3. Defining workflows

Workflows are defined in code, using Python functions, which are decorated to indicate their operation type and configuration. A function decorated as a workflow is the entry point for a run, and a function decorated as a task is an operation to be executed within the run. Workflows can call tasks, tasks can call other tasks, and tasks can also call workflows (which will submit a separate run).

The decorators are intended to be unimposing so that functions can be executed outside of Coflux.

An example

Here's a simple example:

hello.py

import coflux as cf

@cf.task()
def build_greeting(name: str):
    return f"Hello, {name}"

@cf.workflow()
def print_greeting(name: str):
    print(build_greeting(name))

This defines a print_greeting workflow, which takes a name as an argument. When run, it calls the build_greeting task, passing through the name argument. Once it has the result from the task, the result gets printed.

Workflows are defined in repositories. Typically these are Python modules, but they can alternatively be loaded from a Python script, which this guide will demonstrate.

Put the workflow above into hello.py.

Before coming back to more advanced features, let's see how to get this workflow running...