Creating a custom plugin for Apache Airflow PythonVirtualenvOperator - Amazon Managed Workflows for Apache Airflow

Creating a custom plugin for Apache Airflow PythonVirtualenvOperator

The following sample shows how to patch the Apache Airflow PythonVirtualenvOperator with a custom plugin on Amazon Managed Workflows for Apache Airflow.

Version

  • The sample code on this page can be used with Apache Airflow v1 in Python 3.7.

  • You can use the code example on this page with Apache Airflow v2 in Python 3.10.

Prerequisites

To use the sample code on this page, you'll need the following:

Permissions

  • No additional permissions are required to use the code example on this page.

Requirements

To use the sample code on this page, add the following dependencies to your requirements.txt. To learn more, see Installing Python dependencies.

virtualenv

Custom plugin sample code

Apache Airflow will execute the contents of Python files in the plugins folder at startup. This plugin will patch the built-in PythonVirtualenvOperator during that startup process to make it compatible with Amazon MWAA. The following steps show the sample code for the custom plugin.

Apache Airflow v2
  1. In your command prompt, navigate to the plugins directory above. For example:

    cd plugins
  2. Copy the contents of the following code sample and save locally as virtual_python_plugin.py.

    """ Copyright Amazon.com, Inc. or its affiliates. All Rights Reserved. Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. """ from airflow.plugins_manager import AirflowPlugin import airflow.utils.python_virtualenv from typing import List def _generate_virtualenv_cmd(tmp_dir: str, python_bin: str, system_site_packages: bool) -> List[str]: cmd = ['python3','/usr/local/airflow/.local/lib/python3.7/site-packages/virtualenv', tmp_dir] if system_site_packages: cmd.append('--system-site-packages') if python_bin is not None: cmd.append(f'--python={python_bin}') return cmd airflow.utils.python_virtualenv._generate_virtualenv_cmd=_generate_virtualenv_cmd class VirtualPythonPlugin(AirflowPlugin): name = 'virtual_python_plugin'
Apache Airflow v1
  1. In your command prompt, navigate to the plugins directory above. For example:

    cd plugins
  2. Copy the contents of the following code sample and save locally as virtual_python_plugin.py.

    from airflow.plugins_manager import AirflowPlugin from airflow.operators.python_operator import PythonVirtualenvOperator def _generate_virtualenv_cmd(self, tmp_dir): cmd = ['python3','/usr/local/airflow/.local/lib/python3.7/site-packages/virtualenv', tmp_dir] if self.system_site_packages: cmd.append('--system-site-packages') if self.python_version is not None: cmd.append('--python=python{}'.format(self.python_version)) return cmd PythonVirtualenvOperator._generate_virtualenv_cmd=_generate_virtualenv_cmd class EnvVarPlugin(AirflowPlugin): name = 'virtual_python_plugin'

Plugins.zip

The following steps show how to create the plugins.zip.

  1. In your command prompt, navigate to the directory containing virtual_python_plugin.py above. For example:

    cd plugins
  2. Zip the contents within your plugins folder.

    zip plugins.zip virtual_python_plugin.py

Code sample

The following steps describe how to create the DAG code for the custom plugin.

Apache Airflow v2
  1. In your command prompt, navigate to the directory where your DAG code is stored. For example:

    cd dags
  2. Copy the contents of the following code sample and save locally as virtualenv_test.py.

    """ Copyright Amazon.com, Inc. or its affiliates. All Rights Reserved. Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. """ from airflow import DAG from airflow.operators.python import PythonVirtualenvOperator from airflow.utils.dates import days_ago import os os.environ["PATH"] = os.getenv("PATH") + ":/usr/local/airflow/.local/bin" def virtualenv_fn(): import boto3 print("boto3 version ",boto3.__version__) with DAG(dag_id="virtualenv_test", schedule_interval=None, catchup=False, start_date=days_ago(1)) as dag: virtualenv_task = PythonVirtualenvOperator( task_id="virtualenv_task", python_callable=virtualenv_fn, requirements=["boto3>=1.17.43"], system_site_packages=False, dag=dag, )
Apache Airflow v1
  1. In your command prompt, navigate to the directory where your DAG code is stored. For example:

    cd dags
  2. Copy the contents of the following code sample and save locally as virtualenv_test.py.

    """ Copyright Amazon.com, Inc. or its affiliates. All Rights Reserved. Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. """ from airflow import DAG from airflow.operators.python_operator import PythonVirtualenvOperator from airflow.utils.dates import days_ago import os os.environ["PATH"] = os.getenv("PATH") + ":/usr/local/airflow/.local/bin" def virtualenv_fn(): import boto3 print("boto3 version ",boto3.__version__) with DAG(dag_id="virtualenv_test", schedule_interval=None, catchup=False, start_date=days_ago(1)) as dag: virtualenv_task = PythonVirtualenvOperator( task_id="virtualenv_task", python_callable=virtualenv_fn, requirements=["boto3>=1.17.43"], system_site_packages=False, dag=dag, )

Airflow configuration options

If you're using Apache Airflow v2, add core.lazy_load_plugins : False as an Apache Airflow configuration option. To learn more, see Using configuration options to load plugins in 2.

What's next?