WIP: Add support for running in environments that have a running asyncio loop#388
Open
WIP: Add support for running in environments that have a running asyncio loop#388
Conversation
Jupyter's ipykernel starts an asyncio loop in the same thread as it is executing the xircuits workflow. When using AsyncComponent in that environment directly, we would encounter an exception thrown by AsyncToSync telling us to just await the thing. A simple workaround to this is to not use ipykernel based execution by selecting the "Terminal Run" option. However, that requires the user to know about the issue and understand the workaround. This change instead allows the workflow to adapt to its environment.
|
|
treo
changed the title
Contributor
Author
|
It seems that there may be additional constraints that need to be kept in mind when using this approach. In particular, it looks like the converse component library doesn't like it. |
Contributor
Author
|
I've found a way that seems to work, but it will require a patch to converse as well. The good news is that it should simplify the ConverseRun component so it can simply await the server task, but maybe there are issues with not natively async component libraries that start servers? I haven't tested that yet. |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Description
Jupyter's ipykernel starts an asyncio loop in the same thread as it is executing the xircuits workflow. When using AsyncComponent (see #351) in that environment directly, we would encounter an exception thrown by AsyncToSync telling us to just await the thing.
A simple workaround to this is to not use ipykernel based execution by selecting the "Terminal Run" option. However, that requires the user to know about the issue and understand the workaround.
This change instead allows the workflow to adapt to its environment.
Pull Request Type
Type of Change