Fractional part is removed after load data from Teradata to Spark

Question

We're trying to load data from Teradata, the code using is:

sparkSession.read
      .format("jdbc")
      .options(
        Map(
          "url" -> "jdbc:teradata://hostname, user=$username, password=$password",
          "MAYBENULL" -> "ON",
          "SIP_SUPPORT" -> "ON",
          "driver" -> "com.teradata.jdbc.TeraDriver",
          "dbtable" -> $table_name
        )
      )
      .load()

However, some data lost its fractional part after loading. To be more concise, the column in Teradata is in the type of [Number][1] and after loading, the data type in Spark is DecimalType(38,0), the scale value is 0 which means no digits after decimal point.

Data in Teradata is something like,

id column1 column2
1   50.23    100.23
2   25.8     20.669
3   30.2     19.23

The dataframe of Spark is like,

id column1 column2
1   50     100
2   26     21
3   30     19

The meta data of the table in Teradata is like:

CREATE SET TABLE table_name (id BIGINT, column1 NUMBER, column2 NUMBER) PRIMARY INDEX (id);

The Spark version is 2.3.0 and Teradata is 16.20.32.59.

So here comes question, why the automatic conversion happens and how can I keep the data's fractional part in Spark just as it was in Teradata. [1]: https://docs.teradata.com/r/Teradata-Database-SQL-Data-Types-and-Literals/June-2017/Numeric-Data-Types/FLOAT/REAL/DOUBLE-PRECISION-Data-Types