This release includes changes in behavior for the following connectors.
Google BigQuery V2 Connector
Effective in this release, Google BigQuery V2 Connector includes the following change in behavior:
• When you configure a mapping, you cannot use view or materialized view as a target object.
Previously, you could import view or materialized view as a target object.
•If you have an existing mapping that writes a view as the target object, override the target object with the Target Table Name advanced property, and set the optional property DisableMappingDeployment:true at the Google BigQuery V2 connection, the mapping fails with the following error:
Operation failed: Internal Error Occurred. Contact Support : [Cannot create write operation. The node supports read operation only.].
Previously, the mapping ran successfully.
To revert to the previous behavior, remove the optional property DisableMappingDeployment:true at the Google BigQuery V2 connection and run the mapping again.
•When you configure a mapping in advanced mode that writes the data of more than 10 MB to a row, the mapping fails.
Previously, the mapping ran successfully.
To revert to the previous behavior, contact Informatica Global Customer Support.
•When you configure a mapping in advanced mode and write Date, Datetime, and Timestamp values, 0001-01-01 00:00:00 or 0001-01-02 into a Timestamp column in the target, the mapping fails. Also, data corruption might occur when you write Date, Datetime, and Timestamp values less than 1910-01-01 00:00:00.
Previously, the mapping ran successfully.
To revert to the previous behavior, contact Informatica Global Customer Support.
•When you configure a mapping in advanced mode with Write truncate selected in the Write Disposition target advanced property, and the target table contains Date or Datetime columns, the mapping fails.
Previously, the mapping ran successfully.
To revert to the previous behavior, select Write append in the Write Disposition target advanced property and enable the Truncate Target Table target advanced property.
Microsoft Azure Synapse SQL Connector
Effective in this release, when you use the copy command to load data from the staging location to Microsoft Azure Synapse SQL and the data contains Windows newline characters, the Secure Agent writes the newline characters with both the carriage return (/r) and line feed (/n) to the target.
Previously, the Secure Agent wrote only the line feed (/n) to the target causing a change in the length of characters in the target data.
Snowflake Data Cloud Connector
Effective in this release, Snowflake Data Cloud Connector includes the following changes in behavior:
•When you run a mapping task to insert data from a CDC source to a Snowflake target where a primary key is not defined, the mapping task inserts data directly to the target table instead of creating a temporary stage table.
Previously, if the Snowflake target did not have a defined primary key, the mapping task inserted data from a CDC source to a temporary stage table before inserting to the Snowflake target.
•When you run a mapping task to read from a CDC source that does not contain any changed data and write to a Snowflake target for which the recovery mechanism is disabled, the mapping task does not commit data to the target.
Previously, even if a CDC source does not contain any changed data and the recovery mechanism is disabled for a Snowflake target, the mapping task executed the commit request and showed incorrect result.