Faster way to initialize array with text file C++

First you have to think about how 2-dimensional arrays declared and initialized. Simply;

int x[5][2] = {
    {1, 1},     
    {2, 2},     
    {3, 3},     
    {4, 4},     
    {5, 5}
};

if array elements coming from file you have to create the file exacly like; in file.txt;

{1, 1},     
{2, 2},     
{3, 3},     
{4, 4},     
{5, 5}

following the rule of C++ include file preprocessor directive, you can simple write your code like this; Source.cpp file;

#include <iostream>
using namespace std;

int x[5][2] = {
    #include "file.txt"
};

int main(int argc, char* argv[])
{   
    for(int i = 0; i < 5; ++i)
    {
        for(int j = 0; j < 2; j++)          
        {
            cout << "x[" << i + 1 << "][" << j + 1 << "] = " << x[i][j] << " , ";
        }
        cout << endl;
    }

    return 0;
}

If you stick the way of embedding the data within included file, you can use Macros:

1. Write your data in the included file ("myfile.txt" in your example, or "myfile.h").

2. Write the data in in the format of:

   #define DATA 1,2,3,4,5
   

   (you can break it into lines, but keep the comma-separated format).

3. include "myfile.h" at the top of your source file.

4. Use DATA instead of your #include statement:

   float array[] {
       DATA
   };
   

To make this work you should format your text file in a way that it follows the C++ syntax. You can follow number of tutorials in order to see good examples of that along with description of what is happening.

Here is a short example:

• for one dimensional array just type the numbers with comas like this

1.0, 3.4, 3.14, 5.56

• for two and more dimensions enclose each row in brace {}, separating each column with a coma like this

{1.0, 3.14}, {4.1, 5.674}

Above is assuming you're following your own example of

float array[] = {
    #include "textFile.txt"
};

You're doing it the right way IF you don't mind recompiling when the data changes. Of course the data has to have commas because it has to be valid C/C++ syntax. All #include does is paste the file's contents into your source code.

OTOH, if you DO mind recompiling when the data changes, then you need to allocate the memory dynamically and parse in the data file, when the program starts up. The problem you need to overcome here is you don't necessarily know in advance how big the array will be, and if you use a dynamically-expanding array you could waste a fair amount of time in memory allocation and deletion as it grows the array. My solution is, as usual, counterintuitive. I just read the file twice, once to count how many numbers are in it, then allocate the memory, then read the file again, to parse the numbers into the memory. (If reading it once doesn't take very long, neither does reading it twice, especially since it's already in the system cache.)

You say the data is 2-dimensional, but you could put it in the array as 1-dimensional, and just access it in two dimensions, like this:

#define A(row,col) array[(row)*2 + (col)]