Async kernel

Async kernel is a Python Jupyter kernel with concurrent message handling.

Messages are processed fairly whilst preventing asynchronous deadlocks by using a unique message handler per channel, message_type and subshell_id.

Highlights

• Experimental support for Jupyterlite try it online here 👈
• Debugger client
• anyio compatible event loops
  • asyncio (default)
  • trio
• aiologic thread-safe synchronisation primitives
• Easy multi-thread / multi-event loop management
• IPython shell
• Per-subshell user_ns
• GUI event loops
  • inline
  • ipympl
  • tk
  • qt

Documentation

Installation

pip install async-kernel

Asyncio

An asyncio kernel based backend with the name 'async' is installed when the kernel is installed.

Trio backend

To add a kernel spec for a trio backend.

pip install trio
async-kernel -a async-trio --interface.backend=trio

For further detail about kernel spec customisation see command line usage.

Message handling

• When a message is received the msg_handler is called with:

  • 'job' (a dict of msg, received_time and ident)
  • The channel
  • msg_type
  • A function send_reply

• The msg_handler

  • determines the subshell_id and run mode.
  • obtains the handler from the kernel with the same name as the msg_type.
  • determines the run mode
  • creates cached version of the run_handler with a unique version per:
    • The handler
    • channel
    • subshell_id
    • send_reply (constant or per-channel)
  • Obtains the caller associated with the channel and schedules execution of the cached handler

Run mode

The run modes available are:

• RunMode.direct → Caller.call_direct: Run the request directly in the scheduler.
• RunMode.queue → Caller.queue_call: Run the request in a queue dedicated to the subshell, handler & channel.
• RunMode.task → Caller.call_soon: Run the request in a separate task.
• RunMode.thread → Caller.to_thread: Run the request in a separate worker thread.

These are the currently assigned run modes.

SocketID	shell	control
comm_close	direct	direct
comm_info_request	direct	direct
comm_msg	queue	queue
comm_open	direct	direct
complete_request	thread	thread
create_subshell_request	None	thread
debug_request	None	queue
delete_subshell_request	None	thread
execute_request	queue	queue
history_request	thread	thread
inspect_request	thread	thread
interrupt_request	direct	direct
is_complete_request	thread	thread
kernel_info_request	direct	direct
list_subshell_request	None	direct
shutdown_request	None	direct

Origin

Async kernel started as a fork of IPyKernel. Thank you to the original contributors of IPyKernel that made Async kernel possible.