5.9.8 The OVERLAPS predicate
The OVERLAPS predicate tests whether two “events” cover a common point in
time or not, and has the form:
……Þ ( expression , expression ) OVERLAPS ( expression , expression ) Þ…
Each of the two “events” specified on either side of the OVERLAPS keyword
is a period of time between two specified points on the timeline. The two points
can be specified as a pair of datetime values or as one datetime value and an
INTERVAL offset.
Each “event” is defined by a two expressions constituting a row value
expression having two columns.
The first column in each row value expression must be a DATE, TIME or
TIMESTAMP and the value in the first column of the first “event” must be
comparable (see Section 4.6.3) to the value in the first column of the second
“event”.
The second column in each row value expression may be either a DATE, TIME
or TIMESTAMP that is comparable with the value in the first column or an
INTERVAL with a precision that allows it to be added to the value in the first
column.
The value in the first column of each row value expression defines one of the
points on the timeline for the event.
If the value in the second column of the row value expression is a datetime, it
defines the other point on the timeline for the event.
If the value in the second column of the row value expression is an
INTERVAL, the other point on the timeline for the event is defined by adding
the values in the two column of the row value to expression together.
The NULL value is assumed to be a point that is infinitely late in time.
Either of the two points may be the earlier point in time.
If the value in the first column of the row value expression is the NULL value,
then this is assumed to be the later point in time.