Use bytearray() in FrameBuffer

[deedy5]

Changed self.data to use bytearray instead of bytes to improve performance.

import timeit

data = b"x" * 1024

def test_bytes():
    x = b""
    for _ in range(16):
        x += data

def test_bytearray():
    x = bytearray()
    for _ in range(16):
        x += data

bytes_time = timeit.timeit(test_bytes)
bytearray_time = timeit.timeit(test_bytearray)

print(f"{bytes_time=:.3} seconds")
print(f"{bytearray_time=:.3} seconds")

bytes_time=3.0 seconds
bytearray_time=1.09 seconds

[Kriechi]

Thanks for looking into this!

I'm wondering if a memoryview would be even more performant in this situation?
Looking at the code, the data attribute should probably be _data and considered private.

[deedy5]

What is the most efficient way to concatenate many strings together?

To accumulate many bytes objects, the recommended idiom is to extend a bytearray object using in-place concatenation (the += operator):

result = bytearray()
for b in my_bytes_objects:
   result += b

[deedy5]

I'm wondering if a memoryview would be even more performant in this situation?

memoryview don't support extend: https://github.com/python/cpython/blob/main/Objects/memoryobject.c

bytearray supports: https://github.com/python/cpython/blob/3f3e1c4095ae229b0c6d0364f289a20732281a96/Objects/bytearrayobject.c#L2082-L2191

[Kriechi]

Thanks! 🎉