v0.10.5
Converters improvements
General
-
XML Encoding Support: The
_process_one_filefunction now detects and correctly handles XML files with internally-specified encoding (e.g., Windows-1252), ensuring successful parsing and conversion of non-UTF-8 files in transformation pipelines. [#1828] -
Test Enhancements: Updates to test cases (
test_transpiles_informatica_with_sparksql,test_transpiles_all_dbt_project_files) were made to increase reliability and provide better logging. [#1828]
Morpheus transpiler
-
Temporary and Transient Table Support Across Dialects:
-
Enhanced Support for T-SQL
SETStatement Options: -
Fix: CTEs in Subqueries:
-
IR Refinement for
CREATECommands:- Introduces a new
CreateCommandnode to better mirror SQL grammar, consolidating and simplifying previous IR structures (e.g., removingReplaceTableandReplaceTableAsSelect)
- Introduces a new
-
CREATE VIEW Implementation:
BladeBridge Transpiler
-
UPDATE to MERGE Logic:
- Conversion logic for
UPDATE...FROMtoMERGEimplemented - Post-processing Improvements:
convert_update_to_mergefunction now ensures statement termination by checking for trailing semicolons.
- Conversion logic for
-
Oracle Data Type Mapping Fixes:
NUMBERwithout precision now maps toDECIMAL(38,18)instead ofDECIMAL(10,0).- Corrects
Timestampmapping and convertsChar(length)toSTRING. SYSTIMESTAMPis now translated toCURRENT_TIMESTAMP()
-
Datastage SET VARIABLE Handling:
Reconcile Improvements
- Use of Existing Warehouse During Configure-Reconcile:
- The reconcile configuration now checks for an existing
warehouse_idin the user's Databricks config. - If present, it uses the existing SQL warehouse (with
CAN_USEpermission) instead of creating a new one. - Logs warehouse details and defers deletion for reusability. [#1825]
- The reconcile configuration now checks for an existing
Documentation updates
-
Databricks Auth Profiles and
--profileOption:- Users can now specify which Databricks workspace to use with the
--profileflag during installation. - Adds command to list available profiles. [#1813]
- Users can now specify which Databricks workspace to use with the
-
Export Instructions for Microsoft SQL Server and Azure Synapse:
- Step-by-step guides added for extracting view, table, and procedure DDLs using:
- SQL Server Management Studio (SSMS),
- Azure Synapse Studio,
- PowerShell via
Export-AzSynapseSqlScriptfor Synapse Serverless.
- Screenshots and Microsoft documentation links included. [#1812]
- Step-by-step guides added for extracting view, table, and procedure DDLs using:
Dependency Updates:
- Updated `databricks-labs-blueprint` version.
- Added `pytest-timeout` for improved test reliability. [[#1828]](https://github.com/databrickslabs/lakebridge/issues/1828)
Contributors: @eri-adepoju, @sundarshankar89, @asnare, @biswadeepupadhyay-db