[PoC]: Declarative Workflows in V2 SDK #292
Open
Conversation
This file contains 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
mrkaye97
commented
Jan 3, 2025
Comment on lines
+26
to
+27
| workflow_input = greet_workflow.workflow_input(ctx) | ||
| greeting = workflow_input.greeting |
There was a problem hiding this comment.
these type check correctly, so no need to cast or parse from a dict to an ExampleInput
mrkaye97
commented
Jan 3, 2025
Comment on lines
+31
to
+33
| input=language_counter_workflow.construct_spawn_workflow_input( | ||
| input=workflow_input | ||
| ), |
There was a problem hiding this comment.
types are enforced here, so you know the shape of input in your IDE + at type checking time
mrkaye97
commented
Jan 3, 2025
Comment on lines
+24
to
+25
| @greet_workflow.declare() | ||
| async def greet(ctx: Context) -> dict[Literal["message"], str]: |
There was a problem hiding this comment.
single decorator to register a function in Hatchet
mrkaye97
commented
Jan 3, 2025
| workflow_input = greet_workflow.workflow_input(ctx) | ||
| greeting = workflow_input.greeting | ||
|
|
||
| await language_counter_workflow.spawn( |
There was a problem hiding this comment.
refer to the workflow as an object instead of passing a string with the name
mrkaye97
commented
Jan 3, 2025
Comment on lines
+47
to
+49
| greeting = language_counter_workflow.workflow_input(ctx).greeting | ||
|
|
||
| match greeting: |
There was a problem hiding this comment.
type checking here for greeting
grutt
approved these changes
Jan 14, 2025
reillyse
approved these changes
Jan 15, 2025
|
@grutt Think we should merge this or wait for a couple weeks till we have the real V2 hopefully done? |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
This is an initial PoC for declarative workflows in the V2 SDK. We're seeking to accomplish a couple of things here:
I added Pydantic support to the V2 SDK here, and then added a
DeclarativeWorkflow(we can rename this to step, function, etc. if we want) generic-typed class that lets the user declaratively define their workflows and share the definitions across the codebase. The example provided should hopefully give a sense of what's possible with this approach.The TL;DR is that with just a few lines of code in the
workflow.pyin the example, we get type checking for workflow inputs, both when accessing them with<< workflow >>.workflow_input()and when spawning them with<< workflow >>.spawn()or<< workflow >>.run(). We also get "name-inference", meaning we don't need to pass a string value for the name of the workflow (or step or function) when we spawn it, which is another possible cause of errors in the current SDK.Here are a few results from running the example here:
First
greetrun:First child (counter) run
Second greet run:
Second child run:
Pydantic support: