SnowConvert: Teradata¶

Specific CLI arguments¶

The following CLI arguments are specific for executing migrations with SnowConvert for Teradata

--displaceDatabaseAsSchema¶

This flag must be used with the -s parameter. When used it will maintain Teradata’s database name qualification as Snowflake’s data warehouse, contrary to it’s default behavior where it becomes a schema on Snowflake code. Let’s look and example where -s customSchema is included:

SELECT * FROM databaseName.tableName;
Copy
-- Additional Params: -s customSchema
SELECT
* FROM
customSchema.tableName;
Copy
-- Additional Params: -s customSchema --displaceDatabaseAsSchema
SELECT
* FROM
databaseName.customSchema.tableName;
Copy

--CharacterToApproximateNumber <NUMBER>¶

An integer value for the CHARACTER to Approximate Number transformation (Default: 10).

--DefaultDateFormat <STRING>¶

String value for the Default DATE format (Default: "YYYY/MM/DD").

--DefaultTimeFormat <STRING>¶

String value for the Default TIME format (Default: "HH:MI:SS").

--DefaultTimestampFormat <STRING>¶

String value for the Default TIMESTAMP format (Default: "YYYY/MM/DD HH:MI:SS").

--DefaultTimezoneFormat <STRING>¶

String value for the Default TIMEZONE format (Default: "GMT-5").

-p, --scriptTargetLanguage <TARGET_LANGUAGE>¶

The string value specifies the target language to convert Bteq and Mload script files. Currently supported values are SnowScript and Python. The default value is set to Python.

-n, --SessionMode <SESSION_MODE>¶

SnowConvert handles Teradata code in both TERA and ANSI modes. Currently, this is limited to the default case specification of character data and how it affects comparisons.

The string value specifies the Session Mode of the input code. Currently supported values are TERA and ANSI. The default value is set to TERA.

You can learn more about how SnowConvert handles and converts code depending on the session mode, check here.

--replaceDeleteAllToTruncate¶

Flag to indicate whether Delete All statements must be replaced with Truncate or not. This will generate SSC-EWI-TD0037 when the replacement is done. Example:

create table testTable(
    column1 varchar(30)
);

delete testTable all;

delete from testTable;
Copy
CREATE OR REPLACE TABLE testTable (
    column1 varchar(30)
)
COMMENT = '{"origin":"sf_sc","name":"snowconvert","version":{"major":1, "minor":0},{"attributes":{"component":"teradata"}}'
;

DELETE FROM testTable;

DELETE FROM 
    testTable;
Copy
-- Additional Params: --replaceDeleteAllToTruncate
CREATE OR REPLACE TABLE testTable (
    column1 varchar(30)
)
COMMENT = '{"origin":"sf_sc","name":"snowconvert","version":{"major":1, "minor":0},{"attributes":{"component":"teradata"}}'
;

TRUNCATE TABLE testTable;

DELETE FROM 
    testTable;
Copy

--generateStoredProcedureTags¶

Flag to indicate whether the SQL statements SELECT, INSERT, CREATE, DELETE, UPDATE, DROP, MERGE in Stored Procedures will be tagged on the converted code. This feature is used for easy statement identification on the migrated code. Wrapping these statements within these XML-like tags allows for other programs to quickly find and extract them. The decorated code looks like this:

//<SQL_DELETE
EXEC(DELETE FROM SB_EDP_SANDBOX_LAB.PUBLIC.USER_LIST,[])
//SQL_DELETE!>
Copy

--splitPeriodDatatype¶

This flag is used to indicate that the tool should migrate any use of the PERIOD datatype as two separate DATETIME fields that will hold the original period begin and end values, anytime a period field or function is migrated using this flag SSC-FDM-TD0004 will be added to warn about this change.

CREATE TABLE myTable(
   col1 PERIOD(DATE),
   col2 VARCHAR(50),
   col3 PERIOD(TIMESTAMP)
);
Copy
CREATE OR REPLACE TABLE myTable (
   col1 VARCHAR(24) !!!RESOLVE EWI!!! /*** SSC-EWI-TD0053 - SNOWFLAKE DOES NOT SUPPORT THE PERIOD DATATYPE, ALL PERIODS ARE HANDLED AS VARCHAR INSTEAD ***/!!!,
   col2 VARCHAR(50),
   col3 VARCHAR(58) !!!RESOLVE EWI!!! /*** SSC-EWI-TD0053 - SNOWFLAKE DOES NOT SUPPORT THE PERIOD DATATYPE, ALL PERIODS ARE HANDLED AS VARCHAR INSTEAD ***/!!!
)
COMMENT = '{"origin":"sf_sc","name":"snowconvert","version":{"major":1, "minor":0},{"attributes":{"component":"teradata"}}'
;
Copy
-- Additional Params: --splitPeriodDatatype
CREATE OR REPLACE TABLE myTable (
   col1_begin DATE,
   col1_end DATE /*** SSC-FDM-TD0004 - PERIOD DATA TYPES ARE HANDLED AS TWO DATA FIELDS ***/,
   col2 VARCHAR(50),
   col3_begin TIMESTAMP,
   col3_end TIMESTAMP /*** SSC-FDM-TD0004 - PERIOD DATA TYPES ARE HANDLED AS TWO DATA FIELDS ***/
)
COMMENT = '{"origin":"sf_sc","name":"snowconvert","version":{"major":1, "minor":0},{"attributes":{"component":"teradata"}}'
;
Copy

--arrange¶

Flag to indicate whether the input code should be processed before parsing and transformation.

--RenamingFile¶

The path to a .json file that specifies new names for certain objects such as Tables, Views, Procedures, Functions, and Macros. This parameter can’t be used with the customSchema argument. Navigate to the Renaming Feature to learn more about this argument.

--UseCollateForCaseSpecification¶

This flag indicates whether to use COLLATE or UPPER to preserve Case Specification functionality, e.g. CASESPECIFIC or NOT CASESPECIFIC. By default, it is turned off, meaning that the UPPER function will be used to emulate case insensitivity (NOT CASESPECIFIC). To learn more about how Case Specification is handled by SnowConvert check here.

Language: English