flytekit.core.python_function_task

========================================= :mod:flytekit.core.python_function_task

.. currentmodule:: flytekit.core.python_function_task

.. autosummary:: :nosignatures: :template: custom.rst :toctree: generated/

PythonFunctionTask PythonInstanceTask

Directory

Classes

Errors

• EagerException
• FlyteNonRecoverableSystemException
• FlyteValueException

flytekit.core.python_function_task.ABC

Helper class that provides a standard way to create an ABC using inheritance.

flytekit.core.python_function_task.Any

Special type indicating an unconstrained type.

• Any is compatible with every type.
• Any assumed to have all methods.
• All values assumed to be instances of Any.

Note that all the above statements are true from the point of view of static type checkers. At runtime, Any should not be used with instance checks.

flytekit.core.python_function_task.AsyncPythonFunctionTask

This is the base task for eager tasks, as well as normal async tasks Really only need to override the call function.

def AsyncPythonFunctionTask(
    task_config: T,
    task_function: Callable,
    task_type,
    ignore_input_vars: Optional[List[str]],
    execution_mode: ExecutionBehavior,
    task_resolver: Optional[TaskResolverMixin],
    node_dependency_hints: Optional[Iterable[Union['PythonFunctionTask', '_annotated_launch_plan.LaunchPlan', WorkflowBase]]],
    pickle_untyped: bool,
    kwargs,
):

Methods

async_execute()

def async_execute(
    args,
    kwargs,
):

Overrides the base execute function. This function does not handle dynamic at all. Eager and dynamic don’t mix.

compile()

def compile(
    ctx: flytekit.core.context_manager.FlyteContext,
    args,
    kwargs,
):

Generates a node that encapsulates this task in a workflow definition.

compile_into_workflow()

def compile_into_workflow(
    ctx: FlyteContext,
    task_function: Callable,
    kwargs,
):

In the case of dynamic workflows, this function will produce a workflow definition at execution time which will then proceed to be executed.

construct_node_metadata()

def construct_node_metadata()

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

dispatch_execute()

def dispatch_execute(
    ctx: flytekit.core.context_manager.FlyteContext,
    input_literal_map: flytekit.models.literals.LiteralMap,
):

This method translates Flyte’s Type system based input values and invokes the actual call to the executor This method is also invoked during runtime.

• VoidPromise is returned in the case when the task itself declares no outputs.
• Literal Map is returned when the task returns either one more outputs in the declaration. Individual outputs may be none
• DynamicJobSpec is returned when a dynamic workflow is executed

dynamic_execute()

def dynamic_execute(
    task_function: Callable,
    kwargs,
):

By the time this function is invoked, the local_execute function should have unwrapped the Promises and Flyte literal wrappers so that the kwargs we are working with here are now Python native literal values. This function is also expected to return Python native literal values.

Since the user code within a dynamic task constitute a workflow, we have to first compile the workflow, and then execute that workflow.

When running for real in production, the task would stop after the compilation step, and then create a file representing that newly generated workflow, instead of executing it.

execute()

def execute(
    args,
    kwargs,
):

Overrides the base execute function. This function does not handle dynamic at all. Eager and dynamic don’t mix.

find_lhs()

def find_lhs()

get_command()

def get_command(
    settings: SerializationSettings,
):

Returns the command which should be used in the container definition for the serialized version of this task registered on a hosted Flyte platform.

get_config()

def get_config(
    settings: SerializationSettings,
):

Returns the task config as a serializable dictionary. This task config consists of metadata about the custom defined for this task.

get_container()

def get_container(
    settings: SerializationSettings,
):

Returns the container definition (if any) that is used to run the task on hosted Flyte.

get_custom()

def get_custom(
    settings: flytekit.configuration.SerializationSettings,
):

Return additional plugin-specific custom data (if any) as a serializable dictionary.

get_default_command()

def get_default_command(
    settings: SerializationSettings,
):

Returns the default pyflyte-execute command used to run this on hosted Flyte platforms.

get_extended_resources()

def get_extended_resources(
    settings: SerializationSettings,
):

Returns the extended resources to allocate to the task on hosted Flyte.

get_image()

def get_image(
    settings: SerializationSettings,
):

Update image spec based on fast registration usage, and return string representing the image

get_input_types()

def get_input_types()

Returns the names and python types as a dictionary for the inputs of this task.

get_k8s_pod()

def get_k8s_pod(
    settings: SerializationSettings,
):

Returns the kubernetes pod definition (if any) that is used to run the task on hosted Flyte.

get_sql()

def get_sql(
    settings: flytekit.configuration.SerializationSettings,
):

Returns the Sql definition (if any) that is used to run the task on hosted Flyte.

get_type_for_input_var()

def get_type_for_input_var(
    k: str,
    v: typing.Any,
):

Returns the python type for an input variable by name.

get_type_for_output_var()

def get_type_for_output_var(
    k: str,
    v: typing.Any,
):

Returns the python type for the specified output variable by name.

local_execute()

def local_execute(
    ctx: flytekit.core.context_manager.FlyteContext,
    kwargs,
):

This function is used only in the local execution path and is responsible for calling dispatch execute. Use this function when calling a task with native values (or Promises containing Flyte literals derived from Python native values).

local_execution_mode()

def local_execution_mode()

post_execute()

def post_execute(
    user_params: typing.Optional[flytekit.core.context_manager.ExecutionParameters],
    rval: typing.Any,
):

Post execute is called after the execution has completed, with the user_params and can be used to clean-up, or alter the outputs to match the intended tasks outputs. If not overridden, then this function is a No-op

pre_execute()

def pre_execute(
    user_params: typing.Optional[flytekit.core.context_manager.ExecutionParameters],
):

This is the method that will be invoked directly before executing the task method and before all the inputs are converted. One particular case where this is useful is if the context is to be modified for the user process to get some user space parameters. This also ensures that things like SparkSession are already correctly setup before the type transformers are called

This should return either the same context of the mutated context

reset_command_fn()

def reset_command_fn()

Resets the command which should be used in the container definition of this task to the default arguments. This is useful when the command line is overridden at serialization time.

sandbox_execute()

def sandbox_execute(
    ctx: flytekit.core.context_manager.FlyteContext,
    input_literal_map: flytekit.models.literals.LiteralMap,
):

Call dispatch_execute, in the context of a local sandbox execution. Not invoked during runtime.

set_command_fn()

def set_command_fn(
    get_command_fn: Optional[Callable[[SerializationSettings], List[str]]],
):

By default, the task will run on the Flyte platform using the pyflyte-execute command. However, it can be useful to update the command with which the task is serialized for specific cases like running map tasks (“pyflyte-map-execute”) or for fast-executed tasks.

set_resolver()

def set_resolver(
    resolver: TaskResolverMixin,
):

By default, flytekit uses the DefaultTaskResolver to resolve the task. This method allows the user to set a custom task resolver. It can be useful to override the task resolver for specific cases like running tasks in the jupyter notebook.

Properties

flytekit.core.python_function_task.Controller

This controller object is responsible for kicking off and monitoring executions against a Flyte Admin endpoint using a FlyteRemote object. It is used only for running eager tasks. It exposes one async method, add, which should be called by the eager task to run a sub-flyte-entity (task, workflow, or a nested eager task).

The controller maintains a dictionary of entries, where each entry is a list of WorkItems. They are maintained in a list because the number of times and order that each task (or subwf, lp) is called affects the execution name which is consistently hashed.

After calling add, a background thread is started to reconcile the state of this dictionary of WorkItem entries. Executions that should be kicked off will be kicked off, and ones that are running will be checked. This runs in a loop similar to a controller loop in a k8s operator.

def Controller(
    remote: FlyteRemote,
    ss: SerializationSettings,
    tag: str,
    root_tag: str,
    exec_prefix: str,
):

Methods

add()

def add(
    entity: RunnableEntity,
    input_kwargs: dict[str, typing.Any],
):

Add an entity along with the requested inputs to be submitted to Admin for running and return a future

for_sandbox()

def for_sandbox(
    exec_prefix: typing.Optional[str],
):

get_env()

def get_env()

In order for downstream tasks to correctly set the root label, this needs to pass down that information.

get_execution_name()

def get_execution_name(
    entity: RunnableEntity,
    idx: int,
    input_kwargs: dict[str, typing.Any],
):

Make a deterministic name

get_labels()

def get_labels()

These labels keep track of the current and root (in case of nested) eager execution, that is responsible for kicking off this execution.

get_signal_handler()

def get_signal_handler()

TODO: At some point, this loop would be ideally managed by the loop manager, and the signal handler should gracefully initiate shutdown of all loops, calling .cancel() on all tasks, allowing each loop to clean up, starting with the deepest loop/thread first and working up. https://github.com/flyteorg/flyte/issues/6068

launch_execution()

def launch_execution(
    wi: WorkItem,
    idx: int,
):

This function launches executions.

reconcile_one()

def reconcile_one(
    update: Update,
):

This is responsible for processing one work item. Will launch, update, set error on the update object Any errors are captured in the update object.

render_html()

def render_html()

Render the callstack as a deck presentation to be shown after eager workflow execution.

flytekit.core.python_function_task.Deck

Deck enable users to get customizable and default visibility into their tasks.

Deck contains a list of renderers (FrameRenderer, MarkdownRenderer) that can generate a html file. For example, FrameRenderer can render a DataFrame as an HTML table, MarkdownRenderer can convert Markdown string to HTML

Flyte context saves a list of deck objects, and we use renderers in those decks to render the data and create an HTML file when those tasks are executed

Each task has a least three decks (input, output, default). Input/output decks are used to render tasks’ input/output data, and the default deck is used to render line plots, scatter plots or Markdown text. In addition, users can create new decks to render their data with custom renderers.

.. code-block:: python

iris_df = px.data.iris()

@task() def t1() -> str: md_text = ‘#Hello Flyte##Hello Flyte###Hello Flyte’ m = MarkdownRenderer() s = BoxRenderer(“sepal_length”) deck = flytekit.Deck(“demo”, s.to_html(iris_df)) deck.append(m.to_html(md_text)) default_deck = flytekit.current_context().default_deck default_deck.append(m.to_html(md_text)) return md_text

Use Annotated to override default renderer

@task() def t2() -> Annotated[pd.DataFrame, TopFrameRenderer(10)]: return iris_df

def Deck(
    name: str,
    html: typing.Optional[str],
    auto_add_to_deck: bool,
):

Methods

append()

def append(
    html: str,
):

publish()

def publish()

Properties

flytekit.core.python_function_task.Docstring

def Docstring(
    docstring: typing.Optional[str],
    callable_: typing.Optional[typing.Callable],
):

Properties

flytekit.core.python_function_task.EagerAsyncPythonFunctionTask

This is the base eager task (aka eager workflow) type. It replaces the previous experiment eager task type circa Q4 2024. Users unfamiliar with this concept should refer to the documentation for more information. But basically, Python becomes propeller, and every task invocation, creates a stack frame on the Flyte cluster in the form of an execution rather than on the actual memory stack.

def EagerAsyncPythonFunctionTask(
    task_config: T,
    task_function: Callable,
    task_type,
    ignore_input_vars: Optional[List[str]],
    task_resolver: Optional[TaskResolverMixin],
    node_dependency_hints: Optional[Iterable[Union['PythonFunctionTask', '_annotated_launch_plan.LaunchPlan', WorkflowBase]]],
    enable_deck: bool,
    kwargs,
):

Methods

async_execute()

def async_execute(
    args,
    kwargs,
):

Overrides the base execute function. This function does not handle dynamic at all. Eager and dynamic don’t mix.

Some notes on the different call scenarios since it’s a little different than other tasks. a) starting local execution - eager_task() -> last condition of call handler, -> set execution mode and self.local_execute() -> self.execute(native_vals) -> 1) -> task function() or 2) -> self.run_with_backend() # fn name will be changed. b) inside an eager task local execution - calling normal_task() -> call handler detects in eager local execution (middle part of call handler) -> call normal_task’s local_execute() c) inside an eager task local execution - calling async_normal_task() -> produces a coro, which when awaited/run -> call handler detects in eager local execution (middle part of call handler) -> call async_normal_task’s local_execute() -> call AsyncPythonFunctionTask’s async_execute(), which awaits the task function d) inside an eager task local execution - calling another_eager_task() -> produces a coro, which when awaited/run -> call handler detects in eager local execution (middle part of call handler) -> call another_eager_task’s local_execute() -> results are returned instead of being passed to create_native_named_tuple d) eager_task, starting backend execution from entrypoint.py -> eager_task.dispatch_execute(literals) -> eager_task.execute(native values) -> awaits eager_task.run_with_backend() # fn name will be changed e) in an eager task during backend execution, calling any flyte_entity() -> add the entity to the worker queue and await the result.

compile()

def compile(
    ctx: flytekit.core.context_manager.FlyteContext,
    args,
    kwargs,
):

Generates a node that encapsulates this task in a workflow definition.

compile_into_workflow()

def compile_into_workflow(
    ctx: FlyteContext,
    task_function: Callable,
    kwargs,
):

In the case of dynamic workflows, this function will produce a workflow definition at execution time which will then proceed to be executed.

construct_node_metadata()

def construct_node_metadata()

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

dispatch_execute()

def dispatch_execute(
    ctx: flytekit.core.context_manager.FlyteContext,
    input_literal_map: flytekit.models.literals.LiteralMap,
):

This method translates Flyte’s Type system based input values and invokes the actual call to the executor This method is also invoked during runtime.

• VoidPromise is returned in the case when the task itself declares no outputs.
• Literal Map is returned when the task returns either one more outputs in the declaration. Individual outputs may be none
• DynamicJobSpec is returned when a dynamic workflow is executed

dynamic_execute()

def dynamic_execute(
    task_function: Callable,
    kwargs,
):

By the time this function is invoked, the local_execute function should have unwrapped the Promises and Flyte literal wrappers so that the kwargs we are working with here are now Python native literal values. This function is also expected to return Python native literal values.

Since the user code within a dynamic task constitute a workflow, we have to first compile the workflow, and then execute that workflow.

When running for real in production, the task would stop after the compilation step, and then create a file representing that newly generated workflow, instead of executing it.

execute()

def execute(
    kwargs,
):

Overrides the base execute function. This function does not handle dynamic at all. Eager and dynamic don’t mix.

find_lhs()

def find_lhs()

get_as_workflow()

def get_as_workflow()

get_command()

def get_command(
    settings: SerializationSettings,
):

Returns the command which should be used in the container definition for the serialized version of this task registered on a hosted Flyte platform.

get_config()

def get_config(
    settings: SerializationSettings,
):

Returns the task config as a serializable dictionary. This task config consists of metadata about the custom defined for this task.

get_container()

def get_container(
    settings: SerializationSettings,
):

Returns the container definition (if any) that is used to run the task on hosted Flyte.

get_custom()

def get_custom(
    settings: flytekit.configuration.SerializationSettings,
):

Return additional plugin-specific custom data (if any) as a serializable dictionary.

get_default_command()

def get_default_command(
    settings: SerializationSettings,
):

Returns the default pyflyte-execute command used to run this on hosted Flyte platforms.

get_extended_resources()

def get_extended_resources(
    settings: SerializationSettings,
):

Returns the extended resources to allocate to the task on hosted Flyte.

get_image()

def get_image(
    settings: SerializationSettings,
):

Update image spec based on fast registration usage, and return string representing the image

get_input_types()

def get_input_types()

Returns the names and python types as a dictionary for the inputs of this task.

get_k8s_pod()

def get_k8s_pod(
    settings: SerializationSettings,
):

Returns the kubernetes pod definition (if any) that is used to run the task on hosted Flyte.

get_sql()

def get_sql(
    settings: flytekit.configuration.SerializationSettings,
):

Returns the Sql definition (if any) that is used to run the task on hosted Flyte.

get_type_for_input_var()

def get_type_for_input_var(
    k: str,
    v: typing.Any,
):

Returns the python type for an input variable by name.

get_type_for_output_var()

def get_type_for_output_var(
    k: str,
    v: typing.Any,
):

Returns the python type for the specified output variable by name.

local_execute()

def local_execute(
    ctx: flytekit.core.context_manager.FlyteContext,
    kwargs,
):

This function is used only in the local execution path and is responsible for calling dispatch execute. Use this function when calling a task with native values (or Promises containing Flyte literals derived from Python native values).

local_execution_mode()

def local_execution_mode()

post_execute()

def post_execute(
    user_params: typing.Optional[flytekit.core.context_manager.ExecutionParameters],
    rval: typing.Any,
):

Post execute is called after the execution has completed, with the user_params and can be used to clean-up, or alter the outputs to match the intended tasks outputs. If not overridden, then this function is a No-op

pre_execute()

def pre_execute(
    user_params: typing.Optional[flytekit.core.context_manager.ExecutionParameters],
):

This is the method that will be invoked directly before executing the task method and before all the inputs are converted. One particular case where this is useful is if the context is to be modified for the user process to get some user space parameters. This also ensures that things like SparkSession are already correctly setup before the type transformers are called

This should return either the same context of the mutated context

reset_command_fn()

def reset_command_fn()

Resets the command which should be used in the container definition of this task to the default arguments. This is useful when the command line is overridden at serialization time.

run()

def run(
    remote: 'FlyteRemote',
    ss: SerializationSettings,
    kwargs,
):

This is a helper function to help run eager parent tasks locally, pointing to a remote cluster. This is used only for local testing for now.

run_with_backend()

def run_with_backend(
    kwargs,
):

This is the main entry point to kick off a live run. Like if you’re running locally, but want to use a Flyte backend, or running for real on a Flyte backend.

sandbox_execute()

def sandbox_execute(
    ctx: flytekit.core.context_manager.FlyteContext,
    input_literal_map: flytekit.models.literals.LiteralMap,
):

Call dispatch_execute, in the context of a local sandbox execution. Not invoked during runtime.

set_command_fn()

def set_command_fn(
    get_command_fn: Optional[Callable[[SerializationSettings], List[str]]],
):

By default, the task will run on the Flyte platform using the pyflyte-execute command. However, it can be useful to update the command with which the task is serialized for specific cases like running map tasks (“pyflyte-map-execute”) or for fast-executed tasks.

set_resolver()

def set_resolver(
    resolver: TaskResolverMixin,
):

By default, flytekit uses the DefaultTaskResolver to resolve the task. This method allows the user to set a custom task resolver. It can be useful to override the task resolver for specific cases like running tasks in the jupyter notebook.

Properties

flytekit.core.python_function_task.EagerException

Raised when a node in an eager workflow encounters an error.

This exception should be used in an :py:func:@eager <flytekit.core.task.eager> workflow function to catch exceptions that are raised by tasks or subworkflows.

.. code-block:: python

from flytekit import task from flytekit.exceptions.eager import EagerException

@task def add_one(x: int) -> int: if x < 0: raise ValueError(“x must be positive”) return x + 1

@task def double(x: int) -> int: return x * 2

@eager async def eager_workflow(x: int) -> int: try: out = await add_one(x=x) except EagerException:

The ValueError error is caught

and raised as an EagerException

raise return await double(x=out)

flytekit.core.python_function_task.EagerFailureHandlerTask

A Python AutoContainer task should be used as the base for all extensions that want the user’s code to be in the container and the container information to be automatically captured. This base will auto configure the image and image version to be used for all its derivatives.

If you are looking to extend, you might prefer to use PythonFunctionTask or PythonInstanceTask

def EagerFailureHandlerTask(
    name: str,
    container_image: Optional[Union[str, ImageSpec]],
    inputs: typing.Optional[typing.Dict[str, typing.Type]],
    kwargs,
):

Methods

compile()

def compile(
    ctx: flytekit.core.context_manager.FlyteContext,
    args,
    kwargs,
):

Generates a node that encapsulates this task in a workflow definition.

construct_node_metadata()

def construct_node_metadata()

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

dispatch_execute()

def dispatch_execute(
    ctx: FlyteContext,
    input_literal_map: LiteralMap,
):

This task should only be called during remote execution. Because when rehydrating this task at execution time, we don’t have access to the python interface of the corresponding eager task/workflow, we don’t have the Python types to convert the input literal map, but nor do we need them. This task is responsible only for ensuring that all executions are terminated.

execute()

def execute(
    kwargs,
):

This method will be invoked to execute the task.

find_lhs()

def find_lhs()

get_command()

def get_command(
    settings: SerializationSettings,
):

Returns the command which should be used in the container definition for the serialized version of this task registered on a hosted Flyte platform.

get_config()

def get_config(
    settings: SerializationSettings,
):

Returns the task config as a serializable dictionary. This task config consists of metadata about the custom defined for this task.

get_container()

def get_container(
    settings: SerializationSettings,
):

Returns the container definition (if any) that is used to run the task on hosted Flyte.

get_custom()

def get_custom(
    settings: flytekit.configuration.SerializationSettings,
):

Return additional plugin-specific custom data (if any) as a serializable dictionary.

get_default_command()

def get_default_command(
    settings: SerializationSettings,
):

Returns the default pyflyte-execute command used to run this on hosted Flyte platforms.

get_extended_resources()

def get_extended_resources(
    settings: SerializationSettings,
):

Returns the extended resources to allocate to the task on hosted Flyte.

get_image()

def get_image(
    settings: SerializationSettings,
):

Update image spec based on fast registration usage, and return string representing the image

get_input_types()

def get_input_types()

Returns the names and python types as a dictionary for the inputs of this task.

get_k8s_pod()

def get_k8s_pod(
    settings: SerializationSettings,
):

Returns the kubernetes pod definition (if any) that is used to run the task on hosted Flyte.

get_sql()

def get_sql(
    settings: flytekit.configuration.SerializationSettings,
):

Returns the Sql definition (if any) that is used to run the task on hosted Flyte.

get_type_for_input_var()

def get_type_for_input_var(
    k: str,
    v: typing.Any,
):

Returns the python type for an input variable by name.

get_type_for_output_var()

def get_type_for_output_var(
    k: str,
    v: typing.Any,
):

Returns the python type for the specified output variable by name.

local_execute()

def local_execute(
    ctx: flytekit.core.context_manager.FlyteContext,
    kwargs,
):

This function is used only in the local execution path and is responsible for calling dispatch execute. Use this function when calling a task with native values (or Promises containing Flyte literals derived from Python native values).

local_execution_mode()

def local_execution_mode()

post_execute()

def post_execute(
    user_params: typing.Optional[flytekit.core.context_manager.ExecutionParameters],
    rval: typing.Any,
):

Post execute is called after the execution has completed, with the user_params and can be used to clean-up, or alter the outputs to match the intended tasks outputs. If not overridden, then this function is a No-op

pre_execute()

def pre_execute(
    user_params: typing.Optional[flytekit.core.context_manager.ExecutionParameters],
):

This is the method that will be invoked directly before executing the task method and before all the inputs are converted. One particular case where this is useful is if the context is to be modified for the user process to get some user space parameters. This also ensures that things like SparkSession are already correctly setup before the type transformers are called

This should return either the same context of the mutated context

reset_command_fn()

def reset_command_fn()

Resets the command which should be used in the container definition of this task to the default arguments. This is useful when the command line is overridden at serialization time.

sandbox_execute()

def sandbox_execute(
    ctx: flytekit.core.context_manager.FlyteContext,
    input_literal_map: flytekit.models.literals.LiteralMap,
):

Call dispatch_execute, in the context of a local sandbox execution. Not invoked during runtime.

set_command_fn()

def set_command_fn(
    get_command_fn: Optional[Callable[[SerializationSettings], List[str]]],
):

By default, the task will run on the Flyte platform using the pyflyte-execute command. However, it can be useful to update the command with which the task is serialized for specific cases like running map tasks (“pyflyte-map-execute”) or for fast-executed tasks.

set_resolver()

def set_resolver(
    resolver: TaskResolverMixin,
):

By default, flytekit uses the DefaultTaskResolver to resolve the task. This method allows the user to set a custom task resolver. It can be useful to override the task resolver for specific cases like running tasks in the jupyter notebook.

Properties

flytekit.core.python_function_task.EagerFailureTaskResolver

Flytekit tasks interact with the Flyte platform very, very broadly in two steps. They need to be uploaded to Admin, and then they are run by the user upon request (either as a single task execution or as part of a workflow). In any case, at execution time, for most tasks (that is those that generate a container target) the container image containing the task needs to be spun up again at which point the container needs to know which task it’s supposed to run and how to rehydrate the task object.

For example, the serialization of a simple task ::

in repo_root/workflows/example.py

@task def t1(…) -> …: …

might result in a container with arguments like ::

pyflyte-execute –inputs s3://path/inputs.pb –output-prefix s3://outputs/location –raw-output-data-prefix /tmp/data –resolver flytekit.core.python_auto_container.default_task_resolver – task-module repo_root.workflows.example task-name t1

At serialization time, the container created for the task will start out automatically with the pyflyte-execute bit, along with the requisite input/output args and the offloaded data prefix. Appended to that will be two things,

#. the location of the task’s task resolver, followed by two dashes, followed by #. the arguments provided by calling the loader_args function below.

The default_task_resolver declared below knows that

• When loader_args is called on a task, to look up the module the task is in, and the name of the task (the key of the task in the module, either the function name, or the variable it was assigned to).
• When load_task is called, it interprets the first part of the command as the module to call importlib.import_module on, and then looks for a key t1.

This is just the default behavior. Users should feel free to implement their own resolvers.

Methods

get_all_tasks()

def get_all_tasks()

Future proof method. Just making it easy to access all tasks (Not required today as we auto register them)

load_task()

def load_task(
    loader_args: List[str],
):

Given the set of identifier keys, should return one Python Task or raise an error if not found

loader_args()

def loader_args(
    settings: SerializationSettings,
    t: Task,
):

Return a list of strings that can help identify the parameter Task

name()

def name()

task_name()

def task_name(
    t: flytekit.core.base_task.Task,
):

Overridable function that can optionally return a custom name for a given task

Properties

flytekit.core.python_function_task.Enum

Create a collection of name/value pairs.

Example enumeration:

class Color(Enum): … RED = 1 … BLUE = 2 … GREEN = 3

Access them by:

• attribute access:

Color.RED <Color.RED: 1>

• value lookup:

Color(1) <Color.RED: 1>

• name lookup:

Color[‘RED’] <Color.RED: 1>

Enumerations can be iterated over, and know how many members they have:

len(Color) 3

list(Color) [<Color.RED: 1>, <Color.BLUE: 2>, <Color.GREEN: 3>]

Methods can be added to enumerations, and members can have their own attributes – see the documentation for details.

flytekit.core.python_function_task.ExecutionState

This is the context that is active when executing a task or a local workflow. This carries the necessary state to execute. Some required things during execution deal with temporary directories, ExecutionParameters that are passed to the user etc.

Attributes: mode (ExecutionState.Mode): Defines the context in which the task is executed (local, hosted, etc). working_dir (os.PathLike): Specifies the remote, external directory where inputs, outputs and other protobufs are uploaded engine_dir (os.PathLike): branch_eval_mode Optional[BranchEvalMode]: Used to determine whether a branch node should execute. user_space_params Optional[ExecutionParameters]: Provides run-time, user-centric context such as a statsd handler, a logging handler, the current execution id and a working directory.

def ExecutionState(
    working_dir: Union[os.PathLike, str],
    mode: Optional[ExecutionState.Mode],
    engine_dir: Optional[Union[os.PathLike, str]],
    branch_eval_mode: Optional[BranchEvalMode],
    user_space_params: Optional[ExecutionParameters],
):

Methods

branch_complete()

def branch_complete()

Indicates that we are within a conditional / ifelse block and the active branch is not done. Default to SKIPPED

is_local_execution()

def is_local_execution()

take_branch()

def take_branch()

Indicates that we are within an if-else block and the current branch has evaluated to true. Useful only in local execution mode

with_params()

def with_params(
    working_dir: Optional[os.PathLike],
    mode: Optional[Mode],
    engine_dir: Optional[os.PathLike],
    branch_eval_mode: Optional[BranchEvalMode],
    user_space_params: Optional[ExecutionParameters],
):

Produces a copy of the current execution state and overrides the copy’s parameters with passed parameter values.

flytekit.core.python_function_task.FlyteContext

This is an internal-facing context object, that most users will not have to deal with. It’s essentially a globally available grab bag of settings and objects that allows flytekit to do things like convert complex types, run and compile workflows, serialize Flyte entities, etc.

Even though this object as a current_context function on it, it should not be called directly. Please use the :py:class:flytekit.FlyteContextManager object instead.

Please do not confuse this object with the :py:class:flytekit.ExecutionParameters object.

def FlyteContext(
    file_access: FileAccessProvider,
    level: int,
    flyte_client: Optional['friendly_client.SynchronousFlyteClient'],
    compilation_state: Optional[CompilationState],
    execution_state: Optional[ExecutionState],
    serialization_settings: Optional[SerializationSettings],
    in_a_condition: bool,
    origin_stackframe: Optional[traceback.FrameSummary],
    output_metadata_tracker: Optional[OutputMetadataTracker],
    worker_queue: Optional[Controller],
):

Methods

current_context()

def current_context()

This method exists only to maintain backwards compatibility. Please use FlyteContextManager.current_context() instead.

Users of flytekit should be wary not to confuse the object returned from this function with :py:func:flytekit.current_context

enter_conditional_section()

def enter_conditional_section()

get_deck()

def get_deck()

Returns the deck that was created as part of the last execution.

The return value depends on the execution environment. In a notebook, the return value is compatible with IPython.display and should be rendered in the notebook.

.. code-block:: python

with flytekit.new_context() as ctx: my_task(…) ctx.get_deck()

OR if you wish to explicitly display

.. code-block:: python

from IPython import display display(ctx.get_deck())

get_origin_stackframe_repr()

def get_origin_stackframe_repr()

new_builder()

def new_builder()

new_compilation_state()

def new_compilation_state(
    prefix: str,
):

Creates and returns a default compilation state. For most of the code this should be the entrypoint of compilation, otherwise the code should always uses - with_compilation_state

new_execution_state()

def new_execution_state(
    working_dir: Optional[os.PathLike],
):

Creates and returns a new default execution state. This should be used at the entrypoint of execution, in all other cases it is preferable to use with_execution_state

set_stackframe()

def set_stackframe(
    s: traceback.FrameSummary,
):

with_client()

def with_client(
    c: SynchronousFlyteClient,
):

with_compilation_state()

def with_compilation_state(
    c: CompilationState,
):

with_execution_state()

def with_execution_state(
    es: ExecutionState,
):

with_file_access()

def with_file_access(
    fa: FileAccessProvider,
):

with_new_compilation_state()

def with_new_compilation_state()

with_output_metadata_tracker()

def with_output_metadata_tracker(
    t: OutputMetadataTracker,
):

with_serialization_settings()

def with_serialization_settings(
    ss: SerializationSettings,
):

with_worker_queue()

def with_worker_queue(
    wq: Controller,
):

Properties

flytekit.core.python_function_task.FlyteContextManager

FlyteContextManager manages the execution context within Flytekit. It holds global state of either compilation or Execution. It is not thread-safe and can only be run as a single threaded application currently. Context’s within Flytekit is useful to manage compilation state and execution state. Refer to CompilationState and ExecutionState for more information. FlyteContextManager provides a singleton stack to manage these contexts.

Typical usage is

.. code-block:: python

FlyteContextManager.initialize() with FlyteContextManager.with_context(o) as ctx: pass

If required - not recommended you can use

FlyteContextManager.push_context()

but correspondingly a pop_context should be called

FlyteContextManager.pop_context()

Methods

add_signal_handler()

def add_signal_handler(
    handler: typing.Callable[[int, FrameType], typing.Any],
):

current_context()

def current_context()

get_origin_stackframe()

def get_origin_stackframe(
    limit,
):

initialize()

def initialize()

Re-initializes the context and erases the entire context

pop_context()

def pop_context()

push_context()

def push_context(
    ctx: FlyteContext,
    f: Optional[traceback.FrameSummary],
):

size()

def size()

with_context()

def with_context(
    b: FlyteContext.Builder,
):

flytekit.core.python_function_task.FlyteNonRecoverableSystemException

Common base class for all non-exit exceptions.

def FlyteNonRecoverableSystemException(
    exc_value: Exception,
):

FlyteNonRecoverableSystemException is thrown when a system code raises an exception.

Properties

flytekit.core.python_function_task.FlyteTrackedABC

This class exists because if you try to inherit from abc.ABC and TrackedInstance by itself, you’ll get the well-known TypeError: metaclass conflict: the metaclass of a derived class must be a (non-strict) subclass of the metaclasses of all its bases error.

Methods

register()

def register(
    cls,
    subclass,
):

Register a virtual subclass of an ABC.

Returns the subclass, to allow usage as a class decorator.

flytekit.core.python_function_task.FlyteValueException

Inappropriate argument value (of correct type).

def FlyteValueException(
    received_value,
    error_message,
):

Properties

flytekit.core.python_function_task.ImageConfig

We recommend you to use ImageConfig.auto(img_name=None) to create an ImageConfig. For example, ImageConfig.auto(img_name=““ghcr.io/flyteorg/flytecookbook:v1.0.0"”) will create an ImageConfig.

ImageConfig holds available images which can be used at registration time. A default image can be specified along with optional additional images. Each image in the config must have a unique name.

Attributes: default_image (Optional[Image]): The default image to be used as a container for task serialization. images (List[Image]): Optional, additional images which can be used in task container definitions.

def ImageConfig(
    default_image: Optional[Image],
    images: Optional[List[Image]],
):

Methods

auto()

def auto(
    config_file: typing.Union[str, ConfigFile, None],
    img_name: Optional[str],
):

Reads from config file or from img_name Note that this function does not take into account the flytekit default images (see the Dockerfiles at the base of this repo). To pick those up, see the auto_default_image function..

auto_default_image()

def auto_default_image()

create_from()

def create_from(
    default_image: Optional[Image],
    other_images: typing.Optional[typing.List[Image]],
):

find_image()

def find_image(
    name,
):

Return an image, by name, if it exists.

from_dict()

def from_dict(
    kvs: typing.Union[dict, list, str, int, float, bool, NoneType],
    infer_missing,
):

from_images()

def from_images(
    default_image: str,
    m: typing.Optional[typing.Dict[str, str]],
):

Allows you to programmatically create an ImageConfig. Usually only the default_image is required, unless your workflow uses multiple images

.. code:: python

ImageConfig.from_dict( “ghcr.io/flyteorg/flytecookbook:v1.0.0”, { “spark”: “ghcr.io/flyteorg/myspark:…”, “other”: “…”, } )

urn:

from_json()

def from_json(
    s: typing.Union[str, bytes, bytearray],
    parse_float,
    parse_int,
    parse_constant,
    infer_missing,
    kw,
):

schema()

def schema(
    infer_missing: bool,
    only,
    exclude,
    many: bool,
    context,
    load_only,
    dump_only,
    partial: bool,
    unknown,
):

to_dict()

def to_dict(
    encode_json,
):

to_json()

def to_json(
    skipkeys: bool,
    ensure_ascii: bool,
    check_circular: bool,
    allow_nan: bool,
    indent: typing.Union[int, str, NoneType],
    separators: typing.Tuple[str, str],
    default: typing.Callable,
    sort_keys: bool,
    kw,
):

validate_image()

def validate_image(
    _: typing.Any,
    param: str,
    values: tuple,
):

Validates the image to match the standard format. Also validates that only one default image is provided. a default image, is one that is specified as default=<image_uri> or just <image_uri>. All other images should be provided with a name, in the format name=<image_uri> This method can be used with the CLI

flytekit.core.python_function_task.ImageSpec

This class is used to specify the docker image that will be used to run the task.

def ImageSpec(
    name: str,
    python_version: str,
    builder: typing.Optional[str],
    source_root: typing.Optional[str],
    env: typing.Optional[typing.Dict[str, str]],
    registry: typing.Optional[str],
    packages: typing.Optional[typing.List[str]],
    conda_packages: typing.Optional[typing.List[str]],
    conda_channels: typing.Optional[typing.List[str]],
    requirements: typing.Optional[str],
    apt_packages: typing.Optional[typing.List[str]],
    cuda: typing.Optional[str],
    cudnn: typing.Optional[str],
    base_image: typing.Union[str, ForwardRef('ImageSpec'), NoneType],
    platform: str,
    pip_index: typing.Optional[str],
    pip_extra_index_url: typing.Optional[typing.List[str]],
    pip_secret_mounts: typing.Optional[typing.List[typing.Tuple[str, str]]],
    pip_extra_args: typing.Optional[str],
    registry_config: typing.Optional[str],
    entrypoint: typing.Optional[typing.List[str]],
    commands: typing.Optional[typing.List[str]],
    tag_format: typing.Optional[str],
    source_copy_mode: typing.Optional[flytekit.constants.CopyFileDetection],
    copy: typing.Optional[typing.List[str]],
    python_exec: typing.Optional[str],
):

Methods

exist()

def exist()

Check if the image exists in the registry. Return True if the image exists in the registry, False otherwise. Return None if failed to check if the image exists due to the permission issue or other reasons.

force_push()

def force_push()

Builder that returns a new image spec with force push enabled.

from_env()

def from_env(
    pinned_packages: typing.Optional[typing.List[str]],
    kwargs,
):

Create ImageSpec with the environment’s Python version and packages pinned to the ones in the environment.

image_name()

def image_name()

Full image name with tag.

is_container()

def is_container()

Check if the current container image in the pod is built from current image spec. :return: True if the current container image in the pod is built from current image spec, False otherwise.

with_apt_packages()

def with_apt_packages(
    apt_packages: typing.Union[str, typing.List[str]],
):

Builder that returns a new image spec with an additional list of apt packages that will be executed during the building process.

with_commands()

def with_commands(
    commands: typing.Union[str, typing.List[str]],
):

Builder that returns a new image spec with an additional list of commands that will be executed during the building process.

with_copy()

def with_copy(
    src: typing.Union[str, typing.List[str]],
):

Builder that returns a new image spec with the source files copied to the destination directory.

with_packages()

def with_packages(
    packages: typing.Union[str, typing.List[str]],
):

Builder that returns a new image speck with additional python packages that will be installed during the building process.

Properties

flytekit.core.python_function_task.Interface

A Python native interface object, like inspect.signature but simpler.

def Interface(
    inputs: Union[Optional[Dict[str, Type]], Optional[Dict[str, Tuple[Type, Any]]]],
    outputs: Union[Optional[Dict[str, Type]], Optional[Dict[str, Optional[Type]]]],
    output_tuple_name: Optional[str],
    docstring: Optional[Docstring],
):

Methods

remove_inputs()

def remove_inputs(
    vars: Optional[List[str]],
):

This method is useful in removing some variables from the Flyte backend inputs specification, as these are implicit local only inputs or will be supplied by the library at runtime. For example, spark-session etc It creates a new instance of interface with the requested variables removed

with_inputs()

def with_inputs(
    extra_inputs: Dict[str, Type],
):

Use this to add additional inputs to the interface. This is useful for adding additional implicit inputs that are added without the user requesting for them

with_outputs()

def with_outputs(
    extra_outputs: Dict[str, Type],
):

This method allows addition of extra outputs are expected from a task specification

Properties

flytekit.core.python_function_task.LiteralMap

def LiteralMap(
    literals,
):

Methods

from_flyte_idl()

def from_flyte_idl(
    pb2_object,
):

serialize_to_string()

def serialize_to_string()

short_string()

def short_string()

to_flyte_idl()

def to_flyte_idl()

verbose_string()

def verbose_string()

Properties

flytekit.core.python_function_task.OrderedDict

Dictionary that remembers insertion order

flytekit.core.python_function_task.Promise

This object is a wrapper and exists for three main reasons. Let’s assume we’re dealing with a task like ::

@task def t1() -> (int, str): …

#. Handling the duality between compilation and local execution - when the task function is run in a local execution mode inside a workflow function, a Python integer and string are produced. When the task is being compiled as part of the workflow, the task call creates a Node instead, and the task returns two Promise objects that point to that Node. #. One needs to be able to call ::

x = t1().with_overrides(…)

If the task returns an integer or a (int, str) tuple like t1 above, calling with_overrides on the result would throw an error. This Promise object adds that. #. Assorted handling for conditionals.

def Promise(
    var: str,
    val: Union[NodeOutput, _literals_models.Literal],
    type: typing.Optional[_type_models.LiteralType],
):

Methods

deepcopy()

def deepcopy()

eval()

def eval()

is_()

def is_(
    v: bool,
):

is_false()

def is_false()

is_none()

def is_none()

is_true()

def is_true()

with_overrides()

def with_overrides(
    node_name: Optional[str],
    aliases: Optional[Dict[str, str]],
    requests: Optional[Resources],
    limits: Optional[Resources],
    timeout: Optional[Union[int, datetime.timedelta, object]],
    retries: Optional[int],
    interruptible: Optional[bool],
    name: Optional[str],
    task_config: Optional[Any],
    container_image: Optional[str],
    accelerator: Optional[BaseAccelerator],
    cache: Optional[bool],
    cache_version: Optional[str],
    cache_serialize: Optional[bool],
    args,
    kwargs,
):

with_var()

def with_var(
    new_var: str,
):

Properties

flytekit.core.python_function_task.PythonAutoContainerTask

A Python AutoContainer task should be used as the base for all extensions that want the user’s code to be in the container and the container information to be automatically captured. This base will auto configure the image and image version to be used for all its derivatives.

If you are looking to extend, you might prefer to use PythonFunctionTask or PythonInstanceTask

def PythonAutoContainerTask(
    name: str,
    task_config: T,
    task_type,
    container_image: Optional[Union[str, ImageSpec]],
    requests: Optional[Resources],
    limits: Optional[Resources],
    environment: Optional[Dict[str, str]],
    task_resolver: Optional[TaskResolverMixin],
    secret_requests: Optional[List[Secret]],
    pod_template: Optional[PodTemplate],
    pod_template_name: Optional[str],
    accelerator: Optional[BaseAccelerator],
    shared_memory: Optional[Union[L[True], str]],
    resources: Optional[Resources],
    kwargs,
):

Methods

compile()

def compile(
    ctx: flytekit.core.context_manager.FlyteContext,
    args,
    kwargs,
):

Generates a node that encapsulates this task in a workflow definition.

construct_node_metadata()

def construct_node_metadata()

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

dispatch_execute()

def dispatch_execute(
    ctx: flytekit.core.context_manager.FlyteContext,
    input_literal_map: flytekit.models.literals.LiteralMap,
):

This method translates Flyte’s Type system based input values and invokes the actual call to the executor This method is also invoked during runtime.

• VoidPromise is returned in the case when the task itself declares no outputs.
• Literal Map is returned when the task returns either one more outputs in the declaration. Individual outputs may be none
• DynamicJobSpec is returned when a dynamic workflow is executed

execute()

def execute(
    kwargs,
):

This method will be invoked to execute the task.

find_lhs()

def find_lhs()

get_command()

def get_command(
    settings: SerializationSettings,
):

Returns the command which should be used in the container definition for the serialized version of this task registered on a hosted Flyte platform.

get_config()

def get_config(
    settings: SerializationSettings,
):

Returns the task config as a serializable dictionary. This task config consists of metadata about the custom defined for this task.

get_container()

def get_container(
    settings: SerializationSettings,
):

Returns the container definition (if any) that is used to run the task on hosted Flyte.

get_custom()

def get_custom(
    settings: flytekit.configuration.SerializationSettings,
):

Return additional plugin-specific custom data (if any) as a serializable dictionary.

get_default_command()

def get_default_command(
    settings: SerializationSettings,
):

Returns the default pyflyte-execute command used to run this on hosted Flyte platforms.

get_extended_resources()

def get_extended_resources(
    settings: SerializationSettings,
):

Returns the extended resources to allocate to the task on hosted Flyte.

get_image()

def get_image(
    settings: SerializationSettings,
):

Update image spec based on fast registration usage, and return string representing the image

get_input_types()

def get_input_types()

Returns the names and python types as a dictionary for the inputs of this task.

get_k8s_pod()

def get_k8s_pod(
    settings: SerializationSettings,
):

Returns the kubernetes pod definition (if any) that is used to run the task on hosted Flyte.

get_sql()

def get_sql(
    settings: flytekit.configuration.SerializationSettings,
):