Datometry Documentation

The numeric_to_int_emulation hint maps the Teradata NUMERIC data type to the BigQuery INT64 data type.

The BigQuery NUMERIC data type uses a fixed decimal precision and scale, in which the precision is 38 and the scale is 9. Precision is the number of digits that the number contains. Scale is how many of these digits appear after the decimal point. Because the BigQuery NUMERIC data type is more constrained than Teradata's NUMERIC data type, it cannot represent all of the values available to Teradata. For this reason, you can specify if Hyper-Q should map the Teradata NUMERIC data type to the BigQuery INT64 data type.

To learn more about BigQuery numeric data types, see Data types in the Google bigQuery documentation.

Hyper-Q applies the following mapping to emulating Teradata NUMERIC and DECIMAL data types for use with BigQuery.

Using the above mapping rules, a Teradata NUMERIC data type with a precision of 30 and a scale of 6 maps to a NUMERIC data type on BigQuery.

In the following example, the Teradata SQL statement includes a column (column1) that, on Teradata, uses the NUMERIC datatype. Hyper-Q maps this to a column with a data type of INT64.

CREATE TABLE table_name (column1 dec(18,0), column2 dec(19,0), column3 dec(18,1));

The resulting CREATE TABLE statement on BigQuery after Hyper-Q emulation. Note that BigQuery creates column1 with a data type of INT64.

CREATETABLE
    database.table_name (
        'column1' INT64 ,
        'column2' NUMERIC ,
        'column3' NUMERIC 
   ) );

The syntax for numeric_to_int_emulation is:

numeric_to_int_emulation = true | false

where: