Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Deprecate nest asyncio #311

Closed
wants to merge 69 commits into from
Closed

Conversation

unkcpz
Copy link
Member

@unkcpz unkcpz commented Jan 31, 2025

No description provided.

unkcpz and others added 30 commits January 17, 2025 23:35
…and asyncio.Future

hand write wrap to kiwipy future (concurrent.futures.Future)

kiwipy.Future -> concurrent.futures.Future
Move communication to rmq
The refactoring is targeting to decouple the dependencies of using kiwipy+rmq as the communicator for the process control. 
By forming a `Coordinator` protocol contract, the different type of rmq/kiwipy related codes are removed out from plumpy logic. The new contract also pave the way to make it clearly show how a new type coordinator can be implemented (future examples will be the `tatzelwurm` a task broker that has scheduler support and file based task broker require no background service). 
For the prototype of how a coordinator should look like, the `MockCoordinator` in `tests/utils` is the coordinator that store things in memory, and can serve as the lightweight ephemeral daemon without persistent functionality.

Another major change here is hand write the resolver of future by mimic how tho asyncio does for wrapping `concurrent.futures.Future` into `asyncio.Future`. I use the same way to convert `asyncio.Future` into `concurent.futures.Future` (which is the `kiwipy.Future` as alias). 

- move the `aio_pika` import lazily by moving the rmq exceptions to `rmq` module, this can increase the performance of `import aiida; aiida.orm`.
- ~~`CancellableAction` using composite to behave as a Future like object.~~
- use `asyncio.Future` in favor of alias `plumpy.Future` and 
- use `concurrent.futures.Future` instead of alias `kiwipy.Future`.
- Hand write `_chain` and `_copy_future` since we can not just rely on the API of asyncio that is not exposed.
- Forming the `coordinator/Communicator` protocol.
- Just forming the `coordinator/Coordinator` protocol and wrap rmq/communicator as a coordinator that not require changs in kiwipy.
- Mock the coordinator for unit test.
- test against aiida-core see what need to be changed there and improve here. (aiidateam/aiida-core#6675)
- The API for plumpy process can be more compact instead of using kiwipy/rmq "subscriber" concept. (how to replace rpc pattern??)
create_state refact

Hashable initialized + parameters passed to Hashable

Fix pre-commit errors
For the auto_persist attributes, the fn auto_save will take care of save
the state
unkcpz and others added 28 commits January 22, 2025 11:05
Refactoring with de-abstract the save and load interfaces for all savable classes
Savable protocol test for all savable class
A round trip save and load to ensure auto_persist proper store all information.
- state_machine.py
- message.py
- persistence.py
- port.py
- workchain.py
- utils.py
- process_spec.py
- process_States.py
- processes.py
* Use yaml FullLoader that handle UUID type as well

* Use UnsafeLoader explicitly for bundles

* Using msgpack for passing control messages
Using the same interface for creating all message type, create/launch/continue, play/pause/kill/status
It revert the message API did by using MessageBuilder, it cannot handle `MessageBuilder.continue`, it is not a good API design.
The event loop set up of coordinator and controller is from using the same event loop of a single async pytest. 
This maximum the decouple of tests in `tests/rmq/test_communication.py`.
The fixture and the event loop scope of async test are both set to "function".
This can remove the get_event_loop deprecation warning.
fixes #20

As expected, should using the plumpy event loop. But it shows a problem that the custom event loop from tests in other places not set back to the regular event loop.
@unkcpz unkcpz closed this Jan 31, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

1 participant