From 61922d7cba040ac5ec40d20cab05e40f5fbc9e6c Mon Sep 17 00:00:00 2001 From: Francesc Alted Date: Wed, 2 Oct 2024 17:35:43 +0200 Subject: [PATCH] Update latest release in doc --- ANNOUNCE.rst | 26 ++++++++++++++------------ RELEASING.rst | 6 ++++++ doc/python-blosc2.rst | 4 ++-- 3 files changed, 22 insertions(+), 14 deletions(-) diff --git a/ANNOUNCE.rst b/ANNOUNCE.rst index ee4b8c97..50d3d803 100644 --- a/ANNOUNCE.rst +++ b/ANNOUNCE.rst @@ -1,32 +1,34 @@ -Announcing Python-Blosc2 3.0.0-beta.3 +Announcing Python-Blosc2 3.0.0-beta.4 ===================================== -The Blosc development team is pleased to announce the second beta release of -Python-Blosc2 3.0.0. Here, we have fine tuned the performance of the new -evaluation engine, and extended it to operate with other instances than NDArray, -like the `C2Array` class and the new one `Proxy` (access to remote data with caching). -Documentation has been improved and we have added more examples and tutorials. -As always, we would like to get feedback from the community before the final release. +The Blosc development team is pleased to announce the fourth beta release of +Python-Blosc2 3.0.0. Here, documentation has been improved quite a lot and +we have added more examples and tutorials (thanks NumFOCUS for sponsoring this). +Also, there are new CParams, DParams and Storage dataclasses that allow for +a more flexible and powerful way to set parameters for the Blosc2 compressor. -Now, you can evaluate expressions like `a + sin(b) + 1` where `a` and `b` -are NDArray instances. This is a powerful feature that allows for +In new 3.0 release, you can evaluate expressions like `a + sin(b) + 1` where +`a` and `b` are NDArray instances. This is a powerful feature that allows for efficient computations on compressed data, and supports advanced features like reductions, filters, user-defined functions and broadcasting (still -in beta). See this `example `_. +in beta). See this +`example `_. Also, we have added support for memory mapping in `SChunk` and `NDArray` instances. This allows to map super-chunks stored in disk and access them as if they were in memory. When combined with the evaluation engine, this feature allows for very -good performance when working with large datasets. See this `benchmark `_ +good performance when working with large datasets. See this +`benchmark `_ (as it is a Jupyter notebook, you can easily run it in your own computer). Last, but not least, we are using NumPy 2.x as the default for testing procedures and builds. This means that our wheels are built against NumPy 2, so in case you want to use NumPy 1.x, you will need to use NumPy 1.23.0 or later. +As always, we would like to get feedback from the community before the final release. We are providing binary wheels that you can easily install from PyPI with: - pip install blosc2==3.0.0b1 + pip install blosc2==3.0.0b4 For more info, you can have a look at the release notes in: diff --git a/RELEASING.rst b/RELEASING.rst index d7c28c36..a63c0cfb 100644 --- a/RELEASING.rst +++ b/RELEASING.rst @@ -58,6 +58,12 @@ Tagging * Make sure that the tag is passing the tests in continuous integration (this may take more than an hour). +* Update the latest release in the ``doc/python-blosc2.rst`` file with the new version + number (and date?). Do a commit:: + + git commit -a -m "Update latest release in doc" + git push + * Go to ``Blosc/blogsite`` repo, then to "Actions", click on the most recent workflow run (at the top of the list), and then click on the "Re-run all jobs" button to regenerate the documentation and check that it has been diff --git a/doc/python-blosc2.rst b/doc/python-blosc2.rst index d1dd2191..ea59f221 100644 --- a/doc/python-blosc2.rst +++ b/doc/python-blosc2.rst @@ -6,9 +6,9 @@

Version 3.0.0 beta3 released! + style="font-size: 1.5em;">Version 3.0.0 beta4 released on 2024-10-02! - pip install blosc2==3.0.0b3 + pip install blosc2==3.0.0b4

.. raw:: html