CSV

JavaScript · Reference cheat sheet

📋 Overview

CSV is comma-separated text (RFC 4180–style). JavaScript has no built-in CSV parser. For simple rows you can split carefully; for real data use a library such as Papa Parse (browser/Node) or csv-parse / csv-stringify (Node streams).

🔧 Core concepts

• Rows / fields: lines of records; fields separated by , (or ;, \t).
• Quoting: fields with commas/newlines wrap in "; "" escapes a quote.
• Header row: optional first line of column names.
• Manual parse: fine for trusted, simple files without embedded commas.
• Libraries: Papa Parse (Papa.parse / Papa.unparse); Node csv-parse + csv-stringify.

💡 Examples

// Manual — only if no commas inside fields
function parseSimpleCsv(text) {
  return text
    .trim()
    .split(/\r?\n/)
    .map((line) => line.split(","));
}

function toSimpleCsv(rows) {
  return rows.map((r) => r.join(",")).join("\n");
}

const rows = parseSimpleCsv("name,age\nAda,36\n");

// Escape a field for RFC-ish output
function csvField(value) {
  const s = String(value ?? "");
  if (/[",\n\r]/.test(s)) return `"${s.replaceAll('"', '""')}"`;
  return s;
}

function rowsToCsv(rows) {
  return rows.map((row) => row.map(csvField).join(",")).join("\n");
}

// Papa Parse (npm: papaparse)
import Papa from "papaparse";

const parsed = Papa.parse(csvText, { header: true, skipEmptyLines: true });
// parsed.data → array of objects
const out = Papa.unparse(parsed.data);

// Node csv-parse (npm: csv-parse)
import { parse } from "csv-parse/sync";
import { readFile } from "node:fs/promises";

const input = await readFile("data.csv", "utf8");
const records = parse(input, { columns: true, skip_empty_lines: true });

⚠️ Pitfalls

• Naïve split(",") breaks on quoted commas and newlines inside fields.
• Locale exports may use ; as separator — do not assume ,.
• Excel often expects UTF-8 BOM for Unicode; add \uFEFF when needed.
• Large files: prefer streaming parsers (csv-parse stream API), not full-string parse.
• Type coercion: CSV is all strings until you convert numbers/dates yourself.

🔗 Related

• JSON — structured alternative to CSV
• fs — read/write CSV files in Node
• Strings — split / replace helpers
• Async — async file + parse flows
• fetch — download CSV over HTTP