def convert_file_size_to_int()

in src/datasets/utils/py_utils.py [0:0]

• 29 lines of code
• 17 McCabe index (conditional complexity)

def convert_file_size_to_int(size: Union[int, str]) -> int:
    """
    Converts a size expressed as a string with digits an unit (like `"50MB"`) to an integer (in bytes).

    Args:
        size (`int` or `str`): The size to convert. Will be directly returned if an `int`.

    Example:

    ```py
    >>> convert_file_size_to_int("1MiB")
    1048576
    ```
    """
    if isinstance(size, int):
        return size
    if size.upper().endswith("PIB"):
        return int(size[:-3]) * (2**50)
    if size.upper().endswith("TIB"):
        return int(size[:-3]) * (2**40)
    if size.upper().endswith("GIB"):
        return int(size[:-3]) * (2**30)
    if size.upper().endswith("MIB"):
        return int(size[:-3]) * (2**20)
    if size.upper().endswith("KIB"):
        return int(size[:-3]) * (2**10)
    if size.upper().endswith("PB"):
        int_size = int(size[:-2]) * (10**15)
        return int_size // 8 if size.endswith("b") else int_size
    if size.upper().endswith("TB"):
        int_size = int(size[:-2]) * (10**12)
        return int_size // 8 if size.endswith("b") else int_size
    if size.upper().endswith("GB"):
        int_size = int(size[:-2]) * (10**9)
        return int_size // 8 if size.endswith("b") else int_size
    if size.upper().endswith("MB"):
        int_size = int(size[:-2]) * (10**6)
        return int_size // 8 if size.endswith("b") else int_size
    if size.upper().endswith("KB"):
        int_size = int(size[:-2]) * (10**3)
        return int_size // 8 if size.endswith("b") else int_size
    raise ValueError(f"`size={size}` is not in a valid format. Use an integer followed by the unit, e.g., '5GB'.")