Flatten a vector of vectors in C++

This post will discuss how to flatten a vector of vectors in C++.

1. Using `std::insert` function

A simple solution is to iterate over the vector of vectors and insert the contents of each vector into a new vector using the std::insert function. It is overloaded to accept a range of elements to insert at the specified position, in the same order. A typical implementation of this approach would look like this:

#include <iostream>

#include <vector>

template<typename T>

std::vector<T> flatten(std::vector<std::vector<T>> const &vec)

{

std::vector<T> flattened;

for (auto const &v: vec) {

flattened.insert(flattened.end(), v.begin(), v.end());

}

return flattened;

}

int main()

{

std::vector<std::vector<int>> vec {

{ 1, 2, 3 }, { 4, 5 }, { 6, 7, 8, 9 }

};

std::vector<int> flattened = flatten(vec);

for (int &i: flattened) {

std::cout << i << ' ';

}

return 0;

}

Download Run Code

Output:

1 2 3 4 5 6 7 8 9

2. Using Nested Loop

Another option to flatten a nested vector is using the nested for loop. The idea is to pick every element using a nested for-loop and push add it at the end of a new vector. Here’s what the code would look like:

#include <iostream>

#include <vector>

#include <algorithm>

template<typename T>

std::vector<T> flatten(std::vector<std::vector<T>> const &vec)

{

int size = 0;

for (auto &v: vec) {

size += v.size();

}

std::vector<T> flattened;

flattened.reserve(size);

for (auto &v: vec) {

for (auto &e: v) {

flattened.push_back(e);

}

return flattened;

}

int main()

{

std::vector<std::vector<int>> vec {

{ 1, 2, 3 }, { 4, 5 }, { 6, 7, 8, 9 }

};

std::vector<int> flattened = flatten(vec);

for (int &i: flattened) {

std::cout << i << ' ';

}

return 0;

}

Download Run Code

Output:

1 2 3 4 5 6 7 8 9

3. Using `std::accumulate`

Another option to flatten a vector of vectors is using the standard function std::accumulate, defined in the header <numeric>. We can overwrite its default operation by providing a binary predicate in the optional fourth parameter, which performs the flatten operation on two vectors and return the flattened list.

#include <iostream>

#include <vector>

#include <numeric>

int main() {

std::vector<std::vector<int>> vec {

{ 1, 2, 3 }, { 4, 5, 6 }, { 7, 8, 9 }

};

auto flattened = std::accumulate(vec.begin(), vec.end(), decltype(vec)::value_type{},

[](auto &x, auto &y) {

x.insert(x.end(), y.begin(), y.end());

return x;

});

for (int &i: flattened) {

std::cout << i << ' ';

}

return 0;

}

Download Run Code

Output:

1 2 3 4 5 6 7 8 9

4. Using range-v3 library

Here’s a one-liner using the ranges::views::join function from ranges-v3 library, which forms the basis for C++20’s std::ranges.

#include <iostream>

#include <vector>

#include <range/v3/view/join.hpp>

#include <range/v3/range/conversion.hpp>

int main()

{

std::vector<std::vector<int>> vec {

{ 1, 2, 3 }, { 4, 5 }, { 6, 7, 8, 9 }

};

auto flattened = ranges::views::join(vec);

for (int &i: flattened) {

std::cout << i << ' ';

}

return 0;

}

Download Code

Output:

1 2 3 4 5 6 7 8 9

That’s all about flattening a vector of vectors in C++.

Flatten a vector of vectors in C++

1. Using std::insert function

2. Using Nested Loop

3. Using std::accumulate

4. Using range-v3 library

1. Using `std::insert` function

3. Using `std::accumulate`