0.7.0

• 326: Added tests for Python 3.6 - 3.10 for Linux and 3.7 - 3.9 for Windows. Updated devenv and CICD to Python 3.8.
• 321: Moved out Fugue SQL to https://github.com/fugue-project/fugue-sql-antlr, removed version cap of antlr4-python3-runtime
• 323: Removed version cap of DuckDB
• 334: Replaced RLock with SerializableRLock
• 337: Fixed index warning in fugue_dask
• 339: Migrated execution engine parsing to triad conditional_dispatcher
• 341: Added Dask Client to DaskExecutionEngine, and fixed bugs of Dask and Duckdb

Improve Dask

0.7.0.dev5

Migrate execution engine parsing to triad conditional_dispatcher, add…

Remove Dask warnings

0.7.0.dev4

Fix index warning in fugue_dask (#338)

Remove DuckDB cap, replace RLock with SerializableRLock

0.7.0.dev3

Remove duckdb cap (#336)

Move Fugue SQL out of the main project

Refactor fugue sql (#327)

* refactor fugue sql

* update

* update

* update

* update sql dependency

* update sql dependency

* update

* update

Support Py 3.6 - 3.10

Update devenv and cicd (#326)

* update devenv and cicd

* update cicd

* fix 3.9 incompatibility

* update dependencies

* update

* update

0.6.6

• Create a hybrid engine of DuckDB and Dask
• Save Spark-like partitioned parquet files for all engines
• Enable DaskExecutionEngine to transform dataframes with nested columns
• A smarter way to determine default npartitions in Dask
• Support even partitioning on Dask
• Add handling of nested ArrayType on Spark
• Change to plugin approach to avoid explicit import
• Fixed Click version issue
• Added version caps for antlr4-python3-runtime and duckdb as they both released new versions with breaking changes.

Save with partition

0.6.6.dev3

[WIP] Fix issue #285 : save hive partitioned dataset using NativeExec…

Change to plugin approach

Fugue plugin (#311)

* plugin

* update

0.6.6 Dev release

0.6.6.dev1

Add handling for spark array type (#307)