Datetime data types

Data Types and Functions in Snowflake

Jake Roach

Field Data Engineer

DATE data types

Date values store the day on which something occurred

                            2023-04-01
                            YYYY-MM-DD

  CREATE TABLE sales (
      transaction_date DATE  -- Define a column of type `DATE`
  );

DATE data types

SELECT

    TO_DATE('2023-04-01') AS the_date, -- Convert a string into a DATE

    '2023-04-01'::DATE AS casted_date  -- Casted using <value>::DATE

;

                         the_date  | casted_date  
                       ----------- | ------------  
                        2023-04-01 | 2023-04-01

Time data types

Captures the hours, minutes, and seconds that some event occurred

                               08:24:08
                               HH:MM:SS

CREATE TABLE sales (
    transaction_time TIME  -- Create a column with the `TIME` keyword
);

Define a TIME column

Time data types

SELECT

    TO_TIME('08:24:04') AS the_time,

    '08:24:04'::TIME AS casted_time

;

                     |  the_time  | casted_time |
                     | ---------- | ----------- |
                     |  08:24:08  |  08:24:08   |

TO_TIME function
Cast using ::

Timestamp data types

TIMESTAMP captures both date and time

                        2023-04-01 08:24:04
                        YYYY-MM-DD HH:MM:SS

CREATE TABLE (
    transaction_timestamp TIMESTAMP
);

Timestamp data types

SELECT 
    TO_TIMESTAMP('2023-04-01 08:24:04') AS the_timestamp,
    '2023-04-01 08:24:04'::TIMESTAMP AS casted_timestamp          
;

Convert a string to a timestamp using TO_TIMESTAMP or <value>::TIMESTAMP
Most common way to store datetime data

                  the_timestamp     |    casted_timestamp     
              --------------------- | ----------------------  
               2024-04-01 08:24:08  |  2024-04-01 08:24:08

Examples

SELECT
    TO_DATE('2021-05-14') AS the_date,
    TO_TIME('06:13:00') AS the_time,
    TO_TIMESTAMP('2021-05-14T06:13:00') AS the_timestamp,


    -- Extract the DATE from a TIMESTAMP, we could do the same with TIME!
    transaction_timestamp::DATE AS casted_date

;

           the_date   |   the_time  |     the_timestamp    | casted_date  
         ------------ | ----------- | -------------------- | ------------  
          2021-05-14  |   06:13:00  | 2021-05-14T06:13:00  |  2021-05-14

Timezones

By default, DATE, TIME, and TIMESTAMP's are stored without a timezone

TIMESTAMP_NTZ

Default implementation of TIMESTAMP
No timezone is stored

TIMESTAMP_LTZ

Relies on local timezone

TIMESTAMP_TZ

User specifies timezone

¹ https://docs.snowflake.com/en/sql-reference/data-types-datetime#timestamp-ltz-timestamp-ntz-timestamp-tz

Let's practice!

Data Types and Functions in Snowflake