daft.functions.damerau_levenshtein_distance#

def damerau_levenshtein_distance(left: Expression, right: Expression) -> Expression:
    """Compute the Damerau-Levenshtein distance between two strings.

    This extends the Levenshtein distance by also counting transpositions of two
    adjacent characters as a single edit operation (in addition to insertions,
    deletions, and substitutions).

    Note:
        This computes the Optimal String Alignment (OSA) variant, which does not
        allow a substring to be edited more than once. Results may differ from the
        true Damerau-Levenshtein distance for inputs with overlapping transpositions
        (e.g., ``"CA"`` to ``"ABC"`` is 3 under OSA but 2 under true
        Damerau-Levenshtein). OSA does not satisfy the triangle inequality.

    Args:
        left: The left string expression to compare.
        right: The right string expression to compare against.

    Returns:
        The Damerau-Levenshtein (OSA) distance for each pair of strings. Returns null
        when either input is null.

    Examples:
        >>> import daft
        >>> from daft.functions import damerau_levenshtein_distance
        >>> df = daft.from_pydict({"x": ["abc", "abc", ""], "y": ["bac", "acb", "abc"]})
        >>> df = df.with_column("distance", damerau_levenshtein_distance(df["x"], df["y"]))
        >>> df.collect()
        ╭────────┬────────┬──────────╮
        │ x      ┆ y      ┆ distance │
        │ ---    ┆ ---    ┆ ---      │
        │ String ┆ String ┆ Int64    │
        ╞════════╪════════╪══════════╡
        │ abc    ┆ bac    ┆ 1        │
        ├╌╌╌╌╌╌╌╌┼╌╌╌╌╌╌╌╌┼╌╌╌╌╌╌╌╌╌╌┤
        │ abc    ┆ acb    ┆ 1        │
        ├╌╌╌╌╌╌╌╌┼╌╌╌╌╌╌╌╌┼╌╌╌╌╌╌╌╌╌╌┤
        │        ┆ abc    ┆ 3        │
        ╰────────┴────────┴──────────╯
        <BLANKLINE>
        (Showing first 3 of 3 rows)
    """
    return Expression._call_builtin_scalar_fn("damerau_levenshtein_distance", left, right)