v0.3.1: Install and query speed improvements.

What's Changed

• docs: add quickstart and note gcsfs requirement by @ahuang11 in #188
• feat: skip per-dim discovery scans when query is an unfiltered scan by @ghostiee-11 in #193
• Build abi3 wheels so installs don't recompile on macOS/Windows by @alxmrs in #198
• Geospatial SQL benchmark suite: prove core geo/climate ops are relational by @alxmrs in #177
• test: peak-memory integration tests for the lazy round-trip by @ghostiee-11 in #199

Full Changelog: v0.3.0...v0.3.1

v0.3.0: Adding a `result.to_dataset()` interface.

What's Changed

• Support heterogeneous datasets with scalar variables in from_dataset by @ghostiee-11 in #169
• Add lazy SQL → xarray round-trip via XarrayDataFrame.to_dataset (closes #58) by @ghostiee-11 in #167
• Make dask optional when reading Zarr datasets by @alxmrs in #176

Full Changelog: v0.2.3...v0.3.0

v0.2.3: Dataset to multiple table support

What's Changed

• indentation + ruff by @ahuang11 in #164
• Made new xarray context behavior to register multiple tables from a dataset by @eparkle in #166

New Contributors

• @eparkle made their first contribution in #166

Full Changelog: v0.2.2...v0.2.3

v0.2.2: Bugfixes, better Xarray support

What's Changed

• Migrate tests by @ahuang11 in #155
• Skip string coordinates in partition bound computation by @ghostiee-11 in #153
• Docs infra by @ahuang11 in #156
• Documentation fixes: faster, correct URL. by @alxmrs in #157
• Fix null conversion by @ahuang11 in #151
• Fix documentation formatting issues by @alxmrs in #158
• New attribution. by @alxmrs in #159
• Add cftime support by @ahuang11 in #149
• Increment version in prep for the v0.2.2 release by @alxmrs in #160

New Contributors

• @ahuang11 made their first contribution in #155
• @ghostiee-11 made their first contribution in #153

Full Changelog: v0.2.1...v0.2.2

v0.2.1: Xarray-SQL Performance Improvements.

What's Changed

• fix: perf_tests scripts to use pyarrow or LazyArrowStreamTable by @ljstrnadiii in #111
• replace typing module with native collection types by @Ebraam-Ashraf in #120
• Xarray-specific filter predicate pushdown. by @alxmrs in #113
• Convert a Xarray Dataset partition to a pyarrow record batch directly (without going through pandas). by @alxmrs in #134
• Iterating through block into multiple record batches per partition. by @alxmrs in #135
• Iterate through metadata instead of materializing it all up front. by @alxmrs in #136
• Improve concurrency by releasing the GIL between batches. by @alxmrs in #137
• Column Projection Pushdown by @alxmrs in #119
• Upgrade to DataFusion 52.0.0 by @alxmrs in #142
• Increase cargo version (minor). by @alxmrs in #143

New Contributors

• @ljstrnadiii made their first contribution in #111
• @Ebraam-Ashraf made their first contribution in #120

Full Changelog: v0.2.0...v0.2.1

v0.2.0: Native DataFusion integration, tighter public API

What's Changed

• Dask dependency is only needed for tests. by @alxmrs in #80
• Convert tests from unittest to pytest format by @RohanDisa in #76
• Added a roadmap to the readme. by @alxmrs in #90
• Add pre-commit hooks with pyink and mypy by @jayendra-info in #91
• Adding back in lazy xr.Dataset and metadata parsing fix. by @alxmrs in #94
• Bump pypa/gh-action-pypi-publish from 1.4.2 to 1.13.0 in /.github/workflows by @dependabot[bot] in #96
• Lazy datafusion registration via custom table provider by @alxmrs in #100
• README update given deeper integration with DataFusion by @alxmrs in #109
• Parallel Xarray record batch reading via blocks partition factory. by @alxmrs in #108
• Adding Python to the manylinux arm64 release build by @alxmrs in #114
• Explicitly getting tags by @alxmrs in #115
• Pair down size of final wheel via cargo exclude. by @alxmrs in #116
• Improvements inspired by latest datafusion-python build file. by @alxmrs in #117

New Contributors

• @RohanDisa made their first contribution in #76
• @jayendra-info made their first contribution in #91
• @dependabot[bot] made their first contribution in #96

Full Changelog: v0.1.2...v0.2.0

v0.1.2: Documentation and minor improvements

What's Changed

• Minor documentation and typing update. by @alxmrs in #64
• Comprehensive SQL test cases. by @alxmrs in #65
• Refreshed the documentation by @alxmrs in #72
• A few more copy/edits. by @alxmrs in #78
• Lazy read_xarray with iterables and parsing metadata. by @alxmrs in #81

Full Changelog: v0.1.1...v0.1.2

v0.1.1

What's Changed

• More efficient read_xarray implementation with from_map_batched by @alxmrs in #63

Full Changelog: v0.1.0...v0.1.1

v0.1.0 - Migrating to DataFusion

Now that dask-sql is unstable, this repo is experimenting with just using datafusion directly.

What's Changed

• Added install instructions to README.md by @alxmrs in #42
• Added the incredibly helpful collaborators from the Pangeo discord. by @alxmrs in #43
• Fix typo by @davidbrochart in #48
• Reformat code with latest version of pyink. by @alxmrs in #51
• Dataframe includes non-dimension columns. by @alxmrs in #50
• chunks by @RichardScottOZ in #56
• Orient the project around uv. by @alxmrs in #60
• Migrate to DataFusion. by @alxmrs in #61
• Refresh pypy release automation. by @alxmrs in #62

New Contributors

• @davidbrochart made their first contribution in #48
• @RichardScottOZ made their first contribution in #56

Full Changelog: v0.0.2...v0.1.0

Initial Release of xarray-sql.

Initial release of xarray-sql. Users can now query data in Xarray via SQL.

What's Changed

• Rename to xarray-sql. by @alxmrs in #41

Full Changelog: v0.0.1...v0.0.2