overlaps.Rd
%overlaps% determines if two date ranges overlap at all and returns a logical value. Interval returns the number of days of the overlapping part of the two date periods. Inspired by the eponymous SQL-functions.
x %overlaps% y
Overlap(x, y)
Interval(x, y)
%overlaps%
returns TRUE
or FALSE
depending on if the two ranges overlap. The function Overlap
returns the range of the overlapping region as numeric value. This will be 0, if the ranges do not overlap. Interval
returns the width of the empty space between 2 ranges. Again this will be 0 if the ranges overlap.
To handle overlapping ranges there are 4 cases to consider:
range a: |--------------|
range b: |-----|
range c: |--------|
range d: |-----|
1 2 3 4 5 6 7 8
Ranges a and b overlap, the function Overlap
will return the absolute value of the overlapping region (which will be 3 - 2 = 1 in this case). The result will be the same for Overlap(a, b)
and Overlap(b, a)
.Interval
will have a direction. Ranges b and c do not overlap, Overlap
will return 0, %overlaps%
FALSE. Interval
will return 2 for the case Interval(a, b)
and -2 for Interval(b, a)
.
This functions can be of value, if one has to decide, whether confidence intervals overlap or not.
returns a logical vector (match or not for each element of x).
Interval and Overlap return a numeric vector.
as.Date(c("2012-01-03", "2012-02-03")) %overlaps%
as.Date(c("2012-03-01", "2012-03-03"))
#> [1] FALSE
as.Date(c("2012-01-03", "2012-02-03")) %overlaps%
as.Date(c("2012-01-15", "2012-01-21"))
#> [1] TRUE
Interval(as.Date(c("2012-01-03", "2012-02-03")), as.Date(c("2012-03-01", "2012-03-03")))
#> [1] 27
# both ranges are recyled if necessary
as.Date("2012-01-03") %overlaps% as.Date(c("2012-03-01", "2012-03-03"))
#> [1] FALSE
# works with numerics as well
c(1, 18) %overlaps% c(10, 45)
#> [1] TRUE