Creating Your First Workflow

A workflow is a collection of tasks executed together by the engine. Each task represents a unit of work, and workflows allow these tasks to be executed in a controlled and predictable order.

In this guide, you will create a simple workflow with two tasks to understand how task execution and shared execution context work.

Example Scenario

We will create:

• One task that increments a shared counter

• Another task that prints the updated counter value

This example demonstrates task registration order and shared execution context using a mutable object.

from pravaha.core.task import Task
from pravaha.core.executor import TaskExecutor

# Shared execution context
counter = {"value": 0}

@Task(name="increment")
def increment():
    counter["value"] += 1

@Task(name="print_counter_value")
def print_counter_value():
    print("Counter value is:", counter["value"])

# Execute all registered tasks
TaskExecutor.execute()

Expected Output

Counter value is: 1

Execution Order

Tasks are registered in the order they are defined in the source file. By default, the executor runs tasks synchronously in registration order unless dependencies or execution rules are specified.

In this example:

1. increment is registered first and executed first

2. print_counter_value is registered next and executed afterward

Note

This execution order behavior applies only when no explicit dependencies are defined. Dependency-based execution is covered in the next section.