Skip to content

Latest commit

 

History

History
83 lines (61 loc) · 3.5 KB

ANNOUNCE.rst

File metadata and controls

83 lines (61 loc) · 3.5 KB

Announcing Python-Blosc2 4.15.0

Python-Blosc2 4.15.0 is a major feature release introducing Menudet 0.1 beta (a little language for portable computation on arrays and tables), typed ordinary-expression evaluation by default, bundled C-Blosc2 3.6.1 and miniexpr 0.3.1, and faster frame serialization for containers with postfilters.

  • Introducing Menudet 0.1 beta. Menudet is a little language for portable computation on arrays and tables. Use @blosc2.menudet (MenudetFunction) for decorated authors, MenudetKernel for executable specializations, MenudetError for errors, and validate_menudet for source validation. Export normalized Menudet authoring as validated, standalone JSON artifacts with typed signatures and exact captured constants. Native miniexpr executes imports without reconstructing Python functions or falling back to Python. Supports checked integer arithmetic, operand-typed floating computation, control flow, fixed-width strings, logical array coordinates, and ordered block reductions.
  • Typed ordinary-expression evaluation by default. Ordinary lazy expressions now default to typed native execution where admitted, with bounded typed-graph/NumPy fallback otherwise. Intermediate computation dtypes remain separate from output and reduction accumulator dtypes. Unsupported typed execution never silently retries through the uniform evaluator after numerical execution starts.
  • Serializing containers with a postfilter. SChunk.to_cframe() and NDArray.to_cframe() now return stored chunks for every storage, also when a postfilter is set. Containers with a prefilter or postfilter no longer decompress and recompress every chunk during to_cframe(), dramatically improving serialization performance.
  • Bundled C-Blosc2 and miniexpr updates. Bundled C-Blosc2 has been updated to 3.6.1 and miniexpr to 0.3.1. blosc2.print_versions() now also reports the compiled miniexpr version.

Special thanks to Fabian Isensee for his outstanding contributions!

Install it with:

pip install blosc2 --upgrade   # if you prefer wheels
conda install -c conda-forge python-blosc2 mkl  # if you prefer conda and MKL

For more info, see the release notes at:

https://github.com/Blosc/python-blosc2/releases

What is Python-Blosc2?

Python-Blosc2 is a high-performance compressor, compute engine, and format for binary data containers that are portable and open-source. It comes with a lazy expression engine allowing for complex calculations on compressed data, whether stored in memory, on disk, or over the network (e.g., via Caterva2). It is especially optimized for storing and retrieving data from N-dimensional arrays (NDArray) and columnar tables (CTable), bringing a query/indexing layer too. The main use case is fast, compressed, out-of-core numerical data — especially when data is too large to fit comfortably in RAM.

More info: https://www.blosc.org/python-blosc2/getting_started/overview.html

Sources repository

The sources and documentation are managed through GitHub services at:

https://github.com/Blosc/python-blosc2

Python-Blosc2 is distributed using the BSD license, see https://github.com/Blosc/python-blosc2/blob/main/LICENSE.txt for details.

Mastodon feed

Follow https://fosstodon.org/@Blosc2 to get informed about the latest developments.

Enjoy!

  • Blosc Development Team Compress Better, Compute Bigger