rebase.Step
SDK handle for a deployable workflow step.
Signature
class rebase.Step(rebase.Function)Step extends rebase.Function with workflow graph behavior and step execution settings. Most code should create steps with @rebase.step(...) or @project.step(...).
Helper
rebase.step(
fn: Callable | None = None,
*,
project: str | None = None,
name: str | None = None,
description: str | None = None,
default_parameters: dict[str, Any] | None = None,
enabled: bool = True,
retries: int = 0,
timeout_seconds: int | float | None = None,
cache: bool = False,
deploy_source: str | None = None,
) -> Callable[[Callable], rebase.Step] | rebase.StepAdditional Parameters
| Parameter | Type | Description |
|---|---|---|
retries | int | Retry count for workflow step execution. A step that raises is re-run up to this many more times; a step that times out is not (its work cannot be reclaimed). |
timeout_seconds | `int | float |
cache | bool | Whether step caching is enabled. |
deploy_source | `str | None` |
Steps take no image, dependency, or sizing parameters. They run inside their workflow's run and inherit its image and resources.
Methods
step.spawn(**parameters) -> rebase.Run
step.remote(**parameters) -> Any
step.run(**parameters) -> rebase.Run
step.submit(*args, **kwargs) -> rebase.Run| Method | Description |
|---|---|
submit(*args, **kwargs) | Start a step run after binding positional arguments to the wrapped Python function signature. |
spawn(**parameters) | Start a run and return immediately. |
remote(**parameters) | Start a run and wait for the result. |
run(**parameters) | Alias for spawn(**parameters). |
Steps deploy with their workflow. step.deploy() and rebase.deploy(step) raise RebaseWorkflowError; deploy the workflow instead.