New releases of HSDS and h5pyd are out!
The newest HSDS release is 0.9.1. You can get it with pip install hsds or get the Docker image using: docker pull hdfgroup/hsds:0.9.1.
New features in this release:
- Improved support for links to contigous HDF5 datasets
- Added support for fancy selection with multiple dimensions
- Numpy 2.0 support
- Codespaces integration
- Support “Where in” queries
- Support Hyperchunks for multidimensional datasets
- Multi-link methods - get/update multiple links in one request
- Multi-attribute methods - get/update multiple attributes in one request
- Support bitshuffle compression
- Added support for fixed width UTF8 strings
- Enable dataset shape reduction
- Support broadcast operations for dataset reads and writes
Along with the HSDS release, there’s a new h5pyd release as well: version 0.19. You can fetch it from PyPI using: pip install h5pyd.
New and improved h5pyd features:
- Added devcontainer files for running h5pyd (with it’s own HSDS) in Github codespaces
- Several new examples
- Numpy 2.0 support
- Added --h5image option to hsload to load HDF5 files as one dataset
- Enable Fancy selection with multiple coordinates
- Added SWMR compatibility
- Added support for multi-dimensional hyperchunks
- Added MultiManager class (for parallel dataset read/writes)
- Added --names option for hsls
If you have questions about any of the new features, feel free to post questions here.