How can I push_back data in a 2d vector of type int

问题

I have a vector and want to store int data in to it at run time can I store the data in a 2D vector in this manner ?

std::vector<std::vector <int>> normal:
    for(i=0;i<10;i++){
        for(j=0;j<20;j++){
            normal[i].push_back(j);
    }
}

回答1:

Yes, but you also need to push each of the sub-vectors:

std::vector<std::vector<int>> normal;
for(int i=0; i<10; i++)
{
    normal.push_back(std::vector<int>());
    for(int j=0; j<20; j++)
    {    
        normal[i].push_back(j);    
    }
}

回答2:

You are manipulating a vector of vectors. As such, when declaring normal it is empty and does not contain any element.

You can either :

Resize the vector prior to inserting elements

std::vector<std::vector<int> > normal;
normal.resize(20);

for (size_t i = 0; i < normal.size(); ++i)
{
    for (size_t j = 0; j < 20; ++j)
        normal[i].push_back(j);
}

This may be slightly more efficient than pushing an empty vector at each step as proposed in other answers.

Use a flat 2D array

If you want to store a 2D array, this is not the optimal solution, because :

Your array data is spread across N different dynamically allocated buffers (for N lines)
Your array can have a different number of columns per line (because nothing enforces that normal[i].size() == normal[j].size()

Instead, you can use a vector of size N * M (where N is the number of lines and M the number of columns), and access an element at line i and columns j using the index i + j * N :

size_t N = 20;
size_t M = 20;
std::vector<int> normal;
normal.resize(N * M);

for (size_t i = 0; i < N; ++i)
    for (size_t j = 0; j < M; ++j)
        normal[i + j * N] = j;

回答3:

You cannot directly assign to [i] without allocating the outer and inner vectors first. One solution to this would be to create the inner vectors inside your for loop, then once those are populated, push_back to the outer vector.

std::vector<std::vector<int>> normal;
for(i=0;i<10;i++)
{
    std::vector<int> temp;
    for(j=0;j<20;j++)
    {
        temp.push_back(j);
    }
    normal.push_back(temp);
}

回答4:

You have a vector of vectors.

normal[i] Does not exist because you have not created it.

std::vector<std::vector <int> > normal:
for(i=0;i<10;i++){
    normal.emplace_back();
    for(j=0;j<20;j++){
        normal.back().push_back(j);
    }
}

for(i=0;i<10;i++){
    for(j=0;j<20;j++){
        std::cout << normal[i][j] << " ";
    }
    std::cout << std::endl;
}

回答5:

Here is one more approach.

#include <iostream>
#include <iomanip>
#include <vector>
#include <numeric>

int main() 
{
    std::vector<std::vector <int> > normal;
    normal.resize( 10, std::vector<int>( 20 ) );

    for ( auto &v : normal ) std::iota( v.begin(), v.end(), 0 );

    for ( const auto &v : normal )
    {
        for ( int x : v ) std::cout << std::setw( 2 ) << x << ' ';
        std::cout << std::endl;
    }
}

The program output is

 0  1  2  3  4  5  6  7  8  9 10 11 12 13 14 15 16 17 18 19 
 0  1  2  3  4  5  6  7  8  9 10 11 12 13 14 15 16 17 18 19 
 0  1  2  3  4  5  6  7  8  9 10 11 12 13 14 15 16 17 18 19 
 0  1  2  3  4  5  6  7  8  9 10 11 12 13 14 15 16 17 18 19 
 0  1  2  3  4  5  6  7  8  9 10 11 12 13 14 15 16 17 18 19 
 0  1  2  3  4  5  6  7  8  9 10 11 12 13 14 15 16 17 18 19 
 0  1  2  3  4  5  6  7  8  9 10 11 12 13 14 15 16 17 18 19 
 0  1  2  3  4  5  6  7  8  9 10 11 12 13 14 15 16 17 18 19 
 0  1  2  3  4  5  6  7  8  9 10 11 12 13 14 15 16 17 18 19 
 0  1  2  3  4  5  6  7  8  9 10 11 12 13 14 15 16 17 18 19

You can write a corresponding function

#include <iostream>
#include <iomanip>
#include <vector>
#include <numeric>

template <typename T>
T & init_2d( T &container, size_t m, size_t n )
{
    container.resize( m, typename T::value_type( n ) );

    for ( auto &item : container ) std::iota( item.begin(), item.end(), 0 );

    return container;
}

int main() 
{
    std::vector<std::vector<int>> v;

    for ( const auto &v : init_2d( v, 10, 20 ) )
    {
        for ( int x : v ) std::cout << std::setw( 2 ) << x << ' ';
        std::cout << std::endl;
    }

}

回答6:

Allocate n empty vectors, that is, empty vector for each index. Then push_back() can be applied.

int main()
{
    int n = 10;
    std::vector<std::vector<int>> normal;
    normal.resize(n);   //Allocating 'n' empty vectors
    for (int i = 0; i < n; i++)
    {
        for (int j = 0; j < 20; j++)
        {
             normal[i].push_back(j);
        }
    }
    return 0;
}

来源：https://stackoverflow.com/questions/42249303/how-can-i-push-back-data-in-a-2d-vector-of-type-int

标签

c++

multidimensional-array

vector