agents.research.perplexity.pro_search.tasks.models¶

Pydantic models for recursive conditional planning with tree-based task decomposition. Supports dynamic planning, parallel execution, and adaptive replanning.

Classes¶

`ExecutionPlan`	Execution plan for a set of tasks.
`PlanningState`	State for recursive planning workflow.
`PlanningStrategy`	Strategy configuration for the planning process.
`ReplanningAnalysis`	Analysis for replanning decision.
`TaskDecomposition`	Result of decomposing a high-level task.
`TaskDependency`	Dependency relationship between tasks.
`TaskMetadata`	Metadata for task tracking and optimization.
`TaskNode`	Individual task node in the planning tree.
`TaskPriority`	Priority levels for tasks.
`TaskResource`	Resource requirements for a task.
`TaskStatus`	Status of a task in the planning tree.

Module Contents¶

class agents.research.perplexity.pro_search.tasks.models.ExecutionPlan(/, **data)¶

Bases: pydantic.BaseModel

Execution plan for a set of tasks.

Create a new model by parsing and validating input data from keyword arguments.

Raises [ValidationError][pydantic_core.ValidationError] if the input data cannot be validated to form a valid model.

self is explicitly positional-only to allow self as a field name.

Parameters:: data (Any)

calculate_resource_requirements()¶

Calculate total resource requirements.

Return type:: ExecutionPlan

property can_parallelize: bool¶

Check if any parallelization is possible.

Return type:: bool

class agents.research.perplexity.pro_search.tasks.models.PlanningState(/, **data)¶

Bases: pydantic.BaseModel

State for recursive planning workflow.

Create a new model by parsing and validating input data from keyword arguments.

Raises [ValidationError][pydantic_core.ValidationError] if the input data cannot be validated to form a valid model.

self is explicitly positional-only to allow self as a field name.

Parameters:: data (Any)

add_task(task, parent_id=None)¶

Add a task to the tree.

Parameters:

task (TaskNode)
parent_id (str | None)

Return type:

None

update_task_status(task_id, status, result=None)¶

Update task status and handle state transitions.

Parameters:

task_id (str)
status (TaskStatus)
result (dict[str, Any] | None)

Return type:

None

property completion_percentage: float¶

Calculate completion percentage.

Return type:: float

property critical_path: list[str]¶

Get current critical path.

Return type:: list[str]

property executable_tasks: list[TaskNode]¶

Get tasks ready for execution.

Return type:: list[TaskNode]

property is_complete: bool¶

Check if planning goal is achieved.

Return type:: bool

property needs_replanning: bool¶

Check if replanning is needed.

Return type:: bool

class agents.research.perplexity.pro_search.tasks.models.PlanningStrategy(/, **data)¶

Bases: pydantic.BaseModel

Strategy configuration for the planning process.

Create a new model by parsing and validating input data from keyword arguments.

Raises [ValidationError][pydantic_core.ValidationError] if the input data cannot be validated to form a valid model.

self is explicitly positional-only to allow self as a field name.

Parameters:: data (Any)

validate_strategy_coherence()¶

Ensure strategy settings are coherent.

Return type:: PlanningStrategy

class agents.research.perplexity.pro_search.tasks.models.ReplanningAnalysis(/, **data)¶

Bases: pydantic.BaseModel

Analysis for replanning decision.

Create a new model by parsing and validating input data from keyword arguments.

Raises [ValidationError][pydantic_core.ValidationError] if the input data cannot be validated to form a valid model.

self is explicitly positional-only to allow self as a field name.

Parameters:: data (Any)

class agents.research.perplexity.pro_search.tasks.models.TaskDecomposition(/, **data)¶

Bases: pydantic.BaseModel

Result of decomposing a high-level task.

Create a new model by parsing and validating input data from keyword arguments.

Raises [ValidationError][pydantic_core.ValidationError] if the input data cannot be validated to form a valid model.

self is explicitly positional-only to allow self as a field name.

Parameters:: data (Any)

calculate_critical_path()¶

Calculate critical path if not provided.

Return type:: TaskDecomposition

property parallelizable_groups: list[list[str]]¶

Identify groups of tasks that can run in parallel.

Return type:: list[list[str]]

class agents.research.perplexity.pro_search.tasks.models.TaskDependency(/, **data)¶

Bases: pydantic.BaseModel

Dependency relationship between tasks.

Create a new model by parsing and validating input data from keyword arguments.

Raises [ValidationError][pydantic_core.ValidationError] if the input data cannot be validated to form a valid model.

self is explicitly positional-only to allow self as a field name.

Parameters:: data (Any)

class agents.research.perplexity.pro_search.tasks.models.TaskMetadata(/, **data)¶

Bases: pydantic.BaseModel

Metadata for task tracking and optimization.

Create a new model by parsing and validating input data from keyword arguments.

Raises [ValidationError][pydantic_core.ValidationError] if the input data cannot be validated to form a valid model.

self is explicitly positional-only to allow self as a field name.

Parameters:: data (Any)

property can_retry: bool¶

Check if task can be retried.

Return type:: bool

property efficiency_ratio: float | None¶

Calculate efficiency ratio (estimated vs actual).

Return type:: float | None

class agents.research.perplexity.pro_search.tasks.models.TaskNode(/, **data)¶

Bases: pydantic.BaseModel

Individual task node in the planning tree.

Create a new model by parsing and validating input data from keyword arguments.

Raises [ValidationError][pydantic_core.ValidationError] if the input data cannot be validated to form a valid model.

self is explicitly positional-only to allow self as a field name.

Parameters:: data (Any)

can_start(completed_tasks)¶

Check if all dependencies are satisfied.

Parameters:: completed_tasks (set[str])
Return type:: bool

classmethod validate_children_for_type(v, info)¶

Validate children based on task type.

Return type:: Any

property is_complete: bool¶

Check if task is complete.

Return type:: bool

property is_executable: bool¶

Check if task can be executed.

Return type:: bool

property is_leaf: bool¶

Check if this is a leaf node.

Return type:: bool

class agents.research.perplexity.pro_search.tasks.models.TaskPriority¶

Bases: str, enum.Enum

Priority levels for tasks.

Initialize self. See help(type(self)) for accurate signature.

class agents.research.perplexity.pro_search.tasks.models.TaskResource(/, **data)¶

Bases: pydantic.BaseModel

Resource requirements for a task.

Create a new model by parsing and validating input data from keyword arguments.

Raises [ValidationError][pydantic_core.ValidationError] if the input data cannot be validated to form a valid model.

self is explicitly positional-only to allow self as a field name.

Parameters:: data (Any)

property resource_key: str¶

Unique key for this resource requirement.

Return type:: str

class agents.research.perplexity.pro_search.tasks.models.TaskStatus¶

Bases: str, enum.Enum

Status of a task in the planning tree.

Initialize self. See help(type(self)) for accurate signature.