flytekit.remote.entities.FlyteWorkflow#

class flytekit.remote.entities.FlyteWorkflow(id, nodes, interface, output_bindings, metadata, metadata_defaults, subworkflows=None, tasks=None, launch_plans=None, compiled_closure=None, should_register=False)[source]#

A class encapsulating a remote Flyte workflow.

Methods

Parameters:

id (id_models.Identifier)
nodes (List[FlyteNode])
subworkflows (Optional[List[FlyteWorkflow]])
tasks (Optional[List[FlyteTask]])
launch_plans (Optional[Dict[id_models.Identifier, launch_plan_models.LaunchPlanSpec]])
compiled_closure (Optional[compiler_models.CompiledWorkflowClosure])
should_register (bool)

compile(ctx, *args, **kwargs)#

Parameters:: ctx (FlyteContext)

construct_node_metadata()#

Used when constructing the node that encapsulates this task as part of a broader workflow definition.

Return type:: NodeMetadata

execute(**kwargs)#

Return type:: Any

classmethod from_flyte_idl(pb2_object)#

Parameters:: pb2_object – flyteidl.admin.workflow_pb2.WorkflowSpec
Return type:: WorkflowSpec

classmethod get_non_system_nodes(nodes)[source]#

Parameters:: nodes (List[Node])
Return type:: List[Node]

local_execute(ctx, **kwargs)#

Parameters:: ctx (FlyteContext)
Return type:: Tuple[Promise] | Promise | VoidPromise | None

local_execution_mode()#

Return type:: Mode

classmethod promote_from_closure(closure, node_launch_plans=None)[source]#

Extracts out the relevant portions of a FlyteWorkflow from a closure from the control plane.

Parameters:

closure (compiler_models.CompiledWorkflowClosure) – This is the closure returned by Admin
node_launch_plans (Optional[Dict[id_models, launch_plan_models.LaunchPlanSpec]]) – The reason this exists is because the compiled closure doesn’t have launch plans. It only has subworkflows and tasks. Why this is unclear. If supplied, this map of launch plans will be

classmethod promote_from_model(base_model, sub_workflows=None, tasks=None, node_launch_plans=None)[source]#

Parameters:

base_model (WorkflowTemplate)
sub_workflows (Dict[Identifier, WorkflowTemplate] | None)
tasks (Dict[Identifier, FlyteTask] | None)
node_launch_plans (Dict[Identifier, LaunchPlanSpec] | None)

Return type:

FlyteWorkflow

serialize_to_string()#

Return type:: str

short_string()#

Return type:: Text

to_flyte_idl()#

Return type:: flyteidl.admin.workflow_pb2.WorkflowSpec

verbose_string()#

Return type:: Text

Attributes

docs

Description entity for the workflow

Type:: rtype

entity_type_text

failure_node

A catch-all node. This node is executed whenever the execution engine determines the workflow has failed. The interface of this node must match the Workflow interface with an additional input named “error” of type pb.lyft.flyte.core.Error.

Type:: Node failure_node

flyte_nodes

flyte_sub_workflows

flyte_tasks

id: This is an autogenerated id by the system. The id is globally unique across Flyte.

interface: Defines a strongly typed interface for the Workflow (inputs, outputs). This can include some optional parameters.

is_empty

metadata: This contains information on how to run the workflow.

metadata_defaults: This contains information on how to run the workflow. :rtype: WorkflowMetadataDefaults

name

nodes: A list of nodes. In addition, “globals” is a special reserved node id that can be used to consume workflow inputs

outputs: A list of output bindings that specify how to construct workflow outputs. Bindings can pull node outputs or specify literals. All workflow outputs specified in the interface field must be bound in order for the workflow to be validated. A workflow has an implicit dependency on all of its nodes to execute successfully in order to bind final outputs.

python_interface

resource_type

should_register

sub_workflows

list[flytekit.models.core.workflow.WorkflowTemplate]

Type:: rtype

template

flytekit.models.core.workflow.WorkflowTemplate

Type:: rtype