Function parse_bytes

dask/utils.py:1582–1636

Parse byte string to numbers >>> from dask.utils import parse_bytes >>> parse_bytes('100') 100 >>> parse_bytes('100 MB') 100000000 >>> parse_bytes('100M') 100000000 >>> parse_bytes('5kB') 5000 >>> parse_bytes('5.4 kB') 5400 >>> parse_bytes('1kiB')

(s: float | str)

def parse_bytes(s: float | str) -> int:
    """Parse byte string to numbers

    >>> from dask.utils import parse_bytes
    >>> parse_bytes('100')
    100
    >>> parse_bytes('100 MB')
    100000000
    >>> parse_bytes('100M')
    100000000
    >>> parse_bytes('5kB')
    5000
    >>> parse_bytes('5.4 kB')
    5400
    >>> parse_bytes('1kiB')
    1024
    >>> parse_bytes('1e6')
    1000000
    >>> parse_bytes('1e6 kB')
    1000000000
    >>> parse_bytes('MB')
    1000000
    >>> parse_bytes(123)
    123
    >>> parse_bytes('5 foos')
    Traceback (most recent call last):
        ...
    ValueError: Could not interpret 'foos' as a byte unit
    """
    if isinstance(s, (int, float)):
        return int(s)
    s = s.replace(" ", "")
    if not any(char.isdigit() for char in s):
        s = f"1{s}"

    for i in range(len(s) - 1, -1, -1):
        if not s[i].isalpha():
            break
    index = i + 1

    prefix = s[:index]
    suffix = s[index:]

    try:
        n = float(prefix)
    except ValueError as e:
        raise ValueError(f"Could not interpret '{prefix}' as a number") from e

    try:
        multiplier = byte_sizes[suffix.lower()]
    except KeyError as e:
        raise ValueError(f"Could not interpret '{suffix}' as a byte unit") from e

    result = n * multiplier
    return int(result)


byte_sizes = {