
Now, for the second line:

Number[d] will have value: 2
and
Number[d+1] will have value 1

Because I did a count down with d = (d - 1);

So when reading files I will have to refer to previous values like this, so instead of
writing in this example Number[d+1] I will just write Number[1] and this will meen the same thing.

Not to mention doing that would be extremely confusing. First you would mistake it for real code and not some substituted code and secondly, you would have to go find that macro to understand it and third, it would affect all such variables with that exact name, which is very bad, unless you undefine it.

I do not understand what you are trying to do, and the code example does not really help since it is incorrect.

Why not provide a minimal, compilable example of what you are trying to do? You do not need to start with d = 500, just d = 5 will do and then you show why you find it tedious to write the code.

I will give an exact example of how this is done and why I want to do this.

The code look like this and I will read 5 lines from a .txt file that look like this:

1,1
1,2
1,3
1,4
1,5

The code to read these look like this:

Code:

#include "stdafx.h"
#include <iostream>
#include <fstream>
#include <string>  
#include <vector>

using namespace std;
int main () 
{ 
	std::vector<int> Number(5);
	
	std::string Number1;
	int d = 6;						// Put d to its startvalue = 5;
	int Number2 = 0;
	int Number3 = 0;

	ifstream File("File1.txt");
					
	while	(    getline(File, Number1, ',')   )           		
	{
		
		d = (d - 1);				// CountDown, First Value is (6-1) = 5
				 
		File >> Number2;
		File.get();

		Number[d] = Number2;		// For the first line red in the file: Number[d] will have value 1


			if (Number[d+1] == 1)
			{
				Number3 = 1;
			}

	}

return 0;
}

So what am I doing here ? While reading the file I am doing a countdown for int d = 6 so for the first line this will have the number: 5
As seen now below this I am giving Number[d] the value of Number2 wich is the first line in the textfile: 1

For the next line that is red in the file d wil have a value of 4 as it is a countdown so Number[d] will have the value of 2 this time.

This moment Number[d] will have the value of 2 and Number[d+1] will have the value of 1. yes.

Below this I am doing a ifstatement that says if (Number[d+1] == 1) { Number3 = 1};
This is True while reading the second line, yes.

So now you probably are asking why I want to find a way to write Number[1] instead of Number[d+1] as this works fine ?

There is a huge reason why. This is because I will write 100:s of pages of code that will exist of this example. This was a very easy example.
So writing Number[1] instead of Number[d+1] is easier and better for my purpose. So to find a way to do that would be nice how it could be done.
With pointers or other approches.
Hope this understands better what I am trying to do.

So what am I doing here ? While reading the file I am doing a countdown for int d = 6 so for the first line this will have the number: 5
As seen now below this I am giving Number[d] the value of Number2 with is the first line in the textfile: 1

hmm... I am still a little unsure of what it is all about. For example, your comment says "For the first line red in the file: Number[d] will have value 1", but Number2 does not have to be 1. Furthermore, by fixing the vector at size 5 and not checking the number of elements in the loop you risk an out of range problem if the input is not as expected.

There is a huge reason why. This is because I will write 100:s of pages of code that will exist of this example. This was a very easy example.
So writing Number[1] instead of Number[d+1] is easier and better for my purpose.

There is almost no difference between Number[1] and Number[d+1] in terms of how much you have to type. Besides, saving keystrokes is not a valid reason for a design. If you want to create an abstraction (and you probably should), you should abstract away the entire example into functions and/or classes to improve readability and reusability, not just designing for a cosmetic change of Number[d+1] to Number[1] (or something along similiar, pessimistic lines).

I agree with laserlight. Just typing Number[1] will obfuscate the code and will confuse readers.
And your code is flawed, as well: you initialize d to 6, count down to 5 inside the loop, access element 5 which does not exist. Further, you try to access d + 1, 6, which doesn't exist either.
Remember that indexes start from 0, so 0 to 4 exists, but not 5 or 6.

Yes I knew that it began from 0 to 4, this was just a fast writing mistake. This will confuse you and me if we look at this code perheps but the thing is that I am trying to redefine some of the syntax so other people that dont know C++ syntax can write own code that they already know about. This code is that I will redefine. They dont know C++ so for these people it will be confusing to write for example then Number[d+1] instead of Number[1] as they are used to and many other examples. This is the reason why I try to do it. I understand it seems a little bit confusing ofcourse.

This will confuse you and me if we look at this code perheps but the thing is that I am trying to redefine some of the syntax so other people that dont know C++ syntax can write own code that they already know about. This code is that I will redefine. They dont know C++ so for these people it will be confusing to write for example then Number[d+1] instead of Number[1] as they are used to and many other examples.

I suggest you stop now. If they don't know C++ syntax, but can program in another language then a) why can't they continue programming in a language that they know, 2) why can't they learn new syntax - basic syntax is really easy to learn if you know how to program in general?

I don't think this is a little bit confusing - this is very confusing and changing the language syntax is clearly a misuse of #define.

I think I missed that part. It's clearly a bad thing, since it's must clearer to write Number[d+1] than Number[1], since obviously Number[d+1] is not Number[1]. It's an entirely different index. It would only serve to confuse.