RingBuffer

Simple ring/circular buffer for storing numbers. Commonly used for moving-window statistics.

Dependencies

No dependencies.

RingBuffer

Demo code showing off most of RingBuffer's functionality:

from ringbuffer import RingBuffer

buf = RingBuffer(3)

len(buf)  # 0, there are 0 elements currently in the RingBuffer
buf.max_size  # 3, The RingBuffer can hold up to 3 elements.

buf.append(5)
buf.append(10)
buf.append(25)

# Common statistics
buf.mean()  #  13.3333
buf.median()  # 10.0
buf.min()  # 5.0
buf.max()  # 25.0

# Finite Differences
# ``buf.diff()`` returns a generator.
list(buf.diff())  # [5.0, 15.0],  Will be 1-shorter than ``len(buf)``.

# Indexing
buf.append(50)  # Will overwrite the first element, that used to be ``5``
buf[0]  # 10.0, the 0th index will contain the oldest value
buf[-1]  # 50.0, the -1th index will contain the newest value

buf.full  # True, the buffer is currently full

list(
    buf
)  # [10.0, 25.0, 50.0], iterating over RingBuffer will go from oldest to newest.

buf.clear()  # Resets the RingBuffer
buf.full  # False, the RingBuffer has been cleared and is currently empty.

The underlying element size can be configured by specifying the dtype argument:

buf = RingBuffer(3, "b")  # signed character
buf = RingBuffer(3, dtype="d")  # double

All dtype specifiers are the same as for array.array objects.

dtype	C Type	Size
b	signed char	1
B	unsigned char	1
h	signed short	2
H	unsigned short	2
i	signed int	2
I	unsigned int	2
l	signed long	4
L	unsigned long	4
q	signed long long	8
Q	unsigned long long	8
f	float (DEFAULT)	4
d	double	8