daft.functions.list_distinct#

def list_distinct(list_expr: Expression) -> Expression:
    """Returns a list of unique elements in each list, preserving order of first occurrence and ignoring nulls.

    Args:
        list_expr (List Expression): The input list expression

    Returns:
        Expression (List Expression): an expression with lists containing only unique elements

    Examples:
        >>> import daft
        >>> from daft.functions import list_distinct
        >>> df = daft.from_pydict({"a": [[1, 2, 2, 3], [4, 4, 6, 2], [6, 7, 1], [None, 1, None, 1]]})
        >>> df.select(list_distinct(df["a"])).show()
        ╭─────────────╮
        │ a           │
        │ ---         │
        │ List[Int64] │
        ╞═════════════╡
        │ [1, 2, 3]   │
        ├╌╌╌╌╌╌╌╌╌╌╌╌╌┤
        │ [4, 6, 2]   │
        ├╌╌╌╌╌╌╌╌╌╌╌╌╌┤
        │ [6, 7, 1]   │
        ├╌╌╌╌╌╌╌╌╌╌╌╌╌┤
        │ [1]         │
        ╰─────────────╯
        <BLANKLINE>
        (Showing first 4 of 4 rows)

        Note that null values are ignored:

        >>> df = daft.from_pydict({"a": [[None, None], [1, None, 1], [None]]})
        >>> df.select(list_distinct(df["a"])).show()
        ╭─────────────╮
        │ a           │
        │ ---         │
        │ List[Int64] │
        ╞═════════════╡
        │ []          │
        ├╌╌╌╌╌╌╌╌╌╌╌╌╌┤
        │ [1]         │
        ├╌╌╌╌╌╌╌╌╌╌╌╌╌┤
        │ []          │
        ╰─────────────╯
        <BLANKLINE>
        (Showing first 3 of 3 rows)

    """
    return Expression._call_builtin_scalar_fn("list_distinct", list_expr)