Rocketry Documentation

Rocketry is a modern, statement-based scheduling framework for Python. It is designed to be simple to set up while remaining robust and extensible for large-scale applications or autonomous engines.

Key capabilities include:

• Powerful Scheduling: Supports various strategies including cron and custom scheduling statements.
• Concurrency Support: Handles tasks using async, threading, or multiprocessing.
• Task Management: Supports parametrization and task pipelining.
• Extensibility: Features a modifiable runtime session and allows for custom scheduling logic.

Rocketry uses key-value pairs called parameters to pass values to tasks. Parameters can be static instances or dynamic arguments. There is a specific hierarchy of precedence; if the same key exists at multiple levels, the higher level overrides the lower one:

1. Batch parameters (Highest precedence)
2. *Task parameters
3. Function parameters
4. Session parameters (Lowest precedence)

Rocketry's ecosystem is composed of four primary layers that manage the lifecycle of task scheduling and execution:

• Rocketry: The top-level interface used to interact with the system.
• Session: The medium-level interface that manages configuration.
• Scheduler: Manages the overall flow and timing of the system.
• Task: Defines and handles the actual execution logic of a unit of work.

Rocketry's logging system extends the standard Python logging library using Red Bird. It uses a specific logger named rocketry.task to record task actions. To ensure visibility into task lifecycle, this logger should not filter out the INFO level.

Task actions logged include:

• run: Task starts.
• success: Task finishes without exceptions.
• fail: Task finishes with an error.
• terminate: Task is terminated before completion.
• inaction: Task did nothing (requires raising a special exception).
• crash: Task previously crashed silently and is no longer running.

Rocketry uses several specialized subcomponents to refine task behavior:

• Condition: A logic statement that evaluates to true or false based on time or system state, used to determine if a task should run.
• TimePeriod: An abstraction used to define specific time intervals or periods (e.g., 'today', a specific time of day, or a specific week).
• Parameters: Key-value pairs passed into tasks.
• Arguments: The specific values associated with the keys in a Parameters object.

Rocketry's scheduling system uses conditions that evaluate to either True or False. A task is triggered, terminated, or the scheduler is shut down based on these boolean evaluations.

Conditions can be used in three contexts:

1. Start condition: The task runs if the condition is True.
2. End condition: A running task terminates if the condition is True.
3. Shut condition: The scheduler shuts down if the condition is True (useful for testing).

You can combine conditions using logical operations: AND, OR, and NOT, and nest them using parentheses.