Defining Celery tasks in Flask - Step-by-Step Execution

Concept Flow - Defining Celery tasks

Start Flask app

↓

Import Celery

↓

Create Celery instance

↓

Define task function with @app.task

↓

Call task asynchronously

↓

Worker picks task

↓

Task runs in background

↓

Result stored or returned

This flow shows how a Celery task is defined and executed asynchronously in a Flask app.

Execution Sample

Flask

from celery import Celery
app = Celery('tasks', broker='redis://localhost', backend='redis://localhost')

@app.task
def add(x, y):
    return x + y

result = add.delay(4, 6)

Defines a Celery task 'add' and calls it asynchronously with arguments 4 and 6.

Execution Table

Step	Action	Evaluation	Result
1	Import Celery	Celery class imported	Celery available
2	Create Celery instance	app = Celery('tasks', broker='redis://localhost', backend='redis://localhost')	Celery app ready
3	Define task with @app.task	Function 'add' wrapped as task	Task 'add' registered
4	Call add.delay(4, 6)	Task call queued asynchronously	Async task message sent
5	Worker picks task	Worker receives task message	Task ready to run
6	Task runs: add(4, 6)	Function executes	Returns 10
7	Result stored	Result saved in backend	Result 10 available
8	Main program continues	No waiting for task	Program not blocked

💡 Task completes after worker runs function and stores result asynchronously.

Variable Tracker

Variable	Start	After Step 4	After Step 6	Final
app	None	Celery instance	Celery instance	Celery instance
add	Function	Task object	Task object	Task object
result	None	AsyncResult object (pending)	AsyncResult object (ready)	AsyncResult object (ready)

Key Moments - 3 Insights

Why doesn't calling add.delay(4, 6) run the function immediately?

Because add.delay queues the task to run asynchronously by a worker, so the main program continues without waiting. See execution_table step 4 and 8.

What does the @app.task decorator do?

It wraps the function so Celery knows it is a task to run asynchronously. See execution_table step 3.

How is the result of the task accessed?

The result is stored by Celery backend and can be accessed via the AsyncResult object returned by delay(). See execution_table step 7.

Visual Quiz - 3 Questions

Look at the execution table, what is the state of 'result' after step 4?

A None

B Function return value

C AsyncResult object (pending)

D Task function itself

💡 Hint

Check variable_tracker column 'After Step 4' for 'result'

At which step does the worker actually run the task function?

A Step 4

B Step 6

C Step 5

D Step 7

💡 Hint

See execution_table where function executes and returns value

If you remove @app.task decorator, what changes in the execution?

A add.delay() call will fail or not queue task

B Task runs synchronously immediately

C Result is stored automatically

D Worker picks task faster

💡 Hint

Look at step 3 where task registration happens

Concept Snapshot

Defining Celery tasks in Flask:
- Import Celery and create instance with broker
- Use @app.task decorator on function
- Call task asynchronously with .delay(args)
- Worker runs task in background
- Result stored for later retrieval
Key: Tasks run outside main app flow asynchronously.

Full Transcript

This visual trace shows how to define and run Celery tasks in a Flask application. First, Celery is imported and an instance is created with a broker URL. Then, a function is decorated with @app.task to mark it as a Celery task. Calling the task with .delay() queues it asynchronously without blocking the main program. A worker process picks up the task, runs the function, and stores the result. The main app can continue running while the task executes in the background. The result can be accessed later through the AsyncResult object returned by .delay(). This approach helps run time-consuming tasks outside the main request flow.