SnowConvert AI - Teradata - MLOAD¶

Translation references to convert Teradata MLOAD files to Python

Teradata MultiLoad is a command-driven utility for fast, high-volume maintenance on multiple tables and views in Teradata Database.

In order to simulate the MultiLoad functionality for Teradata in Snowflake, MultiLoad files and commands are transformed to Python code, similar to the transformations performed for BTEQ and FastLoad scripts. The generated code uses the Snowflake Python project called snowconvert.helpers which contains the required functions to simulate the MultiLoad statements in Snowflake.

MultiLoad Commands Translation¶

Most of the MultiLoad Commands are considered not relevant in Snowflake, these commands are commented out. Below is the summary list of MultiLoad commands and their transformation status into Snowflake:

CommandsTransformation StatusNote
ACCEPTCommented​
BEGIN MLOADTransformed​​The node is commented out since the transformation occurs in other related statements instead.
BEGIN DELETE MLOADCommented​
DATEFORMCommented​
DELETEPartially transformedCheck known issues.​
DISPLAYCommented​
DML LABELTransformed​
END MLOADTransformed​​Commented out since is not necessary for the transformation of the BEGIN MLOAD.
EOCCommented​
FIELDTransformed​
FILLERTransformedThis command needs to be with a FIELD and LAYOUT command to be converted.
IF, ELSE, and ENDIFCommented​
IMPORTTransformed​
INSERTTransformedThis is taken as a Teradata Statement, so it doesn't appear in this chapter.
LAYOUTTransformedThis command needs to be with a FIELD and FILLER command to be converted.
LOGDATACommented​
LOGMECHCommented​
LOGOFFCommented​
LOGONCommented​
LOGTABLECommented​
PAUSE ACQUISITIONCommented​
RELEASE MLOADCommented​
ROUTE MESSAGESCommented​
RUN FILECommented​
SETCommented​
SYSTEMCommented​
TABLECommented​
UPDATETransformedThis is taken as a Teradata Statement, so it doesn't appear in this chapter.
VERSIONCommented​

However, there are some exceptional commands that must be converted into Python-specific code for them to work as intended in Snowflake. See this section.

If you have any additional questions regarding this documentation, you can email us at snowconvert-support@snowflake.com.

BEGIN MLOAD¶

The transformation for the command .BEGIN MLOAD is a multi-part transformation that requires the .LAYOUT, .FIELD, .FILLER,.DML LABEL, and .IMPORT commands to simulate its behavior correctly.

This transformation is fully explained in the following subsections.

.LAYOUT, .FIELD and .FILLER¶

The transformation for the commands .LAYOUT, .FIELD, and .FILLER will create variable definitions to be used in a future function call of the IMPORT of this layout.

Teradata (MultiLoad)

.LAYOUT INFILE_LAYOUT;
.FIELD TABLE_ID        * INTEGER;
.FIELD TABLE_DESCR     * CHAR(8);
.FILLER COL1           * CHAR(1);
.FIELD TABLE_NBR       * SMALLINT;
.FIELD TABLE_SOMEFIELD * SMALLINT;
Copy

Snowflake (Python)

#*** Generated code is based on the SnowConvert AI Python Helpers version 2.0.6 ***
 
import os
import sys
import snowconvert.helpers
from snowconvert.helpers import Export
from snowconvert.helpers import exec
from snowconvert.helpers import BeginLoading
con = None
def main():
  snowconvert.helpers.configure_log()
  con = snowconvert.helpers.log_on()
  INFILE_LAYOUT_TableName = "INFILE_LAYOUT_TEMP_TABLE"
  INFILE_LAYOUT_Columns = """TABLE_ID INTEGER, 
TABLE_DESCR CHAR(8), 
COL1 CHAR(1), 
TABLE_NBR SMALLINT, 
TABLE_SOMEFIELD SMALLINT"""
  INFILE_LAYOUT_Conditions = """TABLE_ID AS TABLE_ID, TABLE_DESCR AS TABLE_DESCR, COL1 AS COL1, TABLE_NBR AS TABLE_NBR, TABLE_SOMEFIELD AS TABLE_SOMEFIELD"""
  snowconvert.helpers.quit_application()

if __name__ == "__main__":
  main()
Copy

.DML LABEL¶

The transformation for the .DML LABELcommand will create a function containing the statements after the label definition. Note that after the .DML LABEL command there is usually an Insert, Update or Delete.

Teradata (MultiLoad)

-- Example of .DML LABEL with INSERT:
.DML LABEL INSERT_TABLE;                                                                                    
INSERT INTO mydb.mytable( TABLE_ID,TABLE_DESCR,TABLE_NBR ) VALUES( :TABLE_ID,:TABLE_DESCR,:TABLE_NBR );

-- Example of .DML LABEL with DELETE:
.DML LABEL DELETE_TABLE;              
DELETE FROM Employee WHERE EmpNo  = :EmpNo;

-- Example of .DML LABEL with an UPDATE, followed by an INSERT:
.DML LABEL UPSERT_TABLE DO INSERT FOR MISSING UPDATE ROWS;
UPDATE   mydb.mytable SET TABLE_ID = :TABLE_ID WHERE TABLE_DESCR = :somedescription
INSERT INTO mydb.mytable(TABLE_ID, TABLE_DESCR, TABLE_NBR) VALUES(:TABLE_ID, :TABLE_DESCR, :TABLE_NBR );
Copy

Snowflake (Python)

#*** Generated code is based on the SnowConvert AI Python Helpers version 2.0.6 ***
 
import os
import sys
import snowconvert.helpers
from snowconvert.helpers import Export
from snowconvert.helpers import exec
from snowconvert.helpers import BeginLoading
con = None
def main():
  snowconvert.helpers.configure_log()
  con = snowconvert.helpers.log_on()
  def INSERT_TABLE(tempTableName, queryConditions = ""):
    exec(f"""INSERT INTO mydb.mytable (TABLE_ID, TABLE_DESCR, TABLE_NBR)
SELECT
   :TABLE_ID,
   :TABLE_DESCR,
   :TABLE_NBR
FROM {tempTableName} SRC {queryConditions}""")
  exec("""
    DELETE FROM
      Employee
    WHERE
      EmpNo = :EmpNo
    """)
  def UPSERT_TABLE(tempTableName, queryConditions = ""):
    exec(f"""MERGE INTO mydb.mytable TGT USING (SELECT * FROM {tempTableName} {queryConditions}) SRC ON TABLE_DESCR = :somedescription
WHEN MATCHED THEN UPDATE SET
   TABLE_ID = :TABLE_ID
WHEN NOT MATCHED THEN INSERT (TABLE_ID, TABLE_DESCR, TABLE_NBR)
VALUES (:TABLE_ID, :TABLE_DESCR, :TABLE_NBR)""")
  snowconvert.helpers.quit_application()

if __name__ == "__main__":
  main()
Copy

.IMPORT¶

The transformation of the .IMPORT command will create a call to theimport_file_to_temptablehelper to load the data from the file to a temporary table. Then, the calls to all theAPPLYlabels used in the original import will be created. Finally, the calls for anINSERTlabel will be transformed to a query parameter and optionally can have a query condition.

Teradata (MultiLoad)

.IMPORT INFILE INFILE_FILENAME
    LAYOUT INFILE_LAYOUT
    APPLY INSERT_TABLE
    APPLY UPSERT_TABLE
    Apply DELETE_TABLE;
Copy

Snowflake (Python)

#*** Generated code is based on the SnowConvert AI Python Helpers version 2.0.6 ***
 
import os
import sys
import snowconvert.helpers
from snowconvert.helpers import Export
from snowconvert.helpers import exec
from snowconvert.helpers import BeginLoading
con = None
def main():
  snowconvert.helpers.configure_log()
  con = snowconvert.helpers.log_on()
  #** SSC-FDM-0027 - REMOVED NEXT STATEMENT, NOT APPLICABLE IN SNOWFLAKE. TRANSLATED BELOW **
  #.IMPORT INFILE INFILE_FILENAME LAYOUT INFILE_LAYOUT APPLY INSERT_TABLE APPLY UPSERT_TABLE Apply DELETE_TABLE
   
  snowconvert.helpers.import_file_to_temptable(fr"INFILE_FILENAME", INFILE_LAYOUT_TableName, INFILE_LAYOUT_Columns, INFILE_LAYOUT_Conditions, ',')
  INSERT_TABLE(INFILE_LAYOUT_TableName)
  UPSERT_TABLE(INFILE_LAYOUT_TableName)
  DELETE_TABLE(INFILE_LAYOUT_TableName)
  exec(f"""DROP TABLE {INFILE_LAYOUT_TableName}""")
  snowconvert.helpers.quit_application()

if __name__ == "__main__":
  main()
Copy

Large Example¶

Given the transformations shown above for a variety of commands, consider the following example.

With this input data:

id,name,age
1,John,25
2,Maria,29
3,Carlos,31
4,Mike,40
5,Laura,27
Copy

Teradata (MultiLoad)

Query

.begin import mload
        tables
	mySampleTable1
sessions 20
ampcheck none;

.layout myLayOut;
 .field ID * VARCHAR(2) NULLIF ID = '1';
 .field NAME * VARCHAR(25);
 .field AGE * VARCHAR(10);
.dml label insert_data;

INSERT INTO mySampleTable1
 ( 
    ID,
    NAME,
    AGE	
 )
VALUES
 (
    :ID,
    SUBSTRING(:NAME FROM 2),
    :AGE
 );

.import infile sampleData.txt
layout myLayOut
apply insert_data

.end mload;
.logoff;
Copy

Result

ROWIDNAMEAGE
1NULLohn25
22aria29
33arlos31
44ike40
55aura27

Snowflake (Python)

Query

#*** Generated code is based on the SnowConvert AI Python Helpers version 2.0.6 ***
#** SSC-FDM-0007 - MISSING DEPENDENT OBJECT "mySampleTable1" **
 
import os
import sys
import snowconvert.helpers
from snowconvert.helpers import Export
from snowconvert.helpers import exec
from snowconvert.helpers import BeginLoading
con = None
def main():
  snowconvert.helpers.configure_log()
  con = snowconvert.helpers.log_on()
  #** SSC-FDM-0027 - REMOVED NEXT STATEMENT, NOT APPLICABLE IN SNOWFLAKE.  **
  #.begin import mload tables mySampleTable1 sessions 20 ampcheck none
   
  myLayOut_TableName = "myLayOut_TEMP_TABLE"
  myLayOut_Columns = """ID VARCHAR(2), 
NAME VARCHAR(25), 
AGE VARCHAR(10)"""
  myLayOut_Conditions = """CASE
   WHEN UPPER(RTRIM(ID)) = UPPER(RTRIM('1'))
      THEN NULL
   ELSE ID
END AS ID, NAME AS NAME, AGE AS AGE"""
  def insert_data(tempTableName, queryConditions = ""):
    exec(f"""INSERT INTO mySampleTable1 (ID, NAME, AGE)
SELECT
   SRC.ID,
   SUBSTRING(SRC.NAME, 2),
   SRC.AGE
FROM {tempTableName} SRC {queryConditions}""")
  #** SSC-FDM-0027 - REMOVED NEXT STATEMENT, NOT APPLICABLE IN SNOWFLAKE. TRANSLATED BELOW **
  #.import infile sampleData.txt layout myLayOut apply insert_data
   
  snowconvert.helpers.import_file_to_temptable(fr"sampleData.txt", myLayOut_TableName, myLayOut_Columns, myLayOut_Conditions, ',')
  insert_data(myLayOut_TableName)
  exec(f"""DROP TABLE {myLayOut_TableName}""")

  if con is not None:
    con.close()
    con = None
  snowconvert.helpers.quit_application()

if __name__ == "__main__":
  main()
Copy

Result

ROWIDNAMEAGE
1NULLohn25
22aria29
33arlos31
44ike40
55aura27

Known Issues¶

1. Delete statement is partially supported

The DELETE statement is partially supported since the where conditions, when found, are not being converted correctly if pointing to a LAYOUT defined column.

In the example below, :EmpNo is pointing to a LAYOUT defined column. However, the transformation does not take this into account and thus the code will be referencing a column that does not exists.

  exec("""
    DELETE FROM
      Employee
    WHERE
      EmpNo = :EmpNo
    """)
Copy

If you have any additional questions regarding this documentation, you can email us at snowconvert-support@snowflake.com.

Language: English