PyCircularBuffer

• License : MIT

Summary

Simple circular buffer written in python extension.

CircularBuffer will allocate requested size + 2 bytes for two sections of null terminated strings.

May temporary allocate another half of the allocated bytes if you used buffer protocol, I mostly used them for regex.

Installation

Simply pip install pycircularbuffer, compilation from downloaded source is probably needed.

Using

from circularbuffer import CircularBuffer

buf = CircularBuffer(1024)

buf.write(b'some text')
while len(buf) > 0:
    buf.read(1)

from re import match

buf.write(b'hallo')

# python2
with buf:
    match_found = match(br'^ha', buf)

# python3
match_found = match(br'^ha', buf)

# use `result` immediately because regex didn't make memory copy of the
# internal buffer, or run another `match()` on a memory copy, for example:

match_found_str = buf.read(len(match_found.group(0)))
independent_match_found = match(br'^ha', match_found_str)

Warning

Don't to this in python 3: 'a' in buf, instead: b'a' in buf.

Check the return value of write() which is the actual length that was written into the buffer and also the exception that might be raised.

API

Regular methods:

• clear()
• read()
• resize()
• write()
• write_available()
• make_contiguous()

String methods:

• count()
• startswith()
• find()
• index()

Sequence methods:

• __contains__()
• __getitem__()
• __len__()
• __setitem__()

Magic methods:

• __repr__()

• __str__()

  Note: while string representation makes thing easier it always creates memory copy.

• __enter__()

• __exit__()

Buffer protocol:

Note: for python version < 3, you need to use context manager, the with statement, to let CircularBuffer know when you are releasing the buffer.

Using buffer protocol will throw ReservedError exception for CircularBuffer.read().