csv_plus

The fastest, most complete CSV package for Dart.
Encode · Decode · Stream · Query · Transform · Validate

Zero dependencies · Pure Dart · VM, Web & AOT

Why csv_plus? #

Most CSV libraries only do basic parsing. csv_plus gives you a complete toolkit — from raw byte-level decoding to full table operations — in a single, zero-dependency package.

Installation #

dependencies:
  csv_plus: ^latest

import 'package:csv_plus/csv_plus.dart';

Quick Start #

Encode & Decode #

final codec = CsvCodec();

// Encode
final csv = codec.encode([
  ['name', 'age', 'score'],
  ['Alice', 30, 95.5],
  ['Bob', 25, 88.0],
]);

// Decode — types are automatically inferred
final rows = codec.decode(csv);
// rows[1] == ['Alice', 30, 95.5]  ← int and double preserved

Header-Aware Rows #

final people = codec.decodeWithHeaders(csv);
print(people.first['name']); // Alice
print(people.first['age']);  // 30 (int, not String)

CsvTable — Query & Transform #

final table = CsvTable.parse('name,age,city\nAlice,30,NYC\nBob,25,LA\nEve,35,NYC');

// Filter
final nyc = table.where((row) => row['city'] == 'NYC');

// Sort
table.sortBy('age');

// Aggregate
final avgAge = table.avg('age'); // 30.0

// Group
final byCity = table.groupBy('city'); // {NYC: CsvTable, LA: CsvTable}

// Export
print(table.toCsv());
print(table.toFormattedString()); // Pretty-printed aligned table

Stream Large Files #

import 'package:csv_plus/io.dart';

// Stream — constant memory, any file size
await for (final row in CsvFile.stream('huge.csv')) {
  process(row);
}

Features #

Dual-Path Architecture #

csv_plus uses two optimized paths for every operation:

• Batch path — FastEncoder / FastDecoder use codeUnits byte arrays, labeled loops, and first-byte type detection for maximum throughput
• Streaming path — CsvEncoder / CsvDecoder implement StreamTransformer with a chunked state machine for constant-memory processing

Automatic Type Inference #

Raw CSV strings are parsed into native Dart types automatically:

final rows = codec.decode('name,age,active\nAlice,30,true');
// rows[0] == ['name', 'age', 'active']  (String)
// rows[1] == ['Alice', 30, true]         (String, int, bool)

Disable with dynamicTyping: false to get all strings, or use specialized decoders:

codec.decodeStrings(csv)   // List<List<String>>
codec.decodeIntegers(csv)  // List<List<int>>
codec.decodeDoubles(csv)   // List<List<double>>
codec.decodeBooleans(csv)  // List<List<bool>>
codec.decodeFlexible(csv)  // Lenient: trims whitespace, recovers bad quotes

Configuration & Presets #

final codec = CsvCodec();            // Standard CSV (auto-detect on)
final excel = CsvCodec.excel();       // Semicolons + BOM for Excel
final tsv = CsvCodec.tsv();           // Tab-separated
final pipe = CsvCodec.pipe();         // Pipe-separated

// Or customize fully
final custom = CsvCodec(CsvConfig(
  fieldDelimiter: '::',
  quoteMode: QuoteMode.always,
  skipEmptyLines: true,
));

Schema Validation #

final schema = CsvSchema(columns: [
  ColumnDef(name: 'email', type: String, required: true, pattern: r'@'),
  ColumnDef(name: 'age', type: int, nullable: false),
]);

final errors = table.validate(schema);
final isValid = table.conformsTo(schema);

dart:convert Integration #

final adapter = codec.asCodec();    // Codec<List<List<dynamic>>, String>
final rows = adapter.decode(csv);

// Fuse with other codecs
final pipeline = adapter.fuse(utf8);

API Overview #

CsvCodec — Main Facade #

CsvTable — 50+ Methods #

CsvFile — File I/O #

Modular Imports #

Import everything with one line, or pick only what you need:

import 'package:csv_plus/csv_plus.dart';     // Everything
import 'package:csv_plus/codec.dart';         // Just CsvCodec
import 'package:csv_plus/table.dart';         // Just CsvTable
import 'package:csv_plus/io.dart';            // File I/O (dart:io)

Platform Support #

dart:io is isolated in io.dart — core encode/decode/table works everywhere.

Additional Information #

• Documentation: API Reference
• Issues: Report a bug
• Changelog: See CHANGELOG.md for version history

Contributing #

Contributions are welcome! Feel free to open issues or submit pull requests.

License #

MIT License — see LICENSE for details.

Made with ❤️ by Masum