Which method is to access Pairs?

Question

just to clear this doubt of fine i just want to ask that which method is faster or more efficient to access pairs while iterating over a contiguous block of pairs.

I used two method to iterate over a block.

1st

  pair<int, char> arr[3] = {{1, 'a'}, {2, 'b'}, {3, 'c'}};
   
for (int i = 0; i < 3;i++){
        cout << get<0>(arr[i]) << " " << get<1>(arr[i]) << endl;
    }

2nd

 for(const auto &x:arr){
    cout << x.first << " " << x.second << endl;
    get<0>(arr[0]);
}

which one is better and more efficient pls explain if u can.

`arr[i].first`, etc. is another possibility. They're probably all about the same. If it is important to you then I would suggest learning how to profile and benchmark then test with your actual code. — Retired Ninja, Dec 15 '21 at 05:21
Please realize that the code you write is only a description of what you want done. The compiler is then free to optimize the code in any way it sees fit, usually making the code the compiler is running look nothing like your original source code (which is why it is hard to debug optimized code). The only thing is that the resulting code must produce the same results. So unless you look at your compiler's assembly language after optimizations are used, it is impossible to say how two almost identical pieces of C++ code will perform. — PaulMcKenzie, Dec 15 '21 at 05:29

digito_evo · Answer 1 · 2021-12-15T05:51:02.827

2

You can compare them here: https://godbolt.org/z/4dPzKaPWr

As you will see, both have the same assembly code.

edited Dec 15 '21 at 05:51

answered Dec 15 '21 at 05:22

digito_evo

3,216
2
14
42

score 1 · Answer 2 · answered Dec 15 '21 at 05:28

Efficiency takes many forms. I don't expect huge runtime differences in either case (you will have to measure/profile and expect std::cout to slow things down a lot, so be careful what you measure!). But in terms of maintainability (which is also an efficiency). I would use structured bindings on pairs (or not use pairs at all and use my own struct with clearly readable names)

#include <iostream>
#include <utility>

// don't use : using namespace std

struct my_data_t
{
    int number;
    char character;
};


int main()
{
    std::pair<int, char> arr[]{{1, 'a'}, {2, 'b'}, {3, 'c'}};
    my_data_t arr2[]{{1, 'a'}, {2, 'b'}, {3, 'c'}};
    
    // using structured bindings for readabiliey
    for(const auto& [number, character] : arr)
    {
       // for speed, don't use std::endl though
       std::cout << number <<", " << character << "\n"; 
    }
    
    // or using own datastructure
    for(const auto& entry : arr2)
    {
        std::cout << entry.number <<", " << entry.character << "\n"; 
    }
    
    return 0;
}

score 0 · Answer 3 · answered Dec 15 '21 at 05:20

0

They are the same. The template arguments 0 and 1 in get<> must be compile-time constants, An implementation of get<>() refers to first or second for 0 or 1 respectively. The function call to get<>() gets inlined when optimized.

answered Dec 15 '21 at 05:20

extratype

92
3

Which method is to access Pairs?

3 Answers3