USER MANUALS

Data Types for Dates, Timestamps and Intervals

Virtual DataPort has data types to hold datetime and interval values. See the table below:

Name of the Type in Denodo

Description

Name of the Type in the SQL Standard

LOCALDATE

Date without a time zone (year, month and day)

DATE

TIME

Time without a time zone (hour, minute, second and fraction of second with nanosecond precision)

TIME WITHOUT TIME ZONE

TIMESTAMP

Timestamp without a time zone (year, month, day, hour, minute, second and fraction of second with nanosecond precision)

TIMESTAMP WITHOUT TIME ZONE

TIMESTAMPTZ

Timestamp with a time zone displacement

TIMESTAMP WITH TIME ZONE

INTERVALDAYSECOND

Duration of a period of time with a precision that can include any set of contiguous fields other than YEAR or MONTH

day-time intervals

INTERVALYEARMONTH

Duration of a period of time with a precision that includes a YEAR field or a MONTH field, or both

year-month intervals

DATE (deprecated)

Timestamp with a time zone displacement. Although similar to TIMESTAMPTZ, it only has millisecond precision and its behavior is not 100% ANSI SQL-compliant. It should not be used anymore. Maintained for compatibility reasons.

N/A

Virtual DataPort maintains the DATE type from previous versions to keep backward compatibility. However, it will be removed in future major versions so you should plan on stop using it on your new developments. This type is similar to the type TIMESTAMP WITH LOCAL TIMEZONE of Oracle. This data type has been deprecated because it introduces great complexities when dealing with this time, particularly when the client applications or the data sources run on a different time zone than the Denodo server.

Note

For more information about date patterns format, refer to Date and Time Pattern Strings. It is recommended to use date type literals instead of text literals for dates, as using invalid patterns with text literals could lead to unexpected results.

Rules to Build Queries that Deal with Datetime Types

The following subsections list the rules you should follow when executing queries that involve datetime fields:

Most Appropriate Method to Define a Datetime Value

To build a datetime literal you have two options:

  1. Use the syntax of the standard (name of the type followed by the value on a specific format).

  2. Or with a function.

Both methods are equivalent.

Type in Denodo

Option to Build a Value of this Type

Examples

localdate

Literal

DATE '2018-01-15'

Function

to_localdate('yyyy-MM-dd', '2018-01-15')

time

Literal

TIME '21:45:59'

TIME '21:45:59.999'

Function

to_time('HH:mm:ss.SSS','21:45:59.999')

timestamp

Literal

TIMESTAMP '2018-01-15 21:45:01'

TIMESTAMP '2018-01-15 21:45:01.256'

Function

to_timestamp(
    'yyyy-MM-dd HH:mm:ss.SSS'
  , '2018-01-15 21:45:01.256')

date (deprecated)

The same as with timestamptz (see below). From the perspective of clients, both types behave in the same way

timestamptz

Literal

TIMESTAMP WITH TIME ZONE '2018-01-15 21:45:01 +02:00'

TIMESTAMP WITH TIME ZONE '2018-01-15 21:45:01.256
+02:00'

Function

to_timestamptz(
    'yyyy-MM-dd HH:mm:ss.SSS XXX'
  , '2018-01-15 21:45:01.256 +02:00')

interval_year_month

Literal

INTERVAL '145' YEAR

INTERVAL '145-11' YEAR TO MONTH

Function

There are no functions to build intervals

time

Literal

INTERVAL '4 5:12:10.222' DAY TO SECOND

INTERVAL '4 5:12' DAY TO MINUTE

Function

There are no functions to build intervals

Use the Appropriate Function to Obtain the Current Datetime

To compare a datetime value with the current instant of time, use the appropriate function depending on the type you are comparing with.

Function to obtain the current datetime

Data Type

Function to Use Obtain the Current Instant

localdate

CURRENT_DATE() or CURRENT_DATE

time

LOCALTIME

timestamp

LOCALTIMESTAMP or LOCALTIMESTAMP()

timestamptz

CURRENT_TIMESTAMP or NOW()

date (deprecated)

NOW()

LOCALTIMESTAMP was introduced with 8.0u20200201. If you are using Denodo 8.0 GA (no updates), execute CAST(CURRENT_TIMESTAMP AS TIMESTAMP WITHOUT TIME ZONE) instead.

LOCALTIME was introduced with 8.0u20210715. If you are using Denodo 8.0 GA (no updates) or 8.0u20210209, execute CAST(CURRENT_TIMESTAMP AS TIME WITHOUT TIME ZONE) instead.

Converting a Datetime Value to Another Datetime Value

A datetime value of one type can be converted to a datetime value of another type. The table below shows the possible conversions between different datetime data types.

Datetime data type conversions

to LOCALDATE

to TIME

to TIMESTAMP

to TIMESTAMPTZ

from LOCALDATE

trivial

not supported

Copy year, month, day. Set hour, minute, second to (zero)

SV→TS w/o TZ→TS w/ TZ

from TIME

not supported

trivial

Copy date fields from CURRENT_DATE and time fields from SV

SV→TS w/o TZ→TS w/ TZ

from TIMESTAMP

Copy date fields from SV

Copy time fields from SV

trivial

TV.UTC = SV-STZD; TV.TZ=STZD

to TIMESTAMPTZ

SV → TS w/o TZ→LOCALDATE

SV→TS w/o TZ→TIME

SV.UTC + SV.TZ

trivial

  • SV: source value.

  • TV: target value.

  • UTC: UTC component of SV or TV (if and only if the source or target has time zone).

  • TZ: time zone displacement of SV or TV (if and only if the source or target has time zone), STZD is the SQL-session default time zone displacement.

  • “→”: the conversion is conceptually done in several steps. For example, from “timestamptz” to “localdate”, the conversion is “SV → TS w/o TZ→LOCALDATE”. This means that the “timestamptz” value is converted to “timestamp”, and the “timestamp” to “date”.

Rules:

  • All these conversions are done implicitly if necessary, except the conversions from and to timestamptz, which need to be done explicitly (using the function CAST).

  • When converting from timestamp to timestamptz, the time zone added to the timestamptz value is the time zone of the i18n of the Denodo server.

Examples:

  • 2017-10-15 21:45:00 → 2017-10-15 21:45:00 +02:00

    From timestamp to timestamptz. The time zone of the new value is the i18n of the Server.

  • 2017-01-15 21:15:00 → 2017-01-15

    From timestamp to date: remove the time components.

  • 2017-10-15 21:45:01 → 21:45:01

    From timestamp to time: remove the date components.

  • 2017-10-15 21:45:00 +05:00 → 2017-10-15 18:45:00

    From timestamptz to timestamp: convert to local timestamp subtracting the difference between the time zone displacement of the source value and the time zone of the i18n of the Denodo server. In this example, the time zone of the Server is +02:00.

  • 2017-10-15 21:45:00 +05:00 → 2017-01-16

    From timestamptz to localdate: first convert to timestamp (2017-10-16 00:45:00), and then remove the time part. In this example, the time zone of the Server is +08:00. Note that in this example, the day in the source value is different from the day in the target value. That is because in the first stage of the conversion (from timestamptz to timestamp), the time is modified according to the difference in time zones and this can lead to a change in the day as well.

  • 2017-10-15 21:45:00 +05:00 → 18:45:00

    From timestamptz to time: first convert to timestamp (2017-10-15 18:45:00), and then remove the date part. In this example, the time zone of the Server is +02:00.

  • 2017-01-15 → 2017-01-15 00:00:00

    From localdate to timestamp: set the time components to zero.

  • 2017-01-15 → 2017-01-15 00:00:00 +05:00

    From localdate to timestamptz: first convert to timestamp, and then add the time zone of the i18n of the Server. In this example, the time zone of the Server is +05:00.

  • 21:45:01 → 2017-10-15 21:45:01

    From time to timestamp: complete the date part with the current date in the Denodo server.

  • 21:45:00 → 2017-10-15 21:45:00 +02:00

    From time to timestamptz: first convert to timestamp (2017-10-15 21:45:00), and then add the time zone of the i18n of the Server. In this example, the time zone of the Server is +02:00.

Data Types for Intervals

Intervals represent a period of time. There are two types of intervals:

  1. intervalyearmonth: duration of a period of time with a precision that includes a YEAR field or a MONTH field, or both. Examples:

    -- 145 years
    INTERVAL '145' YEAR
    
    -- 145 years and 11 months
    INTERVAL '145-11' YEAR TO MONTH
    
    -- 145 months
    INTERVAL '145' MONTH
    
  2. intervaldaysecond: duration of a period of time with a precision that can include any set of contiguous fields other than YEAR or MONTH.

    -- 4 days, 5 hours, 12 minutes, 10 seconds and 222 milliseconds
    INTERVAL '4 5:12:10.222' DAY TO SECOND
    
    -- 4 days, 5 hours and 12 minutes
    INTERVAL '4 5:12' DAY TO MINUTE
    
    -- 11 hours and 20 minutes
    INTERVAL '11:20' HOUR TO MINUTE
    
    -- 10 minutes and 22 seconds
    INTERVAL '10:22' MINUTE TO SECOND
    
    -- 30 seconds and 123 milliseconds
    INTERVAL '30.123' SECOND
    

Two interval values are only comparable if they are of the same interval type.

Arithmetic Operators for Datetime and Interval Values

You can use arithmetic operators to perform operations over values of type datetime and INTERVAL.

Valid operators involving datetimes and intervals

Operand 1

Operator

Operand 2

Result Type

Datetime

-

Datetime

long *-1

Datetime

+ or -

Interval

Datetime

Time

+ or -

Time

long *-2

Interval

+

Datetime

Datetime

Interval

+ or -

Interval

Interval

Interval

* or /

int or long

Interval

int or long

*

Interval

Interval

*-1: number of days between <operand 1> and <operand 2>. The result is a positive number if <operand 1> is a more recent datetime than <operand 2>. Otherwise, the result is a negative number.

*-2: milliseconds between <operand 1> and <operand 2>. The result is a positive number if <operand 1> is higher than <operand 2>. Otherwise, the result is a negative number.

Add feedback