📊 CSV ↔ JSON Converter

Complete Guide: CSV to JSON Converter

CSV (Comma-Separated Values) and JSON (JavaScript Object Notation) are two of the most common data interchange formats. CSV is the lingua franca of spreadsheets — exported by Excel, Google Sheets, and virtually every database tool. JSON is the native format for web APIs and JavaScript applications. Converting between them is a daily task for developers, data analysts, and anyone building data pipelines.

The RFC 4180 CSV Specification

Despite its name and simplicity, CSV has surprising edge cases defined by RFC 4180:

• Fields containing commas, double quotes, or line breaks must be enclosed in double quotes.
• Double quotes within a field are escaped by doubling them: "" represents a single ".
• The first record may optionally be a header row naming the columns.
• Each record must have the same number of fields.
• Line endings should be CRLF (\r\n) per the spec, though LF-only is widely accepted.

name,age,city
"Smith, John",42,"New York"
"O'Brien, ""Mike""",35,London

Header Row Handling

When a header row is present, converters produce an array of objects, where each row becomes an object keyed by column names:

[
  { "name": "Smith, John", "age": "42", "city": "New York" },
  { "name": "O'Brien, \"Mike\"", "age": "35", "city": "London" }
]

Without headers, or when using the column-array format, the output is an array of arrays:

[["Smith, John", "42", "New York"], ["O'Brien, \"Mike\"", "35", "London"]]

The BOM Character and Excel UTF-8 Issues

Microsoft Excel adds a UTF-8 BOM (Byte Order Mark, EF BB BF in hex, \uFEFF in Unicode) to the beginning of CSV files exported as "CSV UTF-8." This causes the first column header to appear as name instead of name in parsers that don't strip it. Always check for and remove the BOM before processing Excel-sourced CSV files.

Output Format Options

Two primary JSON output structures are useful in different contexts:

• Array of objects — Best for REST APIs, databases, and ORM input. Each object is self-describing. Field order is preserved by convention but not guaranteed in JSON spec.
• Column arrays (also called "columnar" or "records-split") — Best for charting libraries (e.g., Chart.js, D3.js) and analytics. All values for one column are grouped together.

/* Column arrays format */
{
  "name": ["Smith, John", "O'Brien"],
  "age": ["42", "35"],
  "city": ["New York", "London"]
}

Handling Nulls and Empty Cells

CSV has no native null type. Conventions vary: empty string, the literal text NULL, N/A, or - are all used. During conversion, decide whether empty CSV fields become JSON null, an empty string "", or are omitted entirely from the object. Be consistent — mixed null conventions break downstream type checking.

Streaming Large Files

For files exceeding a few hundred megabytes, browser-based conversion is impractical. Use a streaming approach in Node.js:

const { parse } = require('csv-parse');
const fs = require('fs');

fs.createReadStream('large.csv')
  .pipe(parse({ columns: true, bom: true }))
  .on('data', (row) => console.log(JSON.stringify(row)))
  .on('end', () => console.log('Done'));

Python's pandas is another popular alternative: df = pd.read_csv('file.csv'); df.to_json('out.json', orient='records').

• Format and validate your JSON output with the JSON Formatter.
• Compare two JSON outputs side by side with JSON Diff.