streamly-core

Streaming, parsers, arrays, serialization and more

For upgrading to streamly-0.9.0+ please read the <https://github.com/composewell/streamly/blob/streamly-0.10.0/docs/User/Project/Upgrading-0.8-to-0.9.md Streamly-0.9.0 upgrade guide>. Streamly is a high-performance, beginner-friendly standard library for Haskell. It unifies streaming with list transformers and logic programming; unifies streaming with concurrency and reactive programming; unifies arrays with ring arrays, text, bytestring and vector use cases; unifies arrays with builders and binary serialization; generalizes parsers to any input type and unifies attoparsec, parsec use cases with better performance; provides streaming fileIO — all with a clean, consistent, well-integrated and streaming enabled API. Streams are designed to have a list like interface — no steep learning curve, no complex types. Streamly is designed to build general purpose applications in a truly functional manner, from simple hello-world to advanced high-performance systems. The design emphasizes simplicity, modularity, and code reuse with minimal building blocks. Performance is on par with C, tuning is easy, and it’s hard to get it wrong. Streamly is serial by default, with seamless declarative concurrency that scales automatically when needed. It provides prompt and safe resource management, works well with other streaming libraries as well as core libraries like bytestring and text, and is backed by solid documentation. @streamly-core@ is a Haskell standard library built on top of @base@ and GHC boot libraries only. Stream processing abstractions include streams, scans, folds, parsers; and console I/O, file I/O; text processing. Array abstractions include pinned, unpinned, mutable, immutable, boxed and unboxed arrays, and ring arrays. Builders, binary serialization, and deserialization are built-in features of arrays. This package provides a high-performance, unified and ergonomic alternative to many disparate packages, such as @streaming, pipes, conduit, list-t, logict, foldl, attoparsec, array, primitive, vector, vector-algorithms, binary, cereal, store, bytestring, text, stringsearch, interpolate, filepath, and path@. Performant. Unified. Modular. Powerful. Simple. Learn more at <https://streamly.composewell.com the streamly website>.

Install

cabal install streamly-core