daft.functions.decode#

def decode(bytes: Expression, charset: ENCODING_CHARSET) -> Expression:
    """Decodes binary values using the specified character set.

    Note that if the charset is "utf-8" or "utf8", then this is equivalent
    to cast(bytes, daft.DataType.string())

    If an invalid encoding is encountered, an error will be raised.
    To handle invalid encodings, use `try_decode` instead.

    Args:
        bytes (Binary Expression): The expression to decode.
        charset (str): The decoding character set (utf-8, base64).

    Returns:
        Expression (Binary Expression): A binary expression with the decoded values.

    Examples:
        >>> import daft
        >>> from daft.functions import decode
        >>> df = daft.from_pydict({"bytes": [b"aGVsbG8sIHdvcmxkIQ=="]})
        >>> df.select(decode(df["bytes"], "base64")).show()
        ╭──────────────────╮
        │ bytes            │
        │ ---              │
        │ Binary           │
        ╞══════════════════╡
        │ b"hello, world!" │
        ╰──────────────────╯
        <BLANKLINE>
        (Showing first 1 of 1 rows)
    """
    if charset.lower() in ("utf-8", "utf8"):
        return bytes.cast(DataType.string())

    return Expression._call_builtin_scalar_fn("decode", bytes, codec=charset)