Collect data into overlapping fixed-length chunks or blocks

sliding_window.py

from typing import Iterator, Iterable
from collections import deque
from itertools import islice

def sliding_window(
    iterable: Iterable[str], 
    size: int, 
    overlap: int = 1,
) -> Iterator[str]:
    """Collect data into overlapping fixed-length chunks or blocks.
    
    Args:
        iterable: Input string or iterable of strings
        size: Window size
        overlap: Number of overlapping characters between windows
        
    Yields:
        Strings of length n (or less for final window if input exhausted)
        
    Examples:
        >>> list(sliding_window('ABCDEFG', 4, 0))
        ['ABCD', 'EFG']
        >>> list(sliding_window('ABCDEFG', 4, 1))
        ['ABCD', 'DEFG']
        >>> list(sliding_window('ABCDEFG', 4, 2))
        ['ABCD', 'CDEF', 'EFG']
        >>> list(sliding_window('ABCDEFG', 4, 3))
        ['ABCD', 'BCDE', 'CDEF', 'DEFG']
    """        
    if overlap >= size:
        # I know I told you not to do this, but this is an exception (pun intended). Plus  
        # I know what I'm doing ;)
        raise ValueError(f"overlap ({overlap}) must be less than window size ({size})")
    elif overlap < 0:
        raise ValueError(f"overlap ({overlap}) must be non-negative")
    
    step = size - overlap
    iterator = iter(iterable)

    # Sam FYI: I mispoke. islice DOES NOT raise a StopIteration exception, only next does
    window = deque(islice(iterator, overlap), maxlen=size)

    while True:
        new_elements = ''.join(islice(iterator, step))
        
        if not new_elements:
            break

        # The last window may not have enough new elements to fill the step size. In 
        # that case, remove stale elements of the prev window by resizing the deque
        maxlen = overlap + len(new_elements)
        window = deque(window, maxlen=maxlen) if maxlen < size else window
        window.extend(new_elements)
        yield ''.join(window)

@scelarek