ENH: `__contains__` between `pd.Series` and `pd.Interval`

[matteosantama]

Feature Type

• Adding new functionality to pandas

• Changing existing functionality in pandas

• Removing existing functionality in pandas

Problem Description

I would like to check, element-wise, if a pd.Series is contained within a pd.Interval, ie.

pd.Series(range(10)) in pd.Interval(0, 1)

Feature Description

pd.Interval.__contains__ would have to check for a pd.Series type and then call

def __contains__(self, item: Any) -> bool | pd.Series:
    ...
    if isinstance(item, pd.Series):
        return item.apply(lambda element: element in self)

Alternative Solutions

It's easy enough to write pd.Series(range(10)).apply(lambda element: element in pd.Interval(0, 1), but the syntactic sugar proposed here would be nice.

Additional Context

No response

[topper-123]

I'm actually surprised this doesn't work already, making it work should be a small change only, so I'm +1.

Using apply here will be very inefficient and slow and is a no-go. The efficient way to do this will be to use the existing code in Interval.__contains__ or something similar (i.e. use vector operations).

[rhshadrach]

This seems to confuse in with subset. For example, {1, 2} is not in {1, 2, 3}.

[matteosantama]

That example does not have an interval. If you mean to ask

pd.Series([1, 2, 3]) in pd.Interval(1, 2)

the result should be pd.Series([False, True, False]) because 2 is contained in the half-open interval (1, 2] but 1 and 3 are not.