fast-csv-parser

🚀 A high-performance, drop-in replacement for csv-parser powered by Rust

Streaming CSV parser that aims for maximum speed while maintaining 100% compatibility with the original csv-parser API. Built with Rust and N-API for blazing-fast performance on large datasets.

fast-csv-parser is a drop-in replacement - simply change your import and enjoy better performance with zero code changes required.

⚡️ Up to 1.3x faster on large datasets (7K+ rows) 🔧 100% API compatible with the original csv-parser 🦀 Rust-powered native performance 🌐 Cross-platform support (Windows, macOS, Linux, ARM) 📦 Zero dependencies in production 🔤 UTF-16 support - handles UTF-16 LE/BE with BOM detection

Performance Comparison

Performance characteristics vary by file size. Here are real benchmark results:

Large Files (Where It Matters)

Dataset	Rows	Size	csv-parser	fast-csv-parser	Speedup
large-dataset.csv	7,268	1.1MB	59ms	47ms	🚀 1.26x faster
option-maxRowBytes.csv	4,577	700KB	36ms	29ms	🚀 1.24x faster

Small Files (Startup Overhead)

Small files show the Node.js ↔ Rust boundary overhead:

• Files <100 rows: ~0.6x performance (0.1ms → 0.2ms)
• The overhead is constant (~0.1ms) regardless of file size

Performance Profile

          Performance Ratio (higher = better)
     2x  ┤
         │     ╭─ Peak performance zone
   1.5x  ┤   ╱
         │ ╱
     1x  ┼─────────────────────── Break-even point (~1KB)
         │
   0.5x  ┤ Overhead zone
         └─────────────────────────────────────
         0KB    1KB     10KB    100KB   1MB+
                        File Size

💡 Recommendation: Use fast-csv-parser for files >10KB or high-throughput scenarios.

📦 Installation

Using npm:

npm install fast-csv-parser

Using yarn:

yarn add fast-csv-parser

Using pnpm:

pnpm add fast-csv-parser

🚀 Usage Examples

Basic Usage

Simply replace your csv-parser import:

// Before
const csv = require('csv-parser')

// After - that's it!
const csv = require('fast-csv-parser')

// Your existing code works unchanged
const fs = require('fs')
const results = []

fs.createReadStream('data.csv')
  .pipe(csv())
  .on('data', (data) => results.push(data))
  .on('end', () => {
    console.log(results)
    // [
    //   { NAME: 'Daffy Duck', AGE: '24' },
    //   { NAME: 'Bugs Bunny', AGE: '22' }
    // ]
  })

Advanced Usage Examples

Custom Headers and Transformations

const csv = require('fast-csv-parser')

fs.createReadStream('data.csv')
  .pipe(csv({
    headers: ['name', 'age', 'city'],
    skipLines: 1,
    mapHeaders: ({ header }) => header.toUpperCase(),
    mapValues: ({ header, value }) => {
      if (header === 'age') return parseInt(value)
      return value.trim()
    }
  }))
  .on('data', (row) => console.log(row))

Processing TSV Files

const csv = require('fast-csv-parser')

fs.createReadStream('data.tsv')
  .pipe(csv({ separator: '\t' }))
  .on('data', (row) => console.log(row))

Error Handling

const csv = require('fast-csv-parser')

fs.createReadStream('data.csv')
  .pipe(csv({ strict: true }))
  .on('data', (row) => console.log(row))
  .on('headers', (headers) => console.log('Headers:', headers))
  .on('error', (err) => {
    console.error('Parse error:', err.message)
  })
  .on('end', () => console.log('Parsing complete'))

Streaming Transform Pipeline

const csv = require('fast-csv-parser')
const { Transform } = require('stream')

const processor = new Transform({
  objectMode: true,
  transform(row, encoding, callback) {
    // Process each row
    row.processed_at = new Date().toISOString()
    this.push(row)
    callback()
  }
})

fs.createReadStream('input.csv')
  .pipe(csv())
  .pipe(processor)
  .pipe(fs.createWriteStream('output.json'))

📚 API Documentation

fast-csv-parser implements the exact same API as the original csv-parser. All options, events, and behaviors are identical.

csv([options | headers])

Returns: Transform Stream

Options

All original csv-parser options are supported:

• separator (String, default: ,) - Column separator
• quote (String, default: ") - Quote character
• escape (String, default: ") - Escape character
• newline (String, default: \n) - Line ending
• headers (Array|Boolean) - Custom headers or disable header parsing
• mapHeaders (Function) - Transform header names
• mapValues (Function) - Transform cell values
• skipLines (Number, default: 0) - Skip initial lines
• skipComments (Boolean|String, default: false) - Skip comment lines
• maxRowBytes (Number) - Maximum bytes per row
• strict (Boolean, default: false) - Strict column count validation
• raw (Boolean, default: false) - Disable UTF-8 decoding

Example with Options

const csv = require('fast-csv-parser')

fs.createReadStream('data.tsv')
  .pipe(csv({
    separator: '\t',
    mapHeaders: ({ header }) => header.toLowerCase(),
    mapValues: ({ value }) => value.trim()
  }))
  .on('data', (row) => console.log(row))

🎯 Events

data

Emitted for each parsed row (excluding headers).

headers

Emitted after header row is parsed with Array<string> of header names.

end

Emitted when parsing is complete.

error

Emitted on parsing errors.

⚡ Performance Tips

1. Use on large files - The performance benefits are most apparent with files >1MB
2. Enable raw: true for maximum speed if you don't need UTF-8 processing
3. Avoid complex mapValues functions - They can negate performance gains
4. Set appropriate maxRowBytes to avoid memory issues with malformed data

🌐 Platform Support

Pre-built binaries are available for:

• macOS: x64, ARM64 (Apple Silicon)
• Linux: x64, ARM64, ARM7 (GNU & musl)
• Windows: x64, x86, ARM64
• FreeBSD: x64
• Android: ARM64, ARM7

🔄 Migration from csv-parser

Migration is seamless:

// Old code
const csvParser = require('csv-parser')

// New code - just change the import!
const csvParser = require('fast-csv-parser')

// Everything else stays the same

🔤 Encoding Support

fast-csv-parser automatically detects and handles multiple text encodings:

• UTF-8 (default) - with automatic BOM stripping
• UTF-16 LE (Little Endian) - with BOM detection
• UTF-16 BE (Big Endian) - with BOM detection

No configuration needed - encoding is detected automatically from Byte Order Marks (BOM):

const csv = require('fast-csv-parser')

// Works with UTF-8, UTF-16 LE, UTF-16 BE files automatically
fs.createReadStream('data-utf16.csv')  // UTF-16 file
  .pipe(csv())
  .on('data', (row) => console.log(row))

Encoding Details

Encoding	BOM	Detection	Status
UTF-8	EF BB BF	Auto-detected, BOM stripped	✅ Supported
UTF-16 LE	FF FE	Auto-detected	✅ Supported
UTF-16 BE	FE FF	Auto-detected	✅ Supported
ASCII	None	Treated as UTF-8	✅ Supported
Other	-	Not supported	❌

📊 Benchmarks

Run benchmarks yourself:

# Clone the repository
git clone https://github.com/jonaylor89/fast-csv-parser
cd fast-csv-parser

# Install dependencies
npm install

# Run performance comparison against original csv-parser
./benchmark-comparison.js

# Run individual benchmarks
npm run bench

Sample output:

🏁 CSV Parser Performance Comparison
Comparing Rust implementation vs Original JavaScript csv-parser

📊 PERFORMANCE COMPARISON
========================================
File                    Original    Rust     Speedup
large-dataset.csv         59ms      47ms     1.26x ⚡
option-maxRowBytes.csv    36ms      29ms     1.24x ⚡
basic.csv               0.098ms    0.43ms    0.23x

🛠️ Error Handling

Errors are handled identically to the original csv-parser:

fs.createReadStream('data.csv')
  .pipe(csv())
  .on('error', (err) => {
    if (err instanceof RangeError) {
      console.log('Row length mismatch')
    } else {
      console.log('Parse error:', err.message)
    }
  })

📝 TypeScript Support

Full TypeScript definitions are included:

import csv from 'fast-csv-parser'

interface Row {
  name: string
  age: number
}

fs.createReadStream('data.csv')
  .pipe(csv())
  .on('data', (row: Row) => {
    console.log(row.name, row.age)
  })

🖥️ CLI Usage

The CLI is fully compatible with csv-parser:

# Parse CSV to NDJSON
fast-csv-parser data.csv

# Parse TSV
fast-csv-parser -s $'\t' data.tsv

# Custom options
fast-csv-parser --separator=';' --quote='"' data.csv

🔤 Advanced Encoding Usage

For most cases, encoding is handled automatically. For advanced scenarios:

const csv = require('fast-csv-parser')

// Automatic encoding detection (recommended)
fs.createReadStream('data.csv')  // Any supported encoding
  .pipe(csv())
  .on('data', (row) => console.log(row))

// For unsupported encodings, use iconv-lite preprocessing
const iconv = require('iconv-lite')
fs.createReadStream('latin1-data.csv')
  .pipe(iconv.decodeStream('latin1'))
  .pipe(csv())
  .on('data', (row) => console.log(row))

🤔 When to Use

Use fast-csv-parser when:

• Processing large CSV files (>1MB)
• Performance is critical
• You want a drop-in replacement for csv-parser
• Working with data pipelines or ETL processes

Stick with csv-parser when:

• Processing many small files (<1KB each) where startup overhead matters
• Bundle size is extremely critical (csv-parser is pure JS)
• You need cutting-edge features that haven't been ported yet

❓ FAQ

Q: Is this a drop-in replacement?

A: Yes! Just change your import from csv-parser to fast-csv-parser. All options, events, and behaviors are identical.

Q: When should I use this over the original?

A: Use fast-csv-parser when:

• Processing files larger than 10KB
• Performance is critical (ETL pipelines, data processing)
• You're already using csv-parser and want better performance

Q: Are there any breaking changes?

A: No breaking changes. The API is 100% compatible. UTF-16 files are now properly supported with automatic encoding detection.

Q: Why is it slower on small files?

A: There's a ~0.1ms overhead from the Node.js ↔ Rust boundary. For tiny files, this overhead exceeds the parsing time. The crossover point is around 1KB.

Q: Can I use this in the browser?

A: Not currently. This is a native Node.js addon. For browser use, stick with the original csv-parser.

Q: How stable is this?

A: Very stable. It passes all csv-parser tests and maintains the same error handling. The Rust core uses battle-tested CSV parsing libraries.

🛠️ Development

Building from Source

Requirements:

• Node.js 16+
• Rust 1.60+
• Supported platform (see Platform Support above)

# Clone the repository
git clone https://github.com/jonaylor89/fast-csv-parser
cd fast-csv-parser

# Install dependencies
npm install

# Build the native addon
npm run build

# Run tests
npm test

# Run benchmarks
npm run bench

Project Structure

fast-csv-parser/
├── src/                 # Rust source code
│   ├── lib.rs          # N-API bindings
│   └── parser.rs       # Core CSV parsing logic
├── __test__/           # Test files and fixtures
├── examples/           # Usage examples
├── bin/                # CLI tools
├── main.js             # Main entry point with Stream API
└── index.js           # Auto-generated native binding loader (build-safe)

Architecture

1. Rust Core (src/parser.rs) - High-performance CSV parsing
2. N-API Bridge (src/lib.rs) - Node.js ↔ Rust interface
3. Native Loader (index.js) - Auto-generated cross-platform binary loading
4. Main Entry (main.js) - Stream API compatibility layer (build-safe)

🤝 Contributing

Contributions welcome! This project maintains:

• 100% API compatibility with csv-parser
• Comprehensive test coverage (23+ test files)
• Performance benchmarks
• Cross-platform CI/CD
• Rust best practices

Testing

# Run all tests
npm test

# Run specific test
npm test headers.test.mjs

# Run benchmarks
./benchmark-comparison.js

📄 License

MIT License - same as the original csv-parser.

🙏 Acknowledgments

Built on the shoulders of giants:

• csv-parser by @mafintosh - The original, excellent CSV parser
• napi-rs - Rust N-API framework for Node.js addons

Special thanks to the csv-parser community for creating such a robust and well-designed API that made this drop-in replacement possible.

---

🚀 Ready to speed up your CSV processing?

npm install fast-csv-parser

Just replace your import and enjoy the performance boost!

⭐ Star on GitHub • 📦 View on npm • 🐛 Report Issues