C++ Read CSV File: Parse Comma-Separated Data

To read a CSV file in C++, open it with std::ifstream, read each line with std::getline, then split the line on commas. There’s no built-in CSV parser in the standard library, but combining file reading with string splitting handles the common cases in a few lines.

A Sample CSV

Imagine a file called people.csv:

name,age,city
Alice,30,London
Bob,25,Paris

The first line is a header (column names), and each following line is a record with fields separated by commas.

Reading and Splitting Line by Line

The core pattern is: read each line, then feed that line into a stringstream and pull out comma-separated fields with getline:

#include <iostream>
#include <fstream>
#include <sstream>
#include <string>
#include <vector>

int main() {
    std::ifstream file("people.csv");
    if (!file.is_open()) {
        std::cerr << "Could not open file\n";
        return 1;
    }

    std::string line;
    std::getline(file, line);   // read and discard the header

    while (std::getline(file, line)) {       // one record per line
        std::stringstream ss(line);
        std::string field;
        std::vector<std::string> row;

        while (std::getline(ss, field, ',')) // split on commas
            row.push_back(field);

        std::cout << "Name: " << row[0]
                  << ", Age: " << row[1]
                  << ", City: " << row[2] << "\n";
    }
    return 0;
}

There are two loops: the outer getline reads a whole line from the file, and the inner getline with ',' splits that line into fields. Always check is_open() first — trying to read a missing file is a common beginner crash.

Storing the Whole File in a 2D Vector

Often you want the entire table in memory. Store each row as a vector<string>, giving a vector<vector<string>>:

#include <iostream>
#include <fstream>
#include <sstream>
#include <string>
#include <vector>

int main() {
    std::ifstream file("people.csv");
    std::vector<std::vector<std::string>> data;
    std::string line;

    while (std::getline(file, line)) {
        std::stringstream ss(line);
        std::string field;
        std::vector<std::string> row;
        while (std::getline(ss, field, ','))
            row.push_back(field);
        data.push_back(row);
    }

    std::cout << "Read " << data.size() << " rows\n";
    return 0;
}

Now data[1][0] is "Alice" and data[2][2] is "Paris". This structure is flexible, but remember every field is a string for now.

Converting Fields to Numbers

The age column is text like "30". To do math with it, convert with std::stoi (string to int) or std::stod (string to double):

int age = std::stoi(row[1]);   // "30" becomes 30

If a field might be empty or malformed, guard the conversion so one bad row doesn’t crash the program:

try {
    int age = std::stoi(row[1]);
} catch (const std::exception& e) {
    std::cerr << "Bad number: " << row[1] << "\n";
}

A Note on Quoted Fields

This simple parser splits on every comma, which breaks if a field itself contains a comma inside quotes, like "Smith, John". For clean, controlled data it’s perfectly fine. For messy real-world CSVs with quotes and escapes, consider a small library, but the techniques above cover the large majority of beginner tasks.

Related Articles

• C++ Read File Line by Line — fstream reading and writing
• C++ Split String — the splitting logic in depth
• C++ stringstream — parsing fields out of a line
• C++ Vector Tutorial — storing rows and columns
• C++ String to int — converting fields to numbers

Take Your C++ Further

If you’re looking to go deeper with C++, the C++ Better Explained Ebook is perfect for you — whether you’re a complete beginner or looking to solidify your understanding. Just $19.

👉 Get the C++ Better Explained Ebook — $19