zstd, short for Zstandard, is a fast lossless compression algorithm, targeting real-time compression scenarios at zlib-level and better compression ratios. The zstd compression library provides in-memory compression and decompression functions. The library supports compression levels from 1 up to max_compression_level () (currently 22). Levels >= 20, labeled `--ultra`, should be used with caution, as they require more memory.

Compression can be done in:

  • a single step (described as Simple API)
  • a single step, reusing a context (described as Explicit context)
  • unbounded multiple steps (described as Streaming compression)

The compression ratio achievable on small data can be highly improved using a dictionary in:

  • a single step (described as Simple dictionary API)
  • a single step, reusing a dictionary (described as Bulk-processing dictionary API)

Also see LZ4.

exception Error of string
exception Content_size_unknown
exception Content_size_error
exception Not_enough_capacity of int
exception Decompressed_size_exceeds_max_int of Core_kernel.Int64.t
module Output : sig ... end
module Input : sig ... end
val decompressed_size : Input.t -> Core_kernel.Int64.t

Returns the decompressed size of a message. Since decompressed size is an optional field of compressed message, it might raise Content_size_unknown, Content_size_error or Error.

module Compression_context : sig ... end
module Decompression_context : sig ... end
module Simple : sig ... end
module With_explicit_context : sig ... end

This module implements compression and decompression with explicitly managed contexts. When compressing or decompressing many times, it is recommended to allocate a context just once, and re-use it for each successive compression operation. This will make workload friendlier for system's memory.

module Dictionary : sig ... end
module Simple_dictionary : sig ... end
module Bulk_processing_dictionary : sig ... end
module Streaming : sig ... end

The functions exposed in this module differ from the rest of the library, and operates on stream of data (rather than on single messages). Therefore, the API is more complex, and does not use the Input module.