Pyuns

Pyuns is a lightweight, interpreted domain-specific language designed for data transformation and simple ETL tasks within Python environments. Its syntax is Python-like, prioritizing readability and frictionless integration with existing Python code. Pyuns programs run inside a small standalone interpreter or are embedded in a Python process via an API, enabling users to leverage Python libraries when needed.

Originating from a community-driven project in 2018, Pyuns aimed to provide a compact DSL for non-programmers to express data pipelines. The first public release appeared in 2019, followed by major updates that added richer data types, error handling, and streaming capabilities.

Core features include a minimal, Python-like syntax for defining pipelines using a pipe operator, first-class functions, and built-in support for common data sources and sinks such as CSV, JSON, and REST endpoints. It supports dynamic typing with optional type hints, straightforward error propagation, and tight interoperability with Python data structures. The runtime emphasizes lazy evaluation of pipelines and can exploit parallel execution on multi-core systems.

In practice, a Pyuns script reads a CSV file, filters rows, and writes results to another CSV, expressed in concise lines. Pyuns favors declarative pipeline composition, allowing users to replace or reorder steps without modifying surrounding code.

Reception has been modest, with adoption in education and data science communities. It is valued for its approachable syntax but has a smaller ecosystem than Python itself or larger DSLs. Ongoing development focuses on expanding library support and improving performance.

See also: Domain-specific languages and Python integration.