All notable changes to this project will be documented in this file.
The format is based on Keep a Changelog, adheres to Semantic Versioning, and is generated by Changie.
- Cross-database
datemacro
- Update Clone test to reflect core change removing
deferredattribute from nodes
- Fix scenario where using the
--emptyflag causes metadata queries to contain limit clauses
- --limit flag no longer subshells the query. This resolves the dbt Cloud experience issue where limit prevents ordering elements..
- Fix query timer resolution
- Add optional release_connection parameter to connection_named method
- Add get_catalog_for_single_relation macro and capability to enable adapters to optimize catalog generation
- Improvement of the compile error message in the get_fixture-sql.sql when the relation or the model not exist
- Add query recording for adapters which use SQLConnectionManager
- Improve memory efficiency of process_results()
- Enable serialization contexts
- Debug log when
type_codefails to convert to adata_type - Introduce TableLastModifiedMetadataBatch and implement BaseAdapter.calculate_freshness_from_metadata_batch
- Support for sql fixtures in unit testing
- Cross-database
castmacro - Allow adapters to opt out of aliasing the subquery generated by render_limited
- subquery alias generated by render_limited now includes the relation name to mitigate duplicate aliasing
- Fix adapter-specific cast handling for constraint enforcement
- Use
dbt-adaptersthroughout the contributing guide
- Add the option to set the log level of the AdapterRegistered event
- Update dependabot config to cover GHA
- Validate that dbt-core and dbt-adapters remain de-coupled
- remove dbt_version from query comment test fixture
- add support for py3.12
- Add field wrapper to BaseRelation members that were missing it.
- Add "description" and "meta" fields to RelationConfig protocol
- Lazy load agate to improve dbt-core performance
- add BaseAdapater.MAX_SCHEMA_METADATA_RELATIONS
- Pin
black>=24.3inpyproject.toml