prefect.server.orchestration.rules

Prefect's flow and task-run orchestration machinery.

This module contains all the core concepts necessary to implement Prefect's state orchestration engine. These states correspond to intuitive descriptions of all the points that a Prefect flow or task can observe executing user code and intervene, if necessary. A detailed description of states can be found in our concept documentation.

Prefect's orchestration engine operates under the assumption that no governed user code will execute without first requesting Prefect REST API validate a change in state and record metadata about the run. With all attempts to run user code being checked against a Prefect instance, the Prefect REST API database becomes the unambiguous source of truth for managing the execution of complex interacting workflows. Orchestration rules can be implemented as discrete units of logic that operate against each state transition and can be fully observable, extensible, and customizable -- all without needing to store or parse a single line of user code.

BaseOrchestrationRule

Bases: AbstractAsyncContextManager

An abstract base class used to implement a discrete piece of orchestration logic.

An OrchestrationRule is a stateful context manager that directly governs a state transition. Complex orchestration is achieved by nesting multiple rules. Each rule runs against an OrchestrationContext that contains the transition details; this context is then passed to subsequent rules. The context can be modified by hooks that fire before and after a new state is validated and committed to the database. These hooks will fire as long as the state transition is considered "valid" and govern a transition by either modifying the proposed state before it is validated or by producing a side-effect.

A state transition occurs whenever a flow- or task- run changes state, prompting Prefect REST API to decide whether or not this transition can proceed. The current state of the run is referred to as the "initial state", and the state a run is attempting to transition into is the "proposed state". Together, the initial state transitioning into the proposed state is the intended transition that is governed by these orchestration rules. After using rules to enter a runtime context, the OrchestrationContext will contain a proposed state that has been governed by each rule, and at that point can validate the proposed state and commit it to the database. The validated state will be set on the context as context.validated_state, and rules will call the self.after_transition hook upon exiting the managed context.

Examples:

Create a rule:

>>> class BasicRule(BaseOrchestrationRule):
>>>     # allowed initial state types
>>>     FROM_STATES = [StateType.RUNNING]
>>>     # allowed proposed state types
>>>     TO_STATES = [StateType.COMPLETED, StateType.FAILED]
>>>
>>>     async def before_transition(initial_state, proposed_state, ctx):
>>>         # side effects and proposed state mutation can happen here
>>>         ...
>>>
>>>     async def after_transition(initial_state, validated_state, ctx):
>>>         # operations on states that have been validated can happen here
>>>         ...
>>>
>>>     async def cleanup(intitial_state, validated_state, ctx):
>>>         # reverts side effects generated by `before_transition` if necessary
>>>         ...

Use a rule:

>>> intended_transition = (StateType.RUNNING, StateType.COMPLETED)
>>> async with BasicRule(context, *intended_transition):
>>>     # context.proposed_state has been governed by BasicRule
>>>     ...

Use multiple rules:

>>> rules = [BasicRule, BasicRule]
>>> intended_transition = (StateType.RUNNING, StateType.COMPLETED)
>>> async with contextlib.AsyncExitStack() as stack:
>>>     for rule in rules:
>>>         stack.enter_async_context(rule(context, *intended_transition))
>>>
>>>     # context.proposed_state has been governed by all rules
>>>     ...

Attributes:

Name	Type	Description
FROM_STATES	Iterable	list of valid initial state types this rule governs
TO_STATES	Iterable	list of valid proposed state types this rule governs
context		the orchestration context
from_state_type		the state type a run is currently in
to_state_type		the intended proposed state type prior to any orchestration

Parameters:

Name	Type	Description	Default
context	OrchestrationContext	A FlowOrchestrationContext or TaskOrchestrationContext that is passed between rules	required
from_state_type	Optional[StateType]	The state type of the initial state of a run, if this state type is not contained in FROM_STATES, no hooks will fire	required
to_state_type	Optional[StateType]	The state type of the proposed state before orchestration, if this state type is not contained in TO_STATES, no hooks will fire	required

Source code in src/prefect/server/orchestration/rules.py

class BaseOrchestrationRule(contextlib.AbstractAsyncContextManager):
    """
    An abstract base class used to implement a discrete piece of orchestration logic.

    An `OrchestrationRule` is a stateful context manager that directly governs a state
    transition. Complex orchestration is achieved by nesting multiple rules.
    Each rule runs against an `OrchestrationContext` that contains the transition
    details; this context is then passed to subsequent rules. The context can be
    modified by hooks that fire before and after a new state is validated and committed
    to the database. These hooks will fire as long as the state transition is
    considered "valid" and govern a transition by either modifying the proposed state
    before it is validated or by producing a side-effect.

    A state transition occurs whenever a flow- or task- run changes state, prompting
    Prefect REST API to decide whether or not this transition can proceed. The current state of
    the run is referred to as the "initial state", and the state a run is
    attempting to transition into is the "proposed state". Together, the initial state
    transitioning into the proposed state is the intended transition that is governed
    by these orchestration rules. After using rules to enter a runtime context, the
    `OrchestrationContext` will contain a proposed state that has been governed by
    each rule, and at that point can validate the proposed state and commit it to
    the database. The validated state will be set on the context as
    `context.validated_state`, and rules will call the `self.after_transition` hook
    upon exiting the managed context.

    Examples:

        Create a rule:

        >>> class BasicRule(BaseOrchestrationRule):
        >>>     # allowed initial state types
        >>>     FROM_STATES = [StateType.RUNNING]
        >>>     # allowed proposed state types
        >>>     TO_STATES = [StateType.COMPLETED, StateType.FAILED]
        >>>
        >>>     async def before_transition(initial_state, proposed_state, ctx):
        >>>         # side effects and proposed state mutation can happen here
        >>>         ...
        >>>
        >>>     async def after_transition(initial_state, validated_state, ctx):
        >>>         # operations on states that have been validated can happen here
        >>>         ...
        >>>
        >>>     async def cleanup(intitial_state, validated_state, ctx):
        >>>         # reverts side effects generated by `before_transition` if necessary
        >>>         ...

        Use a rule:

        >>> intended_transition = (StateType.RUNNING, StateType.COMPLETED)
        >>> async with BasicRule(context, *intended_transition):
        >>>     # context.proposed_state has been governed by BasicRule
        >>>     ...

        Use multiple rules:

        >>> rules = [BasicRule, BasicRule]
        >>> intended_transition = (StateType.RUNNING, StateType.COMPLETED)
        >>> async with contextlib.AsyncExitStack() as stack:
        >>>     for rule in rules:
        >>>         stack.enter_async_context(rule(context, *intended_transition))
        >>>
        >>>     # context.proposed_state has been governed by all rules
        >>>     ...

    Attributes:
        FROM_STATES: list of valid initial state types this rule governs
        TO_STATES: list of valid proposed state types this rule governs
        context: the orchestration context
        from_state_type: the state type a run is currently in
        to_state_type: the intended proposed state type prior to any orchestration

    Args:
        context: A `FlowOrchestrationContext` or `TaskOrchestrationContext` that is
            passed between rules
        from_state_type: The state type of the initial state of a run, if this
            state type is not contained in `FROM_STATES`, no hooks will fire
        to_state_type: The state type of the proposed state before orchestration, if
            this state type is not contained in `TO_STATES`, no hooks will fire
    """

    FROM_STATES: Iterable = []
    TO_STATES: Iterable = []

    def __init__(
        self,
        context: OrchestrationContext,
        from_state_type: Optional[states.StateType],
        to_state_type: Optional[states.StateType],
    ):
        self.context = context
        self.from_state_type = from_state_type
        self.to_state_type = to_state_type
        self._invalid_on_entry = None

    async def __aenter__(self) -> OrchestrationContext:
        """
        Enter an async runtime context governed by this rule.

        The `with` statement will bind a governed `OrchestrationContext` to the target
        specified by the `as` clause. If the transition proposed by the
        `OrchestrationContext` is considered invalid on entry, entering this context
        will do nothing. Otherwise, `self.before_transition` will fire.
        """

        if await self.invalid():
            pass
        else:
            try:
                entry_context = self.context.entry_context()
                await self.before_transition(*entry_context)
                self.context.rule_signature.append(str(self.__class__))
            except Exception as before_transition_error:
                reason = (
                    f"Aborting orchestration due to error in {self.__class__!r}:"
                    f" !{before_transition_error!r}"
                )
                logger.exception(
                    f"Error running before-transition hook in rule {self.__class__!r}:"
                    f" !{before_transition_error!r}"
                )

                self.context.proposed_state = None
                self.context.response_status = SetStateStatus.ABORT
                self.context.response_details = StateAbortDetails(reason=reason)
                self.context.orchestration_error = before_transition_error

        return self.context

    async def __aexit__(
        self,
        exc_type: Optional[Type[BaseException]],
        exc_val: Optional[BaseException],
        exc_tb: Optional[TracebackType],
    ) -> None:
        """
        Exit the async runtime context governed by this rule.

        One of three outcomes can happen upon exiting this rule's context depending on
        the state of the rule. If the rule was found to be invalid on entry, nothing
        happens. If the rule was valid on entry and continues to be valid on exit,
        `self.after_transition` will fire. If the rule was valid on entry but invalid
        on exit, the rule will "fizzle" and `self.cleanup` will fire in order to revert
        any side-effects produced by `self.before_transition`.
        """

        exit_context = self.context.exit_context()
        if await self.invalid():
            pass
        elif await self.fizzled():
            await self.cleanup(*exit_context)
        else:
            await self.after_transition(*exit_context)
            self.context.finalization_signature.append(str(self.__class__))

    async def before_transition(
        self,
        initial_state: Optional[states.State],
        proposed_state: Optional[states.State],
        context: OrchestrationContext,
    ) -> None:
        """
        Implements a hook that can fire before a state is committed to the database.

        This hook may produce side-effects or mutate the proposed state of a
        transition using one of four methods: `self.reject_transition`,
        `self.delay_transition`, `self.abort_transition`, and `self.rename_state`.

        Note:
            As currently implemented, the `before_transition` hook is not
            perfectly isolated from mutating the transition. It is a standard instance
            method that has access to `self`, and therefore `self.context`. This should
            never be modified directly. Furthermore, `context.run` is an ORM model, and
            mutating the run can also cause unintended writes to the database.

        Args:
            initial_state: The initial state of a transition
            proposed_state: The proposed state of a transition
            context: A safe copy of the `OrchestrationContext`, with the exception of
                `context.run`, mutating this context will have no effect on the broader
                orchestration environment.

        Returns:
            None
        """

    async def after_transition(
        self,
        initial_state: Optional[states.State],
        validated_state: Optional[states.State],
        context: OrchestrationContext,
    ) -> None:
        """
        Implements a hook that can fire after a state is committed to the database.

        Args:
            initial_state: The initial state of a transition
            validated_state: The governed state that has been committed to the database
            context: A safe copy of the `OrchestrationContext`, with the exception of
                `context.run`, mutating this context will have no effect on the broader
                orchestration environment.

        Returns:
            None
        """

    async def cleanup(
        self,
        initial_state: Optional[states.State],
        validated_state: Optional[states.State],
        context: OrchestrationContext,
    ) -> None:
        """
        Implements a hook that can fire after a state is committed to the database.

        The intended use of this method is to revert side-effects produced by
        `self.before_transition` when the transition is found to be invalid on exit.
        This allows multiple rules to be gracefully run in sequence, without logic that
        keeps track of all other rules that might govern a transition.

        Args:
            initial_state: The initial state of a transition
            validated_state: The governed state that has been committed to the database
            context: A safe copy of the `OrchestrationContext`, with the exception of
                `context.run`, mutating this context will have no effect on the broader
                orchestration environment.

        Returns:
            None
        """

    async def invalid(self) -> bool:
        """
        Determines if a rule is invalid.

        Invalid rules do nothing and no hooks fire upon entering or exiting a governed
        context. Rules are invalid if the transition states types are not contained in
        `self.FROM_STATES` and `self.TO_STATES`, or if the context is proposing
        a transition that differs from the transition the rule was instantiated with.

        Returns:
            True if the rules in invalid, False otherwise.
        """
        # invalid and fizzled states are mutually exclusive,
        # `_invalid_on_entry` holds this statefulness
        if self.from_state_type not in self.FROM_STATES:
            self._invalid_on_entry = True
        if self.to_state_type not in self.TO_STATES:
            self._invalid_on_entry = True

        if self._invalid_on_entry is None:
            self._invalid_on_entry = await self.invalid_transition()
        return self._invalid_on_entry

    async def fizzled(self) -> bool:
        """
        Determines if a rule is fizzled and side-effects need to be reverted.

        Rules are fizzled if the transitions were valid on entry (thus firing
        `self.before_transition`) but are invalid upon exiting the governed context,
        most likely caused by another rule mutating the transition.

        Returns:
            True if the rule is fizzled, False otherwise.
        """

        if self._invalid_on_entry:
            return False
        return await self.invalid_transition()

    async def invalid_transition(self) -> bool:
        """
        Determines if the transition proposed by the `OrchestrationContext` is invalid.

        If the `OrchestrationContext` is attempting to manage a transition with this
        rule that differs from the transition the rule was instantiated with, the
        transition is considered to be invalid. Depending on the context, a rule with an
        invalid transition is either "invalid" or "fizzled".

        Returns:
            True if the transition is invalid, False otherwise.
        """

        initial_state_type = self.context.initial_state_type
        proposed_state_type = self.context.proposed_state_type
        return (self.from_state_type != initial_state_type) or (
            self.to_state_type != proposed_state_type
        )

    async def reject_transition(self, state: Optional[states.State], reason: str):
        """
        Rejects a proposed transition before the transition is validated.

        This method will reject a proposed transition, mutating the proposed state to
        the provided `state`. A reason for rejecting the transition is also passed on
        to the `OrchestrationContext`. Rules that reject the transition will not fizzle,
        despite the proposed state type changing.

        Args:
            state: The new proposed state. If `None`, the current run state will be
                returned in the result instead.
            reason: The reason for rejecting the transition
        """

        # don't run if the transition is already validated
        if self.context.validated_state:
            raise RuntimeError("The transition is already validated")

        # the current state will be used if a new one is not provided
        if state is None:
            if self.from_state_type is None:
                raise OrchestrationError(
                    "The current run has no state; this transition cannot be "
                    "rejected without providing a new state."
                )
            self.to_state_type = None
            self.context.proposed_state = None
        else:
            # a rule that mutates state should not fizzle itself
            self.to_state_type = state.type
            self.context.proposed_state = state

        self.context.response_status = SetStateStatus.REJECT
        self.context.response_details = StateRejectDetails(reason=reason)

    async def delay_transition(
        self,
        delay_seconds: int,
        reason: str,
    ):
        """
        Delays a proposed transition before the transition is validated.

        This method will delay a proposed transition, setting the proposed state to
        `None`, signaling to the `OrchestrationContext` that no state should be
        written to the database. The number of seconds a transition should be delayed is
        passed to the `OrchestrationContext`. A reason for delaying the transition is
        also provided. Rules that delay the transition will not fizzle, despite the
        proposed state type changing.

        Args:
            delay_seconds: The number of seconds the transition should be delayed
            reason: The reason for delaying the transition
        """

        # don't run if the transition is already validated
        if self.context.validated_state:
            raise RuntimeError("The transition is already validated")

        # a rule that mutates state should not fizzle itself
        self.to_state_type = None
        self.context.proposed_state = None
        self.context.response_status = SetStateStatus.WAIT
        self.context.response_details = StateWaitDetails(
            delay_seconds=delay_seconds, reason=reason
        )

    async def abort_transition(self, reason: str):
        """
        Aborts a proposed transition before the transition is validated.

        This method will abort a proposed transition, expecting no further action to
        occur for this run. The proposed state is set to `None`, signaling to the
        `OrchestrationContext` that no state should be written to the database. A
        reason for aborting the transition is also provided. Rules that abort the
        transition will not fizzle, despite the proposed state type changing.

        Args:
            reason: The reason for aborting the transition
        """

        # don't run if the transition is already validated
        if self.context.validated_state:
            raise RuntimeError("The transition is already validated")

        # a rule that mutates state should not fizzle itself
        self.to_state_type = None
        self.context.proposed_state = None
        self.context.response_status = SetStateStatus.ABORT
        self.context.response_details = StateAbortDetails(reason=reason)

    async def rename_state(self, state_name):
        """
        Sets the "name" attribute on a proposed state.

        The name of a state is an annotation intended to provide rich, human-readable
        context for how a run is progressing. This method only updates the name and not
        the canonical state TYPE, and will not fizzle or invalidate any other rules
        that might govern this state transition.
        """
        if self.context.proposed_state is not None:
            self.context.proposed_state.name = state_name

    async def update_context_parameters(self, key, value):
        """
        Updates the "parameters" dictionary attribute with the specified key-value pair.

        This mechanism streamlines the process of passing messages and information
        between orchestration rules if necessary and is simpler and more ephemeral than
        message-passing via the database or some other side-effect. This mechanism can
        be used to break up large rules for ease of testing or comprehension, but note
        that any rules coupled this way (or any other way) are no longer independent and
        the order in which they appear in the orchestration policy priority will matter.
        """

        self.context.parameters.update({key: value})

__aenter__() async

Enter an async runtime context governed by this rule.

The with statement will bind a governed OrchestrationContext to the target specified by the as clause. If the transition proposed by the OrchestrationContext is considered invalid on entry, entering this context will do nothing. Otherwise, self.before_transition will fire.

Source code in src/prefect/server/orchestration/rules.py

async def __aenter__(self) -> OrchestrationContext:
    """
    Enter an async runtime context governed by this rule.

    The `with` statement will bind a governed `OrchestrationContext` to the target
    specified by the `as` clause. If the transition proposed by the
    `OrchestrationContext` is considered invalid on entry, entering this context
    will do nothing. Otherwise, `self.before_transition` will fire.
    """

    if await self.invalid():
        pass
    else:
        try:
            entry_context = self.context.entry_context()
            await self.before_transition(*entry_context)
            self.context.rule_signature.append(str(self.__class__))
        except Exception as before_transition_error:
            reason = (
                f"Aborting orchestration due to error in {self.__class__!r}:"
                f" !{before_transition_error!r}"
            )
            logger.exception(
                f"Error running before-transition hook in rule {self.__class__!r}:"
                f" !{before_transition_error!r}"
            )

            self.context.proposed_state = None
            self.context.response_status = SetStateStatus.ABORT
            self.context.response_details = StateAbortDetails(reason=reason)
            self.context.orchestration_error = before_transition_error

    return self.context

__aexit__(exc_type, exc_val, exc_tb) async

Exit the async runtime context governed by this rule.

One of three outcomes can happen upon exiting this rule's context depending on the state of the rule. If the rule was found to be invalid on entry, nothing happens. If the rule was valid on entry and continues to be valid on exit, self.after_transition will fire. If the rule was valid on entry but invalid on exit, the rule will "fizzle" and self.cleanup will fire in order to revert any side-effects produced by self.before_transition.

Source code in src/prefect/server/orchestration/rules.py

async def __aexit__(
    self,
    exc_type: Optional[Type[BaseException]],
    exc_val: Optional[BaseException],
    exc_tb: Optional[TracebackType],
) -> None:
    """
    Exit the async runtime context governed by this rule.

    One of three outcomes can happen upon exiting this rule's context depending on
    the state of the rule. If the rule was found to be invalid on entry, nothing
    happens. If the rule was valid on entry and continues to be valid on exit,
    `self.after_transition` will fire. If the rule was valid on entry but invalid
    on exit, the rule will "fizzle" and `self.cleanup` will fire in order to revert
    any side-effects produced by `self.before_transition`.
    """

    exit_context = self.context.exit_context()
    if await self.invalid():
        pass
    elif await self.fizzled():
        await self.cleanup(*exit_context)
    else:
        await self.after_transition(*exit_context)
        self.context.finalization_signature.append(str(self.__class__))

abort_transition(reason) async

Aborts a proposed transition before the transition is validated.

This method will abort a proposed transition, expecting no further action to occur for this run. The proposed state is set to None, signaling to the OrchestrationContext that no state should be written to the database. A reason for aborting the transition is also provided. Rules that abort the transition will not fizzle, despite the proposed state type changing.

Parameters:

Name	Type	Description	Default
reason	str	The reason for aborting the transition	required

Source code in src/prefect/server/orchestration/rules.py

async def abort_transition(self, reason: str):
    """
    Aborts a proposed transition before the transition is validated.

    This method will abort a proposed transition, expecting no further action to
    occur for this run. The proposed state is set to `None`, signaling to the
    `OrchestrationContext` that no state should be written to the database. A
    reason for aborting the transition is also provided. Rules that abort the
    transition will not fizzle, despite the proposed state type changing.

    Args:
        reason: The reason for aborting the transition
    """

    # don't run if the transition is already validated
    if self.context.validated_state:
        raise RuntimeError("The transition is already validated")

    # a rule that mutates state should not fizzle itself
    self.to_state_type = None
    self.context.proposed_state = None
    self.context.response_status = SetStateStatus.ABORT
    self.context.response_details = StateAbortDetails(reason=reason)

after_transition(initial_state, validated_state, context) async

Implements a hook that can fire after a state is committed to the database.

Parameters:

Name	Type	Description	Default
initial_state	Optional[State]	The initial state of a transition	required
validated_state	Optional[State]	The governed state that has been committed to the database	required
context	OrchestrationContext	A safe copy of the OrchestrationContext, with the exception of context.run, mutating this context will have no effect on the broader orchestration environment.	required

Returns:

Type	Description
None	None

Source code in src/prefect/server/orchestration/rules.py

async def after_transition(
    self,
    initial_state: Optional[states.State],
    validated_state: Optional[states.State],
    context: OrchestrationContext,
) -> None:
    """
    Implements a hook that can fire after a state is committed to the database.

    Args:
        initial_state: The initial state of a transition
        validated_state: The governed state that has been committed to the database
        context: A safe copy of the `OrchestrationContext`, with the exception of
            `context.run`, mutating this context will have no effect on the broader
            orchestration environment.

    Returns:
        None
    """

before_transition(initial_state, proposed_state, context) async

Implements a hook that can fire before a state is committed to the database.

This hook may produce side-effects or mutate the proposed state of a transition using one of four methods: self.reject_transition, self.delay_transition, self.abort_transition, and self.rename_state.

Note

As currently implemented, the before_transition hook is not perfectly isolated from mutating the transition. It is a standard instance method that has access to self, and therefore self.context. This should never be modified directly. Furthermore, context.run is an ORM model, and mutating the run can also cause unintended writes to the database.

Parameters:

Name	Type	Description	Default
initial_state	Optional[State]	The initial state of a transition	required
proposed_state	Optional[State]	The proposed state of a transition	required
context	OrchestrationContext	A safe copy of the OrchestrationContext, with the exception of context.run, mutating this context will have no effect on the broader orchestration environment.	required

Returns:

Type	Description
None	None

Source code in src/prefect/server/orchestration/rules.py

async def before_transition(
    self,
    initial_state: Optional[states.State],
    proposed_state: Optional[states.State],
    context: OrchestrationContext,
) -> None:
    """
    Implements a hook that can fire before a state is committed to the database.

    This hook may produce side-effects or mutate the proposed state of a
    transition using one of four methods: `self.reject_transition`,
    `self.delay_transition`, `self.abort_transition`, and `self.rename_state`.

    Note:
        As currently implemented, the `before_transition` hook is not
        perfectly isolated from mutating the transition. It is a standard instance
        method that has access to `self`, and therefore `self.context`. This should
        never be modified directly. Furthermore, `context.run` is an ORM model, and
        mutating the run can also cause unintended writes to the database.

    Args:
        initial_state: The initial state of a transition
        proposed_state: The proposed state of a transition
        context: A safe copy of the `OrchestrationContext`, with the exception of
            `context.run`, mutating this context will have no effect on the broader
            orchestration environment.

    Returns:
        None
    """

cleanup(initial_state, validated_state, context) async

Implements a hook that can fire after a state is committed to the database.

The intended use of this method is to revert side-effects produced by self.before_transition when the transition is found to be invalid on exit. This allows multiple rules to be gracefully run in sequence, without logic that keeps track of all other rules that might govern a transition.

Parameters:

Name	Type	Description	Default
initial_state	Optional[State]	The initial state of a transition	required
validated_state	Optional[State]	The governed state that has been committed to the database	required
context	OrchestrationContext	A safe copy of the OrchestrationContext, with the exception of context.run, mutating this context will have no effect on the broader orchestration environment.	required

Returns:

Type	Description
None	None

Source code in src/prefect/server/orchestration/rules.py

async def cleanup(
    self,
    initial_state: Optional[states.State],
    validated_state: Optional[states.State],
    context: OrchestrationContext,
) -> None:
    """
    Implements a hook that can fire after a state is committed to the database.

    The intended use of this method is to revert side-effects produced by
    `self.before_transition` when the transition is found to be invalid on exit.
    This allows multiple rules to be gracefully run in sequence, without logic that
    keeps track of all other rules that might govern a transition.

    Args:
        initial_state: The initial state of a transition
        validated_state: The governed state that has been committed to the database
        context: A safe copy of the `OrchestrationContext`, with the exception of
            `context.run`, mutating this context will have no effect on the broader
            orchestration environment.

    Returns:
        None
    """

delay_transition(delay_seconds, reason) async

Delays a proposed transition before the transition is validated.

This method will delay a proposed transition, setting the proposed state to None, signaling to the OrchestrationContext that no state should be written to the database. The number of seconds a transition should be delayed is passed to the OrchestrationContext. A reason for delaying the transition is also provided. Rules that delay the transition will not fizzle, despite the proposed state type changing.

Parameters:

Name	Type	Description	Default
delay_seconds	int	The number of seconds the transition should be delayed	required
reason	str	The reason for delaying the transition	required

Source code in src/prefect/server/orchestration/rules.py

async def delay_transition(
    self,
    delay_seconds: int,
    reason: str,
):
    """
    Delays a proposed transition before the transition is validated.

    This method will delay a proposed transition, setting the proposed state to
    `None`, signaling to the `OrchestrationContext` that no state should be
    written to the database. The number of seconds a transition should be delayed is
    passed to the `OrchestrationContext`. A reason for delaying the transition is
    also provided. Rules that delay the transition will not fizzle, despite the
    proposed state type changing.

    Args:
        delay_seconds: The number of seconds the transition should be delayed
        reason: The reason for delaying the transition
    """

    # don't run if the transition is already validated
    if self.context.validated_state:
        raise RuntimeError("The transition is already validated")

    # a rule that mutates state should not fizzle itself
    self.to_state_type = None
    self.context.proposed_state = None
    self.context.response_status = SetStateStatus.WAIT
    self.context.response_details = StateWaitDetails(
        delay_seconds=delay_seconds, reason=reason
    )

fizzled() async

Determines if a rule is fizzled and side-effects need to be reverted.

Rules are fizzled if the transitions were valid on entry (thus firing self.before_transition) but are invalid upon exiting the governed context, most likely caused by another rule mutating the transition.

Returns:

Type	Description
bool	True if the rule is fizzled, False otherwise.

Source code in src/prefect/server/orchestration/rules.py

async def fizzled(self) -> bool:
    """
    Determines if a rule is fizzled and side-effects need to be reverted.

    Rules are fizzled if the transitions were valid on entry (thus firing
    `self.before_transition`) but are invalid upon exiting the governed context,
    most likely caused by another rule mutating the transition.

    Returns:
        True if the rule is fizzled, False otherwise.
    """

    if self._invalid_on_entry:
        return False
    return await self.invalid_transition()

invalid() async

Determines if a rule is invalid.

Invalid rules do nothing and no hooks fire upon entering or exiting a governed context. Rules are invalid if the transition states types are not contained in self.FROM_STATES and self.TO_STATES, or if the context is proposing a transition that differs from the transition the rule was instantiated with.

Returns:

Type	Description
bool	True if the rules in invalid, False otherwise.

Source code in src/prefect/server/orchestration/rules.py

async def invalid(self) -> bool:
    """
    Determines if a rule is invalid.

    Invalid rules do nothing and no hooks fire upon entering or exiting a governed
    context. Rules are invalid if the transition states types are not contained in
    `self.FROM_STATES` and `self.TO_STATES`, or if the context is proposing
    a transition that differs from the transition the rule was instantiated with.

    Returns:
        True if the rules in invalid, False otherwise.
    """
    # invalid and fizzled states are mutually exclusive,
    # `_invalid_on_entry` holds this statefulness
    if self.from_state_type not in self.FROM_STATES:
        self._invalid_on_entry = True
    if self.to_state_type not in self.TO_STATES:
        self._invalid_on_entry = True

    if self._invalid_on_entry is None:
        self._invalid_on_entry = await self.invalid_transition()
    return self._invalid_on_entry

invalid_transition() async

Determines if the transition proposed by the OrchestrationContext is invalid.

If the OrchestrationContext is attempting to manage a transition with this rule that differs from the transition the rule was instantiated with, the transition is considered to be invalid. Depending on the context, a rule with an invalid transition is either "invalid" or "fizzled".

Returns:

Type	Description
bool	True if the transition is invalid, False otherwise.

Source code in src/prefect/server/orchestration/rules.py

async def invalid_transition(self) -> bool:
    """
    Determines if the transition proposed by the `OrchestrationContext` is invalid.

    If the `OrchestrationContext` is attempting to manage a transition with this
    rule that differs from the transition the rule was instantiated with, the
    transition is considered to be invalid. Depending on the context, a rule with an
    invalid transition is either "invalid" or "fizzled".

    Returns:
        True if the transition is invalid, False otherwise.
    """

    initial_state_type = self.context.initial_state_type
    proposed_state_type = self.context.proposed_state_type
    return (self.from_state_type != initial_state_type) or (
        self.to_state_type != proposed_state_type
    )

reject_transition(state, reason) async

Rejects a proposed transition before the transition is validated.

This method will reject a proposed transition, mutating the proposed state to the provided state. A reason for rejecting the transition is also passed on to the OrchestrationContext. Rules that reject the transition will not fizzle, despite the proposed state type changing.

Parameters:

Name	Type	Description	Default
state	Optional[State]	The new proposed state. If None, the current run state will be returned in the result instead.	required
reason	str	The reason for rejecting the transition	required

Source code in src/prefect/server/orchestration/rules.py

async def reject_transition(self, state: Optional[states.State], reason: str):
    """
    Rejects a proposed transition before the transition is validated.

    This method will reject a proposed transition, mutating the proposed state to
    the provided `state`. A reason for rejecting the transition is also passed on
    to the `OrchestrationContext`. Rules that reject the transition will not fizzle,
    despite the proposed state type changing.

    Args:
        state: The new proposed state. If `None`, the current run state will be
            returned in the result instead.
        reason: The reason for rejecting the transition
    """

    # don't run if the transition is already validated
    if self.context.validated_state:
        raise RuntimeError("The transition is already validated")

    # the current state will be used if a new one is not provided
    if state is None:
        if self.from_state_type is None:
            raise OrchestrationError(
                "The current run has no state; this transition cannot be "
                "rejected without providing a new state."
            )
        self.to_state_type = None
        self.context.proposed_state = None
    else:
        # a rule that mutates state should not fizzle itself
        self.to_state_type = state.type
        self.context.proposed_state = state

    self.context.response_status = SetStateStatus.REJECT
    self.context.response_details = StateRejectDetails(reason=reason)

rename_state(state_name) async

Sets the "name" attribute on a proposed state.

The name of a state is an annotation intended to provide rich, human-readable context for how a run is progressing. This method only updates the name and not the canonical state TYPE, and will not fizzle or invalidate any other rules that might govern this state transition.

Source code in src/prefect/server/orchestration/rules.py

async def rename_state(self, state_name):
    """
    Sets the "name" attribute on a proposed state.

    The name of a state is an annotation intended to provide rich, human-readable
    context for how a run is progressing. This method only updates the name and not
    the canonical state TYPE, and will not fizzle or invalidate any other rules
    that might govern this state transition.
    """
    if self.context.proposed_state is not None:
        self.context.proposed_state.name = state_name

update_context_parameters(key, value) async

Updates the "parameters" dictionary attribute with the specified key-value pair.

This mechanism streamlines the process of passing messages and information between orchestration rules if necessary and is simpler and more ephemeral than message-passing via the database or some other side-effect. This mechanism can be used to break up large rules for ease of testing or comprehension, but note that any rules coupled this way (or any other way) are no longer independent and the order in which they appear in the orchestration policy priority will matter.

Source code in src/prefect/server/orchestration/rules.py

async def update_context_parameters(self, key, value):
    """
    Updates the "parameters" dictionary attribute with the specified key-value pair.

    This mechanism streamlines the process of passing messages and information
    between orchestration rules if necessary and is simpler and more ephemeral than
    message-passing via the database or some other side-effect. This mechanism can
    be used to break up large rules for ease of testing or comprehension, but note
    that any rules coupled this way (or any other way) are no longer independent and
    the order in which they appear in the orchestration policy priority will matter.
    """

    self.context.parameters.update({key: value})

BaseUniversalTransform

Bases: AbstractAsyncContextManager

An abstract base class used to implement privileged bookkeeping logic.

Warning

In almost all cases, use the BaseOrchestrationRule base class instead.

Beyond the orchestration rules implemented with the BaseOrchestrationRule ABC, Universal transforms are not stateful, and fire their before- and after- transition hooks on every state transition unless the proposed state is None, indicating that no state should be written to the database. Because there are no guardrails in place to prevent directly mutating state or other parts of the orchestration context, universal transforms should only be used with care.

Attributes:

Name	Type	Description
FROM_STATES	Iterable	for compatibility with BaseOrchestrationPolicy
TO_STATES	Iterable	for compatibility with BaseOrchestrationPolicy
context		the orchestration context
from_state_type		the state type a run is currently in
to_state_type		the intended proposed state type prior to any orchestration

Parameters:

Name	Type	Description	Default
context	OrchestrationContext	A FlowOrchestrationContext or TaskOrchestrationContext that is passed between transforms	required

Source code in src/prefect/server/orchestration/rules.py

class BaseUniversalTransform(contextlib.AbstractAsyncContextManager):
    """
    An abstract base class used to implement privileged bookkeeping logic.

    Warning:
        In almost all cases, use the `BaseOrchestrationRule` base class instead.

    Beyond the orchestration rules implemented with the `BaseOrchestrationRule` ABC,
    Universal transforms are not stateful, and fire their before- and after- transition
    hooks on every state transition unless the proposed state is `None`, indicating that
    no state should be written to the database. Because there are no guardrails in place
    to prevent directly mutating state or other parts of the orchestration context,
    universal transforms should only be used with care.

    Attributes:
        FROM_STATES: for compatibility with `BaseOrchestrationPolicy`
        TO_STATES: for compatibility with `BaseOrchestrationPolicy`
        context: the orchestration context
        from_state_type: the state type a run is currently in
        to_state_type: the intended proposed state type prior to any orchestration

    Args:
        context: A `FlowOrchestrationContext` or `TaskOrchestrationContext` that is
            passed between transforms
    """

    # `BaseUniversalTransform` will always fire on non-null transitions
    FROM_STATES: Iterable = ALL_ORCHESTRATION_STATES
    TO_STATES: Iterable = ALL_ORCHESTRATION_STATES

    def __init__(
        self,
        context: OrchestrationContext,
        from_state_type: Optional[states.StateType],
        to_state_type: Optional[states.StateType],
    ):
        self.context = context
        self.from_state_type = from_state_type
        self.to_state_type = to_state_type

    async def __aenter__(self):
        """
        Enter an async runtime context governed by this transform.

        The `with` statement will bind a governed `OrchestrationContext` to the target
        specified by the `as` clause. If the transition proposed by the
        `OrchestrationContext` has been nullified on entry and `context.proposed_state`
        is `None`, entering this context will do nothing. Otherwise
        `self.before_transition` will fire.
        """

        await self.before_transition(self.context)
        self.context.rule_signature.append(str(self.__class__))
        return self.context

    async def __aexit__(
        self,
        exc_type: Optional[Type[BaseException]],
        exc_val: Optional[BaseException],
        exc_tb: Optional[TracebackType],
    ) -> None:
        """
        Exit the async runtime context governed by this transform.

        If the transition has been nullified or errorred upon exiting this transforms's context,
        nothing happens. Otherwise, `self.after_transition` will fire on every non-null
        proposed state.
        """

        if not self.exception_in_transition():
            await self.after_transition(self.context)
            self.context.finalization_signature.append(str(self.__class__))

    async def before_transition(self, context) -> None:
        """
        Implements a hook that fires before a state is committed to the database.

        Args:
            context: the `OrchestrationContext` that contains transition details

        Returns:
            None
        """

    async def after_transition(self, context) -> None:
        """
        Implements a hook that can fire after a state is committed to the database.

        Args:
            context: the `OrchestrationContext` that contains transition details

        Returns:
            None
        """

    def nullified_transition(self) -> bool:
        """
        Determines if the transition has been nullified.

        Transitions are nullified if the proposed state is `None`, indicating that
        nothing should be written to the database.

        Returns:
            True if the transition is nullified, False otherwise.
        """

        return self.context.proposed_state is None

    def exception_in_transition(self) -> bool:
        """
        Determines if the transition has encountered an exception.

        Returns:
            True if the transition is encountered an exception, False otherwise.
        """

        return self.context.orchestration_error is not None

__aenter__() async

Enter an async runtime context governed by this transform.

The with statement will bind a governed OrchestrationContext to the target specified by the as clause. If the transition proposed by the OrchestrationContext has been nullified on entry and context.proposed_state is None, entering this context will do nothing. Otherwise self.before_transition will fire.

Source code in src/prefect/server/orchestration/rules.py

async def __aenter__(self):
    """
    Enter an async runtime context governed by this transform.

    The `with` statement will bind a governed `OrchestrationContext` to the target
    specified by the `as` clause. If the transition proposed by the
    `OrchestrationContext` has been nullified on entry and `context.proposed_state`
    is `None`, entering this context will do nothing. Otherwise
    `self.before_transition` will fire.
    """

    await self.before_transition(self.context)
    self.context.rule_signature.append(str(self.__class__))
    return self.context

__aexit__(exc_type, exc_val, exc_tb) async

Exit the async runtime context governed by this transform.

If the transition has been nullified or errorred upon exiting this transforms's context, nothing happens. Otherwise, self.after_transition will fire on every non-null proposed state.

Source code in src/prefect/server/orchestration/rules.py

async def __aexit__(
    self,
    exc_type: Optional[Type[BaseException]],
    exc_val: Optional[BaseException],
    exc_tb: Optional[TracebackType],
) -> None:
    """
    Exit the async runtime context governed by this transform.

    If the transition has been nullified or errorred upon exiting this transforms's context,
    nothing happens. Otherwise, `self.after_transition` will fire on every non-null
    proposed state.
    """

    if not self.exception_in_transition():
        await self.after_transition(self.context)
        self.context.finalization_signature.append(str(self.__class__))

after_transition(context) async

Implements a hook that can fire after a state is committed to the database.

Parameters:

Name	Type	Description	Default
context		the OrchestrationContext that contains transition details	required

Returns:

Type	Description
None	None

Source code in src/prefect/server/orchestration/rules.py

async def after_transition(self, context) -> None:
    """
    Implements a hook that can fire after a state is committed to the database.

    Args:
        context: the `OrchestrationContext` that contains transition details

    Returns:
        None
    """

before_transition(context) async

Implements a hook that fires before a state is committed to the database.

Parameters:

Name	Type	Description	Default
context		the OrchestrationContext that contains transition details	required

Returns:

Type	Description
None	None

Source code in src/prefect/server/orchestration/rules.py

async def before_transition(self, context) -> None:
    """
    Implements a hook that fires before a state is committed to the database.

    Args:
        context: the `OrchestrationContext` that contains transition details

    Returns:
        None
    """

exception_in_transition()

Determines if the transition has encountered an exception.

Returns:

Type	Description
bool	True if the transition is encountered an exception, False otherwise.

Source code in src/prefect/server/orchestration/rules.py

def exception_in_transition(self) -> bool:
    """
    Determines if the transition has encountered an exception.

    Returns:
        True if the transition is encountered an exception, False otherwise.
    """

    return self.context.orchestration_error is not None

nullified_transition()

Determines if the transition has been nullified.

Transitions are nullified if the proposed state is None, indicating that nothing should be written to the database.

Returns:

Type	Description
bool	True if the transition is nullified, False otherwise.

Source code in src/prefect/server/orchestration/rules.py

def nullified_transition(self) -> bool:
    """
    Determines if the transition has been nullified.

    Transitions are nullified if the proposed state is `None`, indicating that
    nothing should be written to the database.

    Returns:
        True if the transition is nullified, False otherwise.
    """

    return self.context.proposed_state is None

FlowOrchestrationContext

Bases: OrchestrationContext

A container for a flow run state transition, governed by orchestration rules.

When a flow- run attempts to change state, Prefect REST API has an opportunity to decide whether this transition can proceed. All the relevant information associated with the state transition is stored in an OrchestrationContext, which is subsequently governed by nested orchestration rules implemented using the BaseOrchestrationRule ABC.

FlowOrchestrationContext introduces the concept of a state being None in the context of an intended state transition. An initial state can be None if a run is is attempting to set a state for the first time. The proposed state might be None if a rule governing the transition determines that no state change should occur at all and nothing is written to the database.

Attributes:

Name	Type	Description
session		a SQLAlchemy database session
run	Any	the flow run attempting to change state
initial_state	Any	the initial state of the run
proposed_state	Any	the proposed state the run is transitioning into
validated_state	Any	a proposed state that has committed to the database
rule_signature	Any	a record of rules that have fired on entry into a managed context, currently only used for debugging purposes
finalization_signature	Any	a record of rules that have fired on exit from a managed context, currently only used for debugging purposes
response_status	Any	a SetStateStatus object used to build the API response
response_details	Any	a StateResponseDetails object use to build the API response

Parameters:

Name	Type	Description	Default
session		a SQLAlchemy database session	required
run		the flow run attempting to change state	required
initial_state		the initial state of a run	required
proposed_state		the proposed state a run is transitioning into	required

Source code in src/prefect/server/orchestration/rules.py

class FlowOrchestrationContext(OrchestrationContext):
    """
    A container for a flow run state transition, governed by orchestration rules.

    When a flow- run attempts to change state, Prefect REST API has an opportunity
    to decide whether this transition can proceed. All the relevant information
    associated with the state transition is stored in an `OrchestrationContext`,
    which is subsequently governed by nested orchestration rules implemented using
    the `BaseOrchestrationRule` ABC.

    `FlowOrchestrationContext` introduces the concept of a state being `None` in the
    context of an intended state transition. An initial state can be `None` if a run
    is is attempting to set a state for the first time. The proposed state might be
    `None` if a rule governing the transition determines that no state change
    should occur at all and nothing is written to the database.

    Attributes:
        session: a SQLAlchemy database session
        run: the flow run attempting to change state
        initial_state: the initial state of the run
        proposed_state: the proposed state the run is transitioning into
        validated_state: a proposed state that has committed to the database
        rule_signature: a record of rules that have fired on entry into a
            managed context, currently only used for debugging purposes
        finalization_signature: a record of rules that have fired on exit from a
            managed context, currently only used for debugging purposes
        response_status: a SetStateStatus object used to build the API response
        response_details:a StateResponseDetails object use to build the API response

    Args:
        session: a SQLAlchemy database session
        run: the flow run attempting to change state
        initial_state: the initial state of a run
        proposed_state: the proposed state a run is transitioning into
    """

    # run: db.FlowRun = ...
    run: Any = ...

    @inject_db
    async def validate_proposed_state(
        self,
        db: PrefectDBInterface,
    ):
        """
        Validates a proposed state by committing it to the database.

        After the `FlowOrchestrationContext` is governed by orchestration rules, the
        proposed state can be validated: the proposed state is added to the current
        SQLAlchemy session and is flushed. `self.validated_state` set to the flushed
        state. The state on the run is set to the validated state as well.

        If the proposed state is `None` when this method is called, no state will be
        written and `self.validated_state` will be set to the run's current state.

        Returns:
            None
        """
        # (circular import)
        from prefect.server.api.server import is_client_retryable_exception

        try:
            await self._validate_proposed_state()
            return
        except Exception as exc:
            logger.exception("Encountered error during state validation")
            self.proposed_state = None

            if is_client_retryable_exception(exc):
                # Do not capture retryable database exceptions, this exception will be
                # raised as a 503 in the API layer
                raise

            reason = f"Error validating state: {exc!r}"
            self.response_status = SetStateStatus.ABORT
            self.response_details = StateAbortDetails(reason=reason)

    @inject_db
    async def _validate_proposed_state(
        self,
        db: PrefectDBInterface,
    ):
        if self.proposed_state is None:
            validated_orm_state = self.run.state
            # We cannot access `self.run.state.data` directly for unknown reasons
            state_data = (
                (
                    await artifacts.read_artifact(
                        self.session, self.run.state.result_artifact_id
                    )
                ).data
                if self.run.state.result_artifact_id
                else None
            )
        else:
            state_payload = self.proposed_state.model_dump_for_orm()
            state_data = state_payload.pop("data", None)

            if state_data is not None and not (
                isinstance(state_data, dict) and state_data.get("type") == "unpersisted"
            ):
                state_result_artifact = core.Artifact.from_result(state_data)
                state_result_artifact.flow_run_id = self.run.id
                await artifacts.create_artifact(self.session, state_result_artifact)
                state_payload["result_artifact_id"] = state_result_artifact.id

            validated_orm_state = db.FlowRunState(
                flow_run_id=self.run.id,
                **state_payload,
            )

        self.session.add(validated_orm_state)
        self.run.set_state(validated_orm_state)

        await self.session.flush()
        if validated_orm_state:
            self.validated_state = states.State.from_orm_without_result(
                validated_orm_state, with_data=state_data
            )
        else:
            self.validated_state = None

    def safe_copy(self):
        """
        Creates a mostly-mutation-safe copy for use in orchestration rules.

        Orchestration rules govern state transitions using information stored in
        an `OrchestrationContext`. However, mutating objects stored on the context
        directly can have unintended side-effects. To guard against this,
        `self.safe_copy` can be used to pass information to orchestration rules
        without risking mutation.

        Note:
            `self.run` is an ORM model, and even when copied is unsafe to mutate

        Returns:
            A mutation-safe copy of `FlowOrchestrationContext`
        """

        return super().safe_copy()

    @property
    def run_settings(self) -> Dict:
        """Run-level settings used to orchestrate the state transition."""

        return self.run.empirical_policy

    async def task_run(self):
        return None

    async def flow_run(self):
        return self.run

run_settings: Dict property

Run-level settings used to orchestrate the state transition.

safe_copy()

Creates a mostly-mutation-safe copy for use in orchestration rules.

Orchestration rules govern state transitions using information stored in an OrchestrationContext. However, mutating objects stored on the context directly can have unintended side-effects. To guard against this, self.safe_copy can be used to pass information to orchestration rules without risking mutation.

Note

self.run is an ORM model, and even when copied is unsafe to mutate

Returns:

Type	Description
	A mutation-safe copy of FlowOrchestrationContext

Source code in src/prefect/server/orchestration/rules.py

def safe_copy(self):
    """
    Creates a mostly-mutation-safe copy for use in orchestration rules.

    Orchestration rules govern state transitions using information stored in
    an `OrchestrationContext`. However, mutating objects stored on the context
    directly can have unintended side-effects. To guard against this,
    `self.safe_copy` can be used to pass information to orchestration rules
    without risking mutation.

    Note:
        `self.run` is an ORM model, and even when copied is unsafe to mutate

    Returns:
        A mutation-safe copy of `FlowOrchestrationContext`
    """

    return super().safe_copy()

validate_proposed_state(db) async

Validates a proposed state by committing it to the database.

After the FlowOrchestrationContext is governed by orchestration rules, the proposed state can be validated: the proposed state is added to the current SQLAlchemy session and is flushed. self.validated_state set to the flushed state. The state on the run is set to the validated state as well.

If the proposed state is None when this method is called, no state will be written and self.validated_state will be set to the run's current state.

Returns:

Type	Description
	None

Source code in src/prefect/server/orchestration/rules.py

@inject_db
async def validate_proposed_state(
    self,
    db: PrefectDBInterface,
):
    """
    Validates a proposed state by committing it to the database.

    After the `FlowOrchestrationContext` is governed by orchestration rules, the
    proposed state can be validated: the proposed state is added to the current
    SQLAlchemy session and is flushed. `self.validated_state` set to the flushed
    state. The state on the run is set to the validated state as well.

    If the proposed state is `None` when this method is called, no state will be
    written and `self.validated_state` will be set to the run's current state.

    Returns:
        None
    """
    # (circular import)
    from prefect.server.api.server import is_client_retryable_exception

    try:
        await self._validate_proposed_state()
        return
    except Exception as exc:
        logger.exception("Encountered error during state validation")
        self.proposed_state = None

        if is_client_retryable_exception(exc):
            # Do not capture retryable database exceptions, this exception will be
            # raised as a 503 in the API layer
            raise

        reason = f"Error validating state: {exc!r}"
        self.response_status = SetStateStatus.ABORT
        self.response_details = StateAbortDetails(reason=reason)

OrchestrationContext

Bases: PrefectBaseModel

A container for a state transition, governed by orchestration rules.

Note

An OrchestrationContext should not be instantiated directly, instead use the flow- or task- specific subclasses, FlowOrchestrationContext and TaskOrchestrationContext.

When a flow- or task- run attempts to change state, Prefect REST API has an opportunity to decide whether this transition can proceed. All the relevant information associated with the state transition is stored in an OrchestrationContext, which is subsequently governed by nested orchestration rules implemented using the BaseOrchestrationRule ABC.

OrchestrationContext introduces the concept of a state being None in the context of an intended state transition. An initial state can be None if a run is is attempting to set a state for the first time. The proposed state might be None if a rule governing the transition determines that no state change should occur at all and nothing is written to the database.

Attributes:

Name	Type	Description
session	Optional[Union[Session, AsyncSession]]	a SQLAlchemy database session
initial_state	Optional[State]	the initial state of a run
proposed_state	Optional[State]	the proposed state a run is transitioning into
validated_state	Optional[State]	a proposed state that has committed to the database
rule_signature	List[str]	a record of rules that have fired on entry into a managed context, currently only used for debugging purposes
finalization_signature	List[str]	a record of rules that have fired on exit from a managed context, currently only used for debugging purposes
response_status	SetStateStatus	a SetStateStatus object used to build the API response
response_details	StateResponseDetails	a StateResponseDetails object use to build the API response

Parameters:

Name	Type	Description	Default
session		a SQLAlchemy database session	required
initial_state		the initial state of a run	required
proposed_state		the proposed state a run is transitioning into	required

Source code in src/prefect/server/orchestration/rules.py

class OrchestrationContext(PrefectBaseModel):
    """
    A container for a state transition, governed by orchestration rules.

    Note:
        An `OrchestrationContext` should not be instantiated directly, instead
        use the flow- or task- specific subclasses, `FlowOrchestrationContext` and
        `TaskOrchestrationContext`.

    When a flow- or task- run attempts to change state, Prefect REST API has an opportunity
    to decide whether this transition can proceed. All the relevant information
    associated with the state transition is stored in an `OrchestrationContext`,
    which is subsequently governed by nested orchestration rules implemented using
    the `BaseOrchestrationRule` ABC.

    `OrchestrationContext` introduces the concept of a state being `None` in the
    context of an intended state transition. An initial state can be `None` if a run
    is is attempting to set a state for the first time. The proposed state might be
    `None` if a rule governing the transition determines that no state change
    should occur at all and nothing is written to the database.

    Attributes:
        session: a SQLAlchemy database session
        initial_state: the initial state of a run
        proposed_state: the proposed state a run is transitioning into
        validated_state: a proposed state that has committed to the database
        rule_signature: a record of rules that have fired on entry into a
            managed context, currently only used for debugging purposes
        finalization_signature: a record of rules that have fired on exit from a
            managed context, currently only used for debugging purposes
        response_status: a SetStateStatus object used to build the API response
        response_details:a StateResponseDetails object use to build the API response

    Args:
        session: a SQLAlchemy database session
        initial_state: the initial state of a run
        proposed_state: the proposed state a run is transitioning into
    """

    model_config = ConfigDict(arbitrary_types_allowed=True)

    session: Optional[Union[sa.orm.Session, AsyncSession]] = ...
    initial_state: Optional[states.State] = ...
    proposed_state: Optional[states.State] = ...
    validated_state: Optional[states.State] = Field(default=None)
    rule_signature: List[str] = Field(default_factory=list)
    finalization_signature: List[str] = Field(default_factory=list)
    response_status: SetStateStatus = Field(default=SetStateStatus.ACCEPT)
    response_details: StateResponseDetails = Field(default_factory=StateAcceptDetails)
    orchestration_error: Optional[Exception] = Field(default=None)
    parameters: Dict[Any, Any] = Field(default_factory=dict)

    @property
    def initial_state_type(self) -> Optional[states.StateType]:
        """The state type of `self.initial_state` if it exists."""

        return self.initial_state.type if self.initial_state else None

    @property
    def proposed_state_type(self) -> Optional[states.StateType]:
        """The state type of `self.proposed_state` if it exists."""

        return self.proposed_state.type if self.proposed_state else None

    @property
    def validated_state_type(self) -> Optional[states.StateType]:
        """The state type of `self.validated_state` if it exists."""
        return self.validated_state.type if self.validated_state else None

    def safe_copy(self):
        """
        Creates a mostly-mutation-safe copy for use in orchestration rules.

        Orchestration rules govern state transitions using information stored in
        an `OrchestrationContext`. However, mutating objects stored on the context
        directly can have unintended side-effects. To guard against this,
        `self.safe_copy` can be used to pass information to orchestration rules
        without risking mutation.

        Returns:
            A mutation-safe copy of the `OrchestrationContext`
        """

        safe_copy = self.model_copy()

        safe_copy.initial_state = (
            self.initial_state.model_copy() if self.initial_state else None
        )
        safe_copy.proposed_state = (
            self.proposed_state.model_copy() if self.proposed_state else None
        )
        safe_copy.validated_state = (
            self.validated_state.model_copy() if self.validated_state else None
        )
        safe_copy.parameters = self.parameters.copy()
        return safe_copy

    def entry_context(self):
        """
        A convenience method that generates input parameters for orchestration rules.

        An `OrchestrationContext` defines a state transition that is managed by
        orchestration rules which can fire hooks before a transition has been committed
        to the database. These hooks have a consistent interface which can be generated
        with this method.
        """

        safe_context = self.safe_copy()
        return safe_context.initial_state, safe_context.proposed_state, safe_context

    def exit_context(self):
        """
        A convenience method that generates input parameters for orchestration rules.

        An `OrchestrationContext` defines a state transition that is managed by
        orchestration rules which can fire hooks after a transition has been committed
        to the database. These hooks have a consistent interface which can be generated
        with this method.
        """

        safe_context = self.safe_copy()
        return safe_context.initial_state, safe_context.validated_state, safe_context

initial_state_type: Optional[states.StateType] property

The state type of self.initial_state if it exists.

proposed_state_type: Optional[states.StateType] property

The state type of self.proposed_state if it exists.

validated_state_type: Optional[states.StateType] property

The state type of self.validated_state if it exists.

entry_context()

A convenience method that generates input parameters for orchestration rules.

An OrchestrationContext defines a state transition that is managed by orchestration rules which can fire hooks before a transition has been committed to the database. These hooks have a consistent interface which can be generated with this method.

Source code in src/prefect/server/orchestration/rules.py

def entry_context(self):
    """
    A convenience method that generates input parameters for orchestration rules.

    An `OrchestrationContext` defines a state transition that is managed by
    orchestration rules which can fire hooks before a transition has been committed
    to the database. These hooks have a consistent interface which can be generated
    with this method.
    """

    safe_context = self.safe_copy()
    return safe_context.initial_state, safe_context.proposed_state, safe_context

exit_context()

A convenience method that generates input parameters for orchestration rules.

An OrchestrationContext defines a state transition that is managed by orchestration rules which can fire hooks after a transition has been committed to the database. These hooks have a consistent interface which can be generated with this method.

Source code in src/prefect/server/orchestration/rules.py

def exit_context(self):
    """
    A convenience method that generates input parameters for orchestration rules.

    An `OrchestrationContext` defines a state transition that is managed by
    orchestration rules which can fire hooks after a transition has been committed
    to the database. These hooks have a consistent interface which can be generated
    with this method.
    """

    safe_context = self.safe_copy()
    return safe_context.initial_state, safe_context.validated_state, safe_context

safe_copy()

Creates a mostly-mutation-safe copy for use in orchestration rules.

Orchestration rules govern state transitions using information stored in an OrchestrationContext. However, mutating objects stored on the context directly can have unintended side-effects. To guard against this, self.safe_copy can be used to pass information to orchestration rules without risking mutation.

Returns:

Type	Description
	A mutation-safe copy of the OrchestrationContext

Source code in src/prefect/server/orchestration/rules.py

def safe_copy(self):
    """
    Creates a mostly-mutation-safe copy for use in orchestration rules.

    Orchestration rules govern state transitions using information stored in
    an `OrchestrationContext`. However, mutating objects stored on the context
    directly can have unintended side-effects. To guard against this,
    `self.safe_copy` can be used to pass information to orchestration rules
    without risking mutation.

    Returns:
        A mutation-safe copy of the `OrchestrationContext`
    """

    safe_copy = self.model_copy()

    safe_copy.initial_state = (
        self.initial_state.model_copy() if self.initial_state else None
    )
    safe_copy.proposed_state = (
        self.proposed_state.model_copy() if self.proposed_state else None
    )
    safe_copy.validated_state = (
        self.validated_state.model_copy() if self.validated_state else None
    )
    safe_copy.parameters = self.parameters.copy()
    return safe_copy

TaskOrchestrationContext

Bases: OrchestrationContext

A container for a task run state transition, governed by orchestration rules.

When a task- run attempts to change state, Prefect REST API has an opportunity to decide whether this transition can proceed. All the relevant information associated with the state transition is stored in an OrchestrationContext, which is subsequently governed by nested orchestration rules implemented using the BaseOrchestrationRule ABC.

TaskOrchestrationContext introduces the concept of a state being None in the context of an intended state transition. An initial state can be None if a run is is attempting to set a state for the first time. The proposed state might be None if a rule governing the transition determines that no state change should occur at all and nothing is written to the database.

Attributes:

Name	Type	Description
session		a SQLAlchemy database session
run	Any	the task run attempting to change state
initial_state	Any	the initial state of the run
proposed_state	Any	the proposed state the run is transitioning into
validated_state	Any	a proposed state that has committed to the database
rule_signature	Any	a record of rules that have fired on entry into a managed context, currently only used for debugging purposes
finalization_signature	Any	a record of rules that have fired on exit from a managed context, currently only used for debugging purposes
response_status	Any	a SetStateStatus object used to build the API response
response_details	Any	a StateResponseDetails object use to build the API response

Parameters:

Name	Type	Description	Default
session		a SQLAlchemy database session	required
run		the task run attempting to change state	required
initial_state		the initial state of a run	required
proposed_state		the proposed state a run is transitioning into	required

Source code in src/prefect/server/orchestration/rules.py

class TaskOrchestrationContext(OrchestrationContext):
    """
    A container for a task run state transition, governed by orchestration rules.

    When a task- run attempts to change state, Prefect REST API has an opportunity
    to decide whether this transition can proceed. All the relevant information
    associated with the state transition is stored in an `OrchestrationContext`,
    which is subsequently governed by nested orchestration rules implemented using
    the `BaseOrchestrationRule` ABC.

    `TaskOrchestrationContext` introduces the concept of a state being `None` in the
    context of an intended state transition. An initial state can be `None` if a run
    is is attempting to set a state for the first time. The proposed state might be
    `None` if a rule governing the transition determines that no state change
    should occur at all and nothing is written to the database.

    Attributes:
        session: a SQLAlchemy database session
        run: the task run attempting to change state
        initial_state: the initial state of the run
        proposed_state: the proposed state the run is transitioning into
        validated_state: a proposed state that has committed to the database
        rule_signature: a record of rules that have fired on entry into a
            managed context, currently only used for debugging purposes
        finalization_signature: a record of rules that have fired on exit from a
            managed context, currently only used for debugging purposes
        response_status: a SetStateStatus object used to build the API response
        response_details:a StateResponseDetails object use to build the API response

    Args:
        session: a SQLAlchemy database session
        run: the task run attempting to change state
        initial_state: the initial state of a run
        proposed_state: the proposed state a run is transitioning into
    """

    # run: db.TaskRun = ...
    run: Any = ...

    @inject_db
    async def validate_proposed_state(
        self,
        db: PrefectDBInterface,
    ):
        """
        Validates a proposed state by committing it to the database.

        After the `TaskOrchestrationContext` is governed by orchestration rules, the
        proposed state can be validated: the proposed state is added to the current
        SQLAlchemy session and is flushed. `self.validated_state` set to the flushed
        state. The state on the run is set to the validated state as well.

        If the proposed state is `None` when this method is called, no state will be
        written and `self.validated_state` will be set to the run's current state.

        Returns:
            None
        """
        # (circular import)
        from prefect.server.api.server import is_client_retryable_exception

        try:
            await self._validate_proposed_state()
            return
        except Exception as exc:
            logger.exception("Encountered error during state validation")
            self.proposed_state = None

            if is_client_retryable_exception(exc):
                # Do not capture retryable database exceptions, this exception will be
                # raised as a 503 in the API layer
                raise

            reason = f"Error validating state: {exc!r}"
            self.response_status = SetStateStatus.ABORT
            self.response_details = StateAbortDetails(reason=reason)

    @inject_db
    async def _validate_proposed_state(
        self,
        db: PrefectDBInterface,
    ):
        if self.proposed_state is None:
            validated_orm_state = self.run.state
            # We cannot access `self.run.state.data` directly for unknown reasons
            state_data = (
                (
                    await artifacts.read_artifact(
                        self.session, self.run.state.result_artifact_id
                    )
                ).data
                if self.run.state.result_artifact_id
                else None
            )
        else:
            state_payload = self.proposed_state.model_dump_for_orm()
            state_data = state_payload.pop("data", None)

            if state_data is not None and not (
                isinstance(state_data, dict) and state_data.get("type") == "unpersisted"
            ):
                state_result_artifact = core.Artifact.from_result(state_data)
                state_result_artifact.task_run_id = self.run.id

                if self.run.flow_run_id is not None:
                    flow_run = await self.flow_run()
                    state_result_artifact.flow_run_id = flow_run.id

                await artifacts.create_artifact(self.session, state_result_artifact)
                state_payload["result_artifact_id"] = state_result_artifact.id

            validated_orm_state = db.TaskRunState(
                task_run_id=self.run.id,
                **state_payload,
            )

        self.session.add(validated_orm_state)
        self.run.set_state(validated_orm_state)

        await self.session.flush()
        if validated_orm_state:
            self.validated_state = states.State.from_orm_without_result(
                validated_orm_state, with_data=state_data
            )
        else:
            self.validated_state = None

    def safe_copy(self):
        """
        Creates a mostly-mutation-safe copy for use in orchestration rules.

        Orchestration rules govern state transitions using information stored in
        an `OrchestrationContext`. However, mutating objects stored on the context
        directly can have unintended side-effects. To guard against this,
        `self.safe_copy` can be used to pass information to orchestration rules
        without risking mutation.

        Note:
            `self.run` is an ORM model, and even when copied is unsafe to mutate

        Returns:
            A mutation-safe copy of `TaskOrchestrationContext`
        """

        return super().safe_copy()

    @property
    def run_settings(self) -> Dict:
        """Run-level settings used to orchestrate the state transition."""

        return self.run.empirical_policy

    async def task_run(self):
        return self.run

    async def flow_run(self):
        return await flow_runs.read_flow_run(
            session=self.session,
            flow_run_id=self.run.flow_run_id,
        )

run_settings: Dict property

Run-level settings used to orchestrate the state transition.

safe_copy()

Creates a mostly-mutation-safe copy for use in orchestration rules.

Orchestration rules govern state transitions using information stored in an OrchestrationContext. However, mutating objects stored on the context directly can have unintended side-effects. To guard against this, self.safe_copy can be used to pass information to orchestration rules without risking mutation.

Note

self.run is an ORM model, and even when copied is unsafe to mutate

Returns:

Type	Description
	A mutation-safe copy of TaskOrchestrationContext

Source code in src/prefect/server/orchestration/rules.py

def safe_copy(self):
    """
    Creates a mostly-mutation-safe copy for use in orchestration rules.

    Orchestration rules govern state transitions using information stored in
    an `OrchestrationContext`. However, mutating objects stored on the context
    directly can have unintended side-effects. To guard against this,
    `self.safe_copy` can be used to pass information to orchestration rules
    without risking mutation.

    Note:
        `self.run` is an ORM model, and even when copied is unsafe to mutate

    Returns:
        A mutation-safe copy of `TaskOrchestrationContext`
    """

    return super().safe_copy()

validate_proposed_state(db) async

Validates a proposed state by committing it to the database.

After the TaskOrchestrationContext is governed by orchestration rules, the proposed state can be validated: the proposed state is added to the current SQLAlchemy session and is flushed. self.validated_state set to the flushed state. The state on the run is set to the validated state as well.

If the proposed state is None when this method is called, no state will be written and self.validated_state will be set to the run's current state.

Returns:

Type	Description
	None

Source code in src/prefect/server/orchestration/rules.py

@inject_db
async def validate_proposed_state(
    self,
    db: PrefectDBInterface,
):
    """
    Validates a proposed state by committing it to the database.

    After the `TaskOrchestrationContext` is governed by orchestration rules, the
    proposed state can be validated: the proposed state is added to the current
    SQLAlchemy session and is flushed. `self.validated_state` set to the flushed
    state. The state on the run is set to the validated state as well.

    If the proposed state is `None` when this method is called, no state will be
    written and `self.validated_state` will be set to the run's current state.

    Returns:
        None
    """
    # (circular import)
    from prefect.server.api.server import is_client_retryable_exception

    try:
        await self._validate_proposed_state()
        return
    except Exception as exc:
        logger.exception("Encountered error during state validation")
        self.proposed_state = None

        if is_client_retryable_exception(exc):
            # Do not capture retryable database exceptions, this exception will be
            # raised as a 503 in the API layer
            raise

        reason = f"Error validating state: {exc!r}"
        self.response_status = SetStateStatus.ABORT
        self.response_details = StateAbortDetails(reason=reason)