How to Assign a List of Strings in Airflow XCOM to a Global Variable using xcom_pull

What will you learn?

In this tutorial, you will learn how to retrieve a list of strings from Apache Airflow’s XCom feature and assign it to a global variable in Python. This knowledge is crucial for passing data between tasks efficiently in your Airflow workflows.

Introduction to the Problem and Solution

When working with Apache Airflow, the need often arises to transfer data between tasks. The XCom (cross-communication) feature provided by Airflow facilitates this data exchange seamlessly. One common scenario involves retrieving a list of strings stored in XCom and assigning it to a global variable within our Python code. This tutorial will guide you through achieving this task effortlessly.

To address this challenge, we will utilize the xcom_pull method available in the Airflow context. By pulling data from XCom in another task and assigning it to a global variable within our Python script, we can conveniently access the list of strings across various stages of our workflow.


from airflow.models import DAG
from airflow.utils.dates import days_ago

# Define your custom function or operator here that generates the desired list of strings

def my_task(**kwargs):
    # Pulling XCom data from another task named 'task_name'
    xcom_data = kwargs['ti'].xcom_pull(task_ids='task_name', key='my_key')

    # Assigning the retrieved list of strings to a global variable 'my_global_variable'
    globals()['my_global_variable'] = xcom_data

# Define your custom tasks/operators within an Airflow DAG here

# Setting up your DAG with appropriate parameters like schedule_interval, default_args etc.

# Copyright PHD


In the provided code snippet: 1. We import necessary modules from Airflow. 2. We define a custom function my_task which uses the xcom_pull method to retrieve data stored by another task (‘task_name’) with key ‘my_key’. 3. The retrieved list is then assigned as a value to the global variable my_global_variable, enabling easy access throughout our script.

By following these steps, you can effectively assign a list of strings fetched from an Airflow XCom object into a global variable for further processing within your Python script.

    How does xcom_pull work in Apache Airflow?
    • When you call xcom_pull, it retrieves an XCom value pushed by another task instance.

    Can I pull multiple values using xcom_pull?

    • Yes, you can pull multiple values by specifying different keys or calling xcom_push multiple times from various tasks.

    Is there any size limit on what I can store/retrieve using XCom?

    • While there isn’t an official size limit mentioned, it’s recommended not to transfer large objects via XCom due to performance reasons.

    How do I handle if no value is found for the specified key during xcom_pull?

    • If no value is found for the given key during xcom_pull, it will return None unless explicitly specified otherwise in your code.

    Can I use xcom_push and xcoom_puull between different DAGs?

    • No, XCom operations are limited within the same Directed Acyclic Graph (DAG), however triggering tasks across DAGs might be possible through some advanced configurations.

    What happens if I try pulling an XCom value that doesn’t exist yet?

    • If you attempt pulling an XCom value before any task has pushed one with matching criteria (like task_id and key), you’ll receive None as output until suitable data gets pushed.


    In conclusion, harnessing Airflow�s XCOM functionality, particularly utilizing xcmo-pull, streamlines communication between different components in your workflow pipeline. Understanding how to extract values stored in one task�s context and utilize them globally within other parts of your program opens up possibilities for creating more dynamic and interconnected workflows efficiently.

    Leave a Comment