zetasql-toolkit

Changelog:

• Avoid NullPointerException when quoting name paths of an unnest expression (#52)
  • Closes #51

Dependency updates

• com.google.cloud:libraries-bom from 26.37.0 to 26.39.0
• com.google.cloud.tools:jib-maven-plugin from 3.3.2 to 3.4.2

Plugin updates

• org.apache.maven.plugins:maven-jar-plugin from 3.3.0 to 3.4.1
• org.apache.maven.plugins:maven-source-plugin from 3.3.0 to 3.3.1
• org.apache.maven.plugins:maven-javadoc-plugin from 3.5.0 to 3.6.3
• org.apache.maven.plugins:maven-surefire-plugin from 3.1.0 to 3.2.5
• org.apache.maven.plugins:maven-gpg-plugin from 3.1.0 to 3.2.4
• org.codehaus.mojo:properties-maven-plugin from 1.1.0 to 1.2.1

Release version 0.5.0 of the ZetaSQL Toolkit

Changelog:

• Use the DOUBLE typekind for FLOAT64 columns
  • Closes #25
• Extract column level lineage for SELECT statement (#29)
  • Closes #28
• Make the ZetaSQLTypeParser case insensitive (#35)
  • Closes #32
• Add reflection-based patching of GRPC's default max nesting depth (#36)
  • Closes #31
• Rewrite queries to fully quote resource name paths before analysis
  • For example, FROM project.dataset.table is now rewritten to FROM `project.dataset.table`
  • This has no impact to end users, but massively simplifies how the ZetaSQL toolkit builds catalogs
• Remove unnecessary slf4j dependency from zetasql-toolkit-core
  • Closes #24

Release version 0.4.0 of the ZetaSQL Toolkit

Changelog:

• Properly interpret BigQuery resources that use the JSON type (#19)
  • Closes #17
• Compile against JDK 8 instead of JDK 11 (#20)
• Extend support for column-level lineage (#22) with:
  • Lineage for CREATE VIEW statements
  • Field-level lineage for STRUCT columns and operations
  • Proper handling of scoping for WITH clauses
  • Proper handling of columns generated in a ResolvedSetOperationScan (Closes #21)

Release version 0.4.0 of the ZetaSQL Toolkit

Changelog:

• Return AnalyzedStatement objects from ZetaSQLToolkitAnalyzer.analyzerStatements()
  • AnalyzedStatement objects contain the parsed statement and, optionally, the resolved statement
  • Statements are only resolved when they are supported by the ZetaSQL Analyzer. Resolution will stop altogether for a script after finding any scripting constructs apart from variable declaration and assignments.
• Resolve variable declaration and assignments when performing analysis.
  • This will validate assignments to variables and properly define them in the Catalog
• Support building testing catalogs using a JSON representation of resources
  • See CatalogResources::fromJson, BigQueryCatalog::usingResources, SpannerCatalog::usingResources
• Add initial support for understanding column-level lineage produced by analyzed queries
  • See the ExtractColumnLevelLineage example
• Add method BigQueryCatalog::addAllResourcesUsedInQuery
• Separate the project into multiple artifacts
  • zetasql-toolkit-core containing the core of the Toolkit
  • zetasql-toolkit-bigquery containing all BigQuery-specific features
  • zetasql-toolkit-spanner containing all Spanner-specific features

Release version 0.3.3 of the ZetaSQL Toolkit

Changelog:

• Support bigquery resources with dashes (-) in their names

Release version 0.3.2 of the ZetaSQL Toolkit

Changelog:

• add methods in the BigQueryCatalog and the SpannerCatalog now ignore resources that have already been added to said catalog

Release version 0.3.1 of the ZetaSQL Toolkit

Changelog:

• Bugfix - Ensure catalog mutations are always applied during analysis (#6)

Initial release for the ZetaSQL Toolkit through Maven Central