libcompress library
Pure Dart implementations of common compression algorithms
This library provides pure Dart implementations of LZ4, Snappy, GZIP, and Zstd compression, with no native dependencies. It supports both in-memory (block) compression and stream-based processing.
Basic Usage
import 'package:libcompress/libcompress.dart';
// Block-based compression
final codec = CodecFactory.codec(CodecType.lz4);
final compressed = codec.compress(data);
final decompressed = codec.decompress(compressed);
// Stream-based compression
final streamCodec = CodecFactory.streaming(CodecType.lz4);
final compressedStream = streamCodec.compress(inputStream);
Available Codecs
- LZ4: Fast compression with good ratios. Supports levels 1-9 (level 9 uses HC mode).
- Snappy: Very fast compression. Supports raw block and framing formats.
- GZIP: Full DEFLATE implementation with GZIP framing (RFC 1952).
- Zstd: Zstandard (RFC 8878) without dictionaries, including compressed blocks with Huffman/FSE encoding and sequence encoding.
All codecs are CLI-compatible with their respective command-line tools (lz4, snzip, gzip, zstd).
Classes
- CodecFactory
- Factory for creating compression codecs
- CompressionCodec
- Base class for compression codecs
- CompressionOptions
- Base class for compression codec configuration options
- CompressionStreamCodec
- Base abstract class for streaming compression codecs
- Crc32
- CRC32 checksum implementation for data integrity verification
- Crc32c
- CRC32C (Castagnoli) checksum implementation
- GzipCodec
- GZIP compression codec (RFC 1952)
- GzipOptions
- GZIP-specific compression options
- GzipStreamCodec
- Gzip streaming codec
- Lz4Codec
- LZ4 compression codec
- Lz4Options
- LZ4-specific compression options
- Lz4StreamCodec
- LZ4 streaming codec
- NoopCodec
- Pass-through codec that performs no compression
- NoopStreamCodec
- Pass-through streaming codec that performs no compression
- SnappyCodec
- Snappy compression codec Native Dart implementation based on the Snappy specification
- SnappyIncrementalDecoder
-
Incremental Snappy framing decoder: parses and decodes one framing chunk at
a time. Chunk-type validation (incl. the leading stream identifier) is
enforced by
SnappyStreamDecoder. - SnappyOptions
- Snappy-specific compression options
- SnappyStreamCodec
- Snappy streaming codec
- ZstdBlockHeader
- Parsed Zstd block header information
- ZstdCodec
- Zstandard (Zstd) compression codec
- ZstdFrameHeader
- Parsed Zstd frame header information
- ZstdFrameHeaderDescriptor
- Frame header descriptor flags parsed from first header byte
- ZstdOptions
- Zstd-specific compression options
- ZstdStreamCodec
- Zstd streaming codec
Enums
- CodecMode
- Codec capability modes
- CodecType
- Supported compression codec types
- CompressionLevel
- Compression level presets for common use cases
- ZstdBlockType
- Block types in Zstandard format
Constants
- defaultMaxBufferSize → const int
- Default maximum buffer size for stream decoders (64MB)
- gzipDefaultMaxDecompressedSize → const int
- Default maximum decompressed size for GZIP (256 MB)
- lz4BlockSize1M → const int
- lz4BlockSize256K → const int
- lz4BlockSize4M → const int
- lz4BlockSize64K → const int
- Standard LZ4 frame block sizes
- lz4DefaultBlockSize → const int
- lz4DefaultMaxBufferSize → const int
- Default maximum buffer size for stream decoders (64MB)
- snappyDefaultMaxBufferSize → const int
- Default maximum buffer size for stream decoders (64MB)
- zstdDefaultMaxBufferSize → const int
- Default maximum buffer size for stream decoders (64MB)
- zstdDefaultMaxDecompressedSize → const int
- Default maximum decompressed size (256 MB) Prevents memory exhaustion from malicious contentSize headers
- zstdMagicNumber → const int
- Zstandard magic number File bytes: 28 B5 2F FD (little-endian) When read as uint32LE: 0xFD2FB528
- zstdMaxBlockSize → const int
- Maximum block size (128 KB)
- zstdSkippableFrameMagicBase → const int
-
Base magic number for skippable frames (0x184D2A50 + n, where n in
0, 15) - zstdSkippableFrameMagicMask → const int
- Mask used to detect skippable magic numbers.
Functions
-
guardFormat<
T> (T body(), CompressionFormatException wrap(String message)) → T -
Runs
body, converting the input-validation errors a decoder throws on malformed data into a codec-specific CompressionFormatException viawrap:StateError(e.g. exhausted bit reader),ArgumentError(incl. its subtypesRangeError/IndexError, e.g. an out-of-window back-reference), plainFormatException, andOutputLimitException(a size limit was hit). -
validateLevel(
int level, int min, int max) → void -
Validates
levelis in[min, max], throwing ArgumentError otherwise. Shared by codec constructors so they fail fast at the API boundary rather than later during the first compression. -
validateOptionalPositive(
int? value, String name) → void -
Validates a nullable limit is either null (unlimited) or
> 0. -
validatePositive(
int value, String name) → void -
Validates
valueis> 0. -
validateRange(
int value, int min, int max, String name) → void -
Validates
valueis in[min, max](e.g. a block/chunk size).
Exceptions / Errors
- CompressionFormatException
- Base exception class for compression format errors
- DeflateFormatException
- Exception thrown during DEFLATE compression/decompression
- GzipFormatException
- Exception thrown when invalid GZIP data is encountered
- Lz4FormatException
- Exception thrown when invalid LZ4 data is encountered
- SnappyFormatException
- Exception thrown when invalid Snappy data is encountered
- ZstdFormatException
- Exception thrown when invalid Zstandard data is encountered