I have 2 questions about Lakebridge solution,

• Synapse with dedicated pool Conversion
  • We were conducting a PoC for Synapse to DBX migration using Lakebridge. What we have observed is that the conversions are not correct. I was anticipating all tables will be converted fine at least but that is not the case. See below source and converted outputs. For complex stored procedures to notebooks I expected manual conversion but you guys can correct me.

Synapse:
CREATE TABLE [extra1].[dimension_City2] (
    [City Key]                   INT            NOT NULL,
    [WWI City ID]                INT            NULL,
    [City]                       NVARCHAR (255) NULL,
    [Latest Recorded Population] BIGINT         NULL,
    [Valid From]                 DATETIME2 (7)  NULL,
    [Valid To]                   DATETIME2 (7)  NULL
)
WITH (CLUSTERED COLUMNSTORE INDEX, DISTRIBUTION = HASH([City Key]));


GO

CREATE OR REPLACE TABLE `extra1`.`dimension_City2` (
    `City Key`                   INT            NOT NULL,
    `WWI City ID`                INT,
    `City` STRING,
    `Latest Recorded Population` BIGINT,
    `Valid From`                 TIMESTAMP  ,
    `Valid To`                   TIMESTAMP  
)
WITH(CLUSTERED COLUMNSTORE INDEX, DISTRIBUTION = HASH(`City Key`));

​

• Note: for source_dialect we have tried both mssql and synapse. Sharing one of the config generated post following the steps.

catalog_name: remorph
error_file_path: ./another_try/errors.log
input_source: ./mssql/DatabaseProjectsqlpool-dwh
output_folder:./another_try/output
schema_name: transpiler
skip_validation: true
source_dialect: mssql
transpiler_config_path: ./.databricks/labs/remorph-transpilers/bladebridge/lib/config.yml
transpiler_options:
  overrides-file: Bladebridge
version: 3

• Custom Transpiler: I was thinking of building a custom transpiler setup. Can you share where I can find documentation on how to start building with it and what I need to take into account.

Documentation referred from official link