Changelog
0.1.3 19th April, 2024¶
#53 - Docs parser is now more forgiving, empty descriptions are tolerated and normal scaladoc syntax is allowed
#50 - typedExpressionRunner - audited capture of expressions with the same type
#51 - Spark 3.5.0 support - NOTE ViewLoaderAnalysisException and MissingViewAnalysisException now have Exception causes
#27 - Delta 3.0.0 (Spark 3.5.0) support - compatible version
#55 - DBR 14.0/1 - Snake Yaml 2.0 support
#58 - Migrate custom runtime usage to Shim
#59 - DBR 13.3 LTS support
#57 - DBR 14.3 support
#61 - Use sparkutils frameless for 3.5, 13.3, 14.x builds - Due to encoding and shim changes this frameless fork version is not binary compatible with typelevel frameless proper
#62 - SPARK-47509 workaround for Subqueries in lambdas - most common patterns are supported with 4.0 / 14.3 DBR
#63 - Use actual struct functions where possible for drop_field/update_field functions - required due to 14.3 DBR introduced plan on local relations
0.1.2.1 4th September, 2023¶
Maven Central build issues, code wise the same as 0.1.2.
0.1.2 4th September, 2023¶
#48 - Bug fix - Enable Sub Queries in all runner types
0.1.1 9th July, 2023¶
#42 - Improve expression runner to store yaml, unlike json, to_yaml and from_yaml allow complete support for roundtripping of Spark data types
0.1.0 10th June, 2023¶
#29 - Quality OptimzerRule's run with Databricks sql display
#35 - agg_expr and associated lambda support in the functions package
#36 - improved update_field, added drop_field based on the Spark withField (3.4.1 impl)
#34 - simplified quality package usage, column functions are now under the functions package.
#32 - expressionRunner - saves the results of expressions as strings, suitable for aggregate statistics
#28 - rule_result function - retrieves a rule result directly from a dq or expressionRunner result
#15 - Addition of the loadXConfigs and loadX functions for maps and blooms, simplifying configuration management
#24 - Remove saferId / rowid functions - use unique_id where required
#18 - ViewLoader - simple view configuration via DataFrames
#30 - 3.3.2 and 3.4.1 builds - simple version bumps
#20 - 3.5.0 starting support
0.0.3 17th June, 2023¶
#25 - Use builtIn function registration by default - allows global views to be created using Quality functions
0.0.2 2nd June, 2023¶
#16 - Remove winutils requirements for testing and usage
#13 - Support 3.4's sub query usage in rules/trigger, output expressions and lambdas
#12 - Introduce the use of underscores instead of relying on camel case for function sql names, inline with Spark built-in functions
#10 - Base64 functions added for RowID encoding and decoding via base64 (more suitable for BI tools)
#9 - Add AsymmetricFilterExpressions with AsUUID and IDBase64 implementation, allows expressions used in field selects to be reversed, support added for optimiser rules through the SparkExtension
#8 - Add set syntax for easier defaulting sql, removing duplicative cruft from intention
#7 - SparkSessionExtension to auto register Quality functions - does not work in 2.4, starting with this release 2.4 support is deprecated
#6 - Simple as_uuid function
#5 - Spark 3.4 and DBR 12.2 LTS support
#4 - comparableMaps / reverseComparableMaps functions, allowing map comparison / set operations (e.g. sort, distinct etc.)
0.0.1 8th March, 2023¶
Initial OSS version.
(many internal versions in between)
the Quality exploration starts 25th April, 2020¶
Start of investigations into how to manage DQ more effectively within Spark and the mesh platform.
Created: April 22, 2024 08:24:32