daft.functions.to_unix_epoch#

def to_unix_epoch(expr: Expression, time_unit: str | TimeUnit | None = None) -> Expression:
    """Converts a datetime column to a Unix timestamp with the specified time unit. (default: seconds).

    See [daft.datatype.TimeUnit](https://docs.daft.ai/en/stable/api/datatypes/all_datatypes/#daft.datatype.DataType.timeunit) for more information on time units and valid values.

    Examples:
        >>> import daft
        >>> from datetime import date
        >>>
        >>> df = daft.from_pydict(
        ...     {
        ...         "dates": [
        ...             date(2001, 1, 1),
        ...             date(2001, 1, 2),
        ...             date(2001, 1, 3),
        ...             None,
        ...         ]
        ...     }
        ... )
        >>> df.with_column("timestamp", df["dates"].to_unix_epoch("ns")).show()
        ╭────────────┬────────────────────╮
        │ dates      ┆ timestamp          │
        │ ---        ┆ ---                │
        │ Date       ┆ Int64              │
        ╞════════════╪════════════════════╡
        │ 2001-01-01 ┆ 978307200000000000 │
        ├╌╌╌╌╌╌╌╌╌╌╌╌┼╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌┤
        │ 2001-01-02 ┆ 978393600000000000 │
        ├╌╌╌╌╌╌╌╌╌╌╌╌┼╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌┤
        │ 2001-01-03 ┆ 978480000000000000 │
        ├╌╌╌╌╌╌╌╌╌╌╌╌┼╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌┤
        │ None       ┆ None               │
        ╰────────────┴────────────────────╯
        <BLANKLINE>
        (Showing first 4 of 4 rows)
    """
    return Expression._call_builtin_scalar_fn("to_unix_epoch", expr, time_unit=time_unit)