Data Type Functions

Jun 20, 2018

Starting in Drill 1.14, Drill supports the following functions that return data type information:

Returns the data type of a column (using the SQL names) whether the column is NULL or not. You can use the SQL name in a CAST statement, for example:

          sqlTypeOf( CAST(x AS <data type> ))  
          //Returns <data type> as the type name.  

If the type is DECIMAL, the type also includes precision and scale, for example:

          DECIMAL(6, 3)  

Returns the cardinality (mode) of the column as "NOT NULL", "NULLABLE", or "ARRAY". Drill data types include a cardinality, for example Optional Int or Required VarChar.

Similar to typeOf(), but returns the internal Drill names even if the value is NULL.

Usage Notes

The data type functions are useful for data conversions. For example, if you know a column value is Nullable Int, you can assume that the data type is one that Drill derived. You can then merge the Drill-derived data type with the data type from another file that has actual values.

Usage Examples

The follow examples show you how you can use the data type functions:

Example 1
This example shows the Drill internal type, nullable int, for a missing column:

   SELECT sqlTypeOf(a) AS a_type, modeOf(a) AS a_mode FROM `json/all-null.json`;

   |  a_type  |  a_mode   |

Example 2
This example shows arrays (repeated) types:

   SELECT sqlTypeOf(columns) as col_type, modeOf(columns) as col_mode
   FROM `csv/cust.csv`;

   |      col_type      | col_mode  |
   Example 3: This example shows non-null types:
   SELECT sqlTypeOf(`name`) AS name_type, 
   modeOf(`name`) AS name_mode FROM `csvh/cust.csvh`;

   |     name_type      | name_mode  |