ByteConverter #

High-performance byte unit converter for Dart with automatic caching and fluent API inspired by ByteSize library from C#.

Features #

• 🚀 High-performance with cached calculations
• 📦 Decimal (KB, MB, GB, TB, PB) and Binary (KiB, MiB, GiB, TiB, PiB) units
• 🔢 Math operations (+, -, *, /)
• 🔄 JSON serialization
• 💫 Fluent API with extensions
• 📐 Precise number formatting
• 🧮 Storage units (sectors, blocks, pages)
• 📈 Network transfer rates
• ⏱️ Time-based calculations
• ✨ NEW: BigInt support for arbitrary precision calculations
• 🌌 NEW: Large units (EB, ZB, YB) for massive data handling
• 🧩 NEW: Parsing from strings ("1.5 GB", "2GiB", "100 Mbps")
• 🧭 NEW: Auto human-readable formatting with SI/IEC/JEDEC and bits
• 🚦 NEW: DataRate helper (MB/s, Mbps) for network speeds
• 🗣️ NEW: Full-form unit parsing (e.g., "megabytes", "kibibits") and locale-friendly input (NBSP/commas/underscores)
• 🧷 NEW: Advanced formatting options (full-form names, custom separators/spacers, min/max fraction digits, signed, forced unit)
• 🧮 NEW: Extended IEC rate units through YiB/s

Installation #

dependencies:
  byte_converter: ^2.0.0

Quick Start #

// Basic Usage
final size = 1.5.gigaBytes;
print(size); // 1.5 GB

// Math Operations
final total = 1.5.gigaBytes + 500.megaBytes;
print(total); // 2 GB

// Binary Units
final ram = 16.gibiBytes;
print(ram.toHumanReadable(SizeUnit.GB)); // 17.18 GB

// Network Rates
final speed = 100.megaBytes; // size
print(speed.megaBitsPerSecond); // 800 Mbps (size to rate helpers)

// DataRate (network speeds)
final rate = DataRate.parse('100 Mbps');
print(rate.toHumanReadableAuto()); // 100 Mb/s
print(rate.toHumanReadableAuto(useBytes: true)); // 12.5 MB/s

// Storage Units
final disk = 4.kibiBytes;
print(disk.sectors); // 8 sectors

BigInt Support for Large Data #

For scenarios requiring arbitrary precision or handling extremely large values:

// BigInt Constructor
final dataCenter = BigByteConverter.fromExaBytes(BigInt.from(5));
print(dataCenter); // 5 EB

// Ultra-precise calculations
final precise = BigInt.parse('123456789012345678901234567890').bytes;
print(precise.asBytes); // Exact value preserved

// Large units support
final cosmic = BigByteConverter.fromYottaBytes(BigInt.one);
print(cosmic); // 1 YB

// BigInt Extensions
final huge = BigInt.from(1024).exaBytes;
print(huge.exaBytesExact); // Exact BigInt result

// Conversion between types
final normal = ByteConverter(1048576);
final big = BigByteConverter.fromByteConverter(normal);
final backToNormal = big.toByteConverter();

When to Use BigInt vs Regular Converter #

• ByteConverter: General use cases, good performance, handles up to ~15 digits precision
• BigByteConverter: Exact calculations, data center scales, scientific computing, crypto applications

Advanced Usage #

final data = ByteConverter.fromGigaBytes(1.5);

// Precision Control
print(data.toHumanReadable(SizeUnit.MB, precision: 3)); // 1536.000 MB

// Auto Humanize (SI/IEC/JEDEC + bits)
print(1024.bytes.toHumanReadableAuto(standard: ByteStandard.iec)); // 1 KiB
print(1024.bytes.toHumanReadableAuto(standard: ByteStandard.si)); // 1.02 KB

// Transfer Time
final downloadTime = data.downloadTimeAt(10.megaBitsPerSecond);
print(downloadTime); // Duration

// Storage Alignment
final aligned = data.roundToBlock();
print(aligned.isWholeBlock); // true

// BigInt exact arithmetic
// Parsing
final p1 = ByteConverter.parse('1.5 GB');
final p2 = BigByteConverter.parse('2TiB', standard: ByteStandard.iec);
final p3 = DataRate.parse('12.5 MB/s');
print(p1); // 1.5 GB
print(p2); // 2 TiB
print(p3); // 12.5 MB/s
final bigData = BigByteConverter.fromGigaBytes(BigInt.from(1000));
print(bigData.gigaBytesExact); // BigInt.from(1000) - no precision loss
print(bigData.gigaBytes); // 1000.0 - converted to double

// Unified parse API (auto normal vs big)
final parsed = parseByteSizeAuto('1 EB', thresholdBytes: 1e12);
if (parsed.isBig) {
  final b = (parsed as ParsedBig).value; // BigByteConverter
  print(b.toHumanReadableAuto());
} else {
  final n = (parsed as ParsedNormal).value; // ByteConverter
  print(n.toHumanReadableAuto());
}

// Reusable formatter options
final opts = ByteFormatOptions(
  standard: ByteStandard.iec,
  useBytes: true,
  precision: 1,
  showSpace: true,
);
print(ByteConverter(1024).toHumanReadableAutoWith(opts)); // 1 KiB
print(DataRate.megaBitsPerSecond(100).toHumanReadableAutoWith(opts)); // 100 Mb/s

Parsing and localization (NEW) #

The parser understands both symbols and full-form names, and is resilient to locale separators:

// Full-form units (bytes and IEC bytes/bits)
final a = ByteConverter.parse('1.5 megabytes'); // SI bytes
final b = ByteConverter.parse('2 kibibytes', standard: ByteStandard.iec); // IEC bytes
final c = ByteConverter.parse('10 megabits'); // SI bits -> bytes

// Locale-friendly numbers: NBSP, commas as decimal separator, underscores as group separators
final nb = ByteConverter.parse('1\u00A0234,56 KB'); // "1 234,56 KB" -> 1,234.56 KB
final us = ByteConverter.parse('12_345.67 MB');

Data rates support IEC units up to YiB/s and standard SI symbols:

final p = DataRate.parse('1 PiB/s', standard: ByteStandard.iec); // parses correctly
final e = DataRate.parse('2 EiB/s', standard: ByteStandard.iec);

Advanced humanize formatting (NEW) #

You can precisely control how sizes and rates are formatted via ByteFormatOptions or method parameters:

• fullForm: Use full names (e.g., "kilobytes", "megabits").
• fullForms: Override specific names (e.g., translate to "kilo-octets").
• separator: Decimal separator (e.g., ",").
• spacer: String between number and unit (e.g., "" for no space).
• minimumFractionDigits/maximumFractionDigits: Clamp fraction digits.
• signed: Always show a sign (+, -, or space for 0).
• forceUnit: Force a specific unit symbol instead of auto-scaling (supports bit units like Mb).

Examples:

// ByteConverter full-form output with custom override
final size = ByteConverter(1500);
print(size.toHumanReadableAutoWith(const ByteFormatOptions(
  standard: ByteStandard.si,
  useBytes: true,
  fullForm: true,
))); // 1.5 kilobytes

print(size.toHumanReadableAutoWith(const ByteFormatOptions(
  standard: ByteStandard.si,
  useBytes: true,
  fullForm: true,
  fullForms: {'kilobytes': 'kilo-octets'},
))); // 1.5 kilo-octets

// Custom separators/spacer, fixed fraction digits, signed, forced unit
final s = ByteConverter(1920);
print(s.toHumanReadableAuto(
  standard: ByteStandard.si,
  useBits: false,
  minimumFractionDigits: 1,
  maximumFractionDigits: 1,
  separator: ',',
  spacer: '',
  signed: true,
  forceUnit: 'KB',
)); // +1,9KB

// DataRate with forced unit and sign
final r = DataRate.megaBitsPerSecond(1920);
print(r.toHumanReadableAuto(
  standard: ByteStandard.si,
  useBytes: false,
  minimumFractionDigits: 1,
  maximumFractionDigits: 1,
  separator: ',',
  spacer: '',
  signed: true,
  forceUnit: 'Mb',
)); // +1920,0Mb/s

Unit Support #

Regular Units (ByteConverter) #

• Decimal: B, KB, MB, GB, TB, PB (auto humanize supports EB, ZB, YB too)
• Binary: B, KiB, MiB, GiB, TiB, PiB

Extended Units (BigByteConverter) #

• Decimal: B, KB, MB, GB, TB, PB, EB, ZB, YB
• Binary: B, KiB, MiB, GiB, TiB, PiB, EiB, ZiB, YiB

Upgrade notes #

This release introduces several new formatting and parsing capabilities while preserving backwards compatibility:

• New parsing paths accept full-form names (e.g., "megabytes", "kibibits") in addition to symbols.
• Locale-friendly number parsing: NBSP, underscores, and a trailing-decimal comma or dot are handled.
• Advanced formatting options are now available across ByteConverter, BigByteConverter, and DataRate humanize methods.
• DataRate.toHumanReadableAuto now delegates to the shared humanizer to ensure consistent separators, signs, and forced unit behavior.

Behavior notes:

• Forcing a unit with forceUnit disables auto-scaling. You can also force bit units like Mb.
• When signed is true, zero values display a leading space to keep alignment with positive/negative outputs.
• If both minimumFractionDigits and maximumFractionDigits are provided, the value is clamped to that range; otherwise precision is used.

ByteFormatOptions quick reference #

Tip: Use toHumanReadableAutoWith(ByteFormatOptions(...)) to apply the same formatting consistently across sizes and rates.

Performance #

• 🚀 Cached calculations for frequent operations
• 🧠 Lazy initialization for better memory usage
• 🔒 Immutable design for thread safety
• ⚡ Optimized string formatting

License #

This project is licensed under MIT License. Read about it here: MIT License

Features and bugs #

Please file feature requests and bugs here issue tracker.