This is a release to fix some packaging issues with using uv to install py_dataset post libdataset depreciation.

This release includes some documentation improvements and a patch to include dsquery support via the dataset.query method.

Full Changelog: v2.2.3...v2.2.3.1

This release of py_dataset no longer uses the C Shared library. Instead is requires that dataset v2.2.3 or better is installed on the system. It now wraps the command line dataset comment.

Several functions now return False due to the feature being depreciated in the past. These include import/export of CSV data from frames, sync actions with csv data, frame grids, create_objects and update_objects (notice the plural). object_path always returns an empty string since Pairtree storage is no longer the default and will be phased out.

Full Changelog: v2.1.2...v2.2.3

Maintenance release for recent version of dataset.

This release updates py_dataset to version v2.1.1 of libdatast from the dataset project. See release-notes-v2.1.1.md for changes. The are a number of function signatures that have update, some behavioral change in methods, depreciated functions and a few new ones. The versioning model changed in dataset between v1 and v2, this release reflects the new model. With it comes support for using SQL stores for the JSON objects as well as experimental versioning of objects and attachments based on a common increment setting for the semver.

This release includes version 1.0.1 of dataset. It adds support for M1 macs and removes sort and filter options for keys.

This is a minor maintenance release, fixed comments in source code to reflect the behaviors for "frame refresh" versus "frame reframe" in frames.go and libdataset.go. Also updated documentation adding back individual pages for frame "refresh" and "reframe".

This release includes updated documentation is available via pypi. The version number has been increased to match the underlying dataset shared library version number.

Added missing functions from libdataset.go (e.g. is_open(), close_all(), open(), frame()). Renamed make_objects() to create_objects() to be more consistent with naming scheme.

This fixes a bug in fame_delete() calls to libdataset where the frame was being cleared and not deleted.