# PyTables 3.9.0 is released!

**URL:** <https://forum.hdfgroup.org/t/pytables-3-9-0-is-released/11590>\
**Category:** HDF5 Ancillary Tools\
**Created:** [October 6, 2023, 12:16pm UTC](https://forum.hdfgroup.org/t/pytables-3-9-0-is-released/11590 "2023-10-06T12:16:39Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![faltet](https://forum.hdfgroup.org/user_avatar/forum.hdfgroup.org/faltet/32/1746_2.png) [@faltet](https://forum.hdfgroup.org/u/faltet)\
**Post date:** [October 6, 2023, 12:16pm UTC](https://forum.hdfgroup.org/t/pytables-3-9-0-is-released/11590/1 "2023-10-06T12:16:39Z")

</div>

We are happy to announce PyTables 3.9.0.

# What’s new

After nearly one year since the previous release, PyTables 3.9.0  
includes an assortment of improvements and fixes contributed by many  
people. A big thank you to everyone involved in keeping PyTables alive  
& kicking!

This release enhances chunked and extensible arrays (CArray/EArray)  
compressed with Blosc2 with the optimized read operation added in  
PyTables 3.8.0 (based on HDF5 direct chunking). Slicing such  
multidimensional arrays (even along inner dimensions) has been greatly  
sped up thanks to Blosc2 NDim 2-level partitioning of chunks into  
smaller multidimensional blocks, which avoids the decompression of whole  
chunks. This is enabled by providing the Blosc2 cframe with a b2nd  
metalayer when writing each array chunk. Please check out  
“Multidimensional slicing and chunk/block sizes” in the optimization  
tips section of the User’s Guide for more information. This development  
was funded by a NumFOCUS grant, with the support of the Blosc project.

More info about the optimized read operation on hyper-slices in our  
recent presentation at the 2023 European HDF User Group (HUG) plugins  
and data compression summit:

> **[2023 European HDF User Group (HUG) plugins and data compression summit](https://indico.desy.de/event/39343/contributions/151494/)**
>
> On September 19-21, the 2023 European HDF5 Users Group (HUG) summit was was held both in-person and virtually at DESY in Hamburg, Germany. Thanks to all those who attended and presented! Slides from the presentations are available here, and videos...

PyTables 3.9.0 also adds support for column-level attributes (e.g. for  
units) via their `attrs` field. This has been in the queue for more  
than two years, but it’s finally here!

PyTables now supports the forthcoming Python 3.12, with binary wheels  
available and extensive automatic testing. Please note that wheels and  
testing for Python 3.8 have been dropped due to issues with  
dependencies. Users or distribution packagers may still use Python 3.8,  
but they are encouraged to perform tests on their own.

Regarding Blosc2 compression support, PyTables may now use either the  
python-blosc2 package (which is handy for packaged Python wheels), or  
just the c-blosc2 library (which may be preferable for wider software  
distributions). Newer versions of Blosc2 and other dependencies are  
required that fix various issues.

In case you want to know more in detail what has changed in this  
version, please refer to: [PyTables Release Notes — PyTables 3.10.2 documentation](http://www.pytables.org/release_notes.html)

You can install it via pip or download a source package with generated  
PDF and HTML docs from:

> **[Release Release v3.9.0 · PyTables/PyTables](https://github.com/PyTables/PyTables/releases/v3.9.0)**
>
> Changes from 3.8.0 to 3.9.0
> New features
> 
> Apply optimized slice read to Blosc2-compressed CArray and EArray, with Blosc2 NDim 2-level partitioning for multidimensional arrays (#1056). See "Multidi...

For an online version of the manual, visit:

> **[PyTables User’s Guide — PyTables 3.10.2 documentation](http://www.pytables.org/usersguide/index.html)**

# What it is?

PyTables is a library for managing hierarchical datasets and  
designed to efficiently cope with extremely large amounts of data with  
support for full 64-bit file addressing. PyTables runs on top of  
the HDF5 library and NumPy package for achieving maximum throughput and  
convenient use. PyTables includes OPSI, a new indexing technology,  
allowing to perform data lookups in tables exceeding 10 gigarows  
(10\*\*10 rows) in less than a tenth of a second.

# Resources

About PyTables: [http://www.pytables.org](http://www.pytables.org)

About the HDF5 library: [The HDF5® Library & File Format - The HDF Group - ensuring long-term access and usability of HDF data and supporting users of HDF technologies](http://hdfgroup.org/HDF5/)

About NumPy: [http://numpy.scipy.org/](http://numpy.scipy.org/)

# Acknowledgments

Thanks to many users who provided feature improvements, patches, bug  
reports, support and suggestions. See the `THANKS` file in the  
distribution package for a (incomplete) list of contributors. Most  
specially, a lot of kudos go to the HDF5 and NumPy makers.  
Without them, PyTables simply would not exist.

# Share your experience

Let us know of any bugs, suggestions, gripes, kudos, etc. you may have.

* * *

**Enjoy data!**

– The PyTables Developers
