aiofiles is an Apache2 licensed library, written in Python, for handling local disk files in asyncio applications.

Ordinary local file IO is blocking, and cannot easily and portably made asynchronous. This means doing file IO may interfere with asyncio applications, which shouldn't block the executing thread. aiofiles helps with this by introducing asynchronous versions of files that support delegating operations to a separate thread pool.

async with aiofiles.open('filename', mode='r') as f:
    contents = await f.read()
print(contents)
'My file contents'

Or, using the old syntax:

f = yield from aiofiles.open('filename', mode='r')
try:
    contents = yield from f.read()
finally:
    yield from f.close()
print(contents)
'My file contents'

• a file API very similar to Python's standard, blocking API
• support for buffered and unbuffered binary files, and buffered text files
• support for async/await (PEP 492) constructs

To install aiofiles, simply:

$ pip install aiofiles

Files are opened using the aiofiles.open() coroutine, which in addition to mirroring the builtin open accepts optional loop and executor arguments. If loop is absent, the default loop will be used, as per the set asyncio policy. If executor is not specified, the default event loop executor will be used.

In case of success, an asynchronous file object is returned with an API identical to an ordinary file, except the following methods are coroutines and delegate to an executor:

• close
• flush
• isatty
• read
• readall
• read1
• readinto
• readline
• readlines
• seek
• seekable
• tell
• truncate
• writable
• write
• writelines

In case of failure, one of the usual exceptions will be raised.

The aiofiles.os module contains executor-enabled coroutine versions of several useful os functions that deal with files:

• stat
• sendfile

When using Python 3.5 or greater, aiofiles file objects can be used as asynchronous context managers. Asynchronous iteration is also supported.

async with aiofiles.open('filename') as f:
    async for line in f:
        ...

When using Python 3.3 or 3.4, be aware that the closing of a file may block, and yielding from a coroutine while exiting from a context manager isn't possible, so aiofiles file objects can't be used as (ordinary, non-async) context managers. Use the try/finally construct from the introductory section to ensure files are closed.

When using Python 3.3 or 3.4, iteration is also unsupported. To iterate over a file, call readline repeatedly until an empty result is returned. Keep in mind readline doesn't strip newline characters.

f = yield from aiofiles.open('filename')
try:
    while True:
        line = yield from f.readline()
        if not line:
            break
        line = line.strip()
        ...
finally:
    yield from f.close()

Real file IO can be mocked by patching aiofiles.threadpool.sync_open as desired. The return type also needs to be registered with the aiofiles.threadpool.wrap dispatcher:

threadpool.wrap.register(mock.MagicMock)(
    lambda *args, **kwargs: threadpool.AsyncBufferedIOBase(*args, **kwargs))

async def test_stuff():
    data = 'data'
    mock_file = mock.MagicMock()

    with mock.patch('aiofiles.threadpool.sync_open', return_value=mock_file) as mock_open:
        async with aiofiles.open('filename', 'w') as f:
            await f.write(data)

        mock_file.write.assert_called_once_with(data)

• The LICENSE is now included in the sdist. #31

• Introduced a changelog.
• aiofiles.os.sendfile will now work if the standard os module contains a sendfile function.

Contributions are very welcome. Tests can be run with tox, please ensure the coverage at least stays the same before you submit a pull request.