Airflow-DAGs schreiben

Auf GitHub ansehen Feedback

import datetime

from airflow import models
default_dag_args = {
    # The start_date describes when a DAG is valid / can be run. Set this to a
    # fixed point in time rather than dynamically, since it is evaluated every
    # time a DAG is parsed. See:
    # https://airflow.apache.org/faq.html#what-s-the-deal-with-start-date
    'start_date': datetime.datetime(2018, 1, 1),
}

# Define a DAG (directed acyclic graph) of tasks.
# Any task you create within the context manager is automatically added to the
# DAG object.
with models.DAG(
        'composer_sample_simple_greeting',
        schedule_interval=datetime.timedelta(days=1),
        default_args=default_dag_args) as dag:

Airflow 1

Auf GitHub ansehen Feedback

import datetime

from airflow import models
default_dag_args = {
    # The start_date describes when a DAG is valid / can be run. Set this to a
    # fixed point in time rather than dynamically, since it is evaluated every
    # time a DAG is parsed. See:
    # https://airflow.apache.org/faq.html#what-s-the-deal-with-start-date
    'start_date': datetime.datetime(2018, 1, 1),
}

# Define a DAG (directed acyclic graph) of tasks.
# Any task you create within the context manager is automatically added to the
# DAG object.
with models.DAG(
        'composer_sample_simple_greeting',
        schedule_interval=datetime.timedelta(days=1),
        default_args=default_dag_args) as dag:

Operatoren, mit denen die auszuführende Arbeit beschrieben wird. Eine Instanziierung eines Operators wird als Aufgabe bezeichnet.

Airflow 2

Auf GitHub ansehen Feedback

from airflow.operators import bash_operator
from airflow.operators import python_operator
    def greeting():
        import logging
        logging.info('Hello World!')

    # An instance of an operator is called a task. In this case, the
    # hello_python task calls the "greeting" Python function.
    hello_python = python_operator.PythonOperator(
        task_id='hello',
        python_callable=greeting)

    # Likewise, the goodbye_bash task calls a Bash script.
    goodbye_bash = bash_operator.BashOperator(
        task_id='bye',
        bash_command='echo Goodbye.')

Airflow 1

Auf GitHub ansehen Feedback

from airflow.operators import bash_operator
from airflow.operators import python_operator
    def greeting():
        import logging
        logging.info('Hello World!')

    # An instance of an operator is called a task. In this case, the
    # hello_python task calls the "greeting" Python function.
    hello_python = python_operator.PythonOperator(
        task_id='hello',
        python_callable=greeting)

    # Likewise, the goodbye_bash task calls a Bash script.
    goodbye_bash = bash_operator.BashOperator(
        task_id='bye',
        bash_command='echo Goodbye.')

Aufgabenbeziehungen, die beschreiben, in welcher Reihenfolge die Ausführung erfolgen soll:

Airflow 2

Auf GitHub ansehen Feedback

# Define the order in which the tasks complete by using the >> and <<
# operators. In this example, hello_python executes before goodbye_bash.
hello_python >> goodbye_bash

Airflow 1

Auf GitHub ansehen Feedback

# Define the order in which the tasks complete by using the >> and <<
# operators. In this example, hello_python executes before goodbye_bash.
hello_python >> goodbye_bash

Beispiel für einen vollständigen DAG-Workflow in Python

Der folgende Workflow ist eine vollständige funktionierende DAG-Vorlage, die sich aus zwei Aufgaben zusammensetzt: einer hello_python-Aufgabe und einer goodbye_bash-Aufgabe:

Airflow 2

Auf GitHub ansehen Feedback

from __future__ import print_function

import datetime

from airflow import models
from airflow.operators import bash_operator
from airflow.operators import python_operator

default_dag_args = {
    # The start_date describes when a DAG is valid / can be run. Set this to a
    # fixed point in time rather than dynamically, since it is evaluated every
    # time a DAG is parsed. See:
    # https://airflow.apache.org/faq.html#what-s-the-deal-with-start-date
    'start_date': datetime.datetime(2018, 1, 1),
}

# Define a DAG (directed acyclic graph) of tasks.
# Any task you create within the context manager is automatically added to the
# DAG object.
with models.DAG(
        'composer_sample_simple_greeting',
        schedule_interval=datetime.timedelta(days=1),
        default_args=default_dag_args) as dag:
    def greeting():
        import logging
        logging.info('Hello World!')

    # An instance of an operator is called a task. In this case, the
    # hello_python task calls the "greeting" Python function.
    hello_python = python_operator.PythonOperator(
        task_id='hello',
        python_callable=greeting)

    # Likewise, the goodbye_bash task calls a Bash script.
    goodbye_bash = bash_operator.BashOperator(
        task_id='bye',
        bash_command='echo Goodbye.')

    # Define the order in which the tasks complete by using the >> and <<
    # operators. In this example, hello_python executes before goodbye_bash.
    hello_python >> goodbye_bash

Airflow 1

Auf GitHub ansehen Feedback

from __future__ import print_function

import datetime

from airflow import models
from airflow.operators import bash_operator
from airflow.operators import python_operator

default_dag_args = {
    # The start_date describes when a DAG is valid / can be run. Set this to a
    # fixed point in time rather than dynamically, since it is evaluated every
    # time a DAG is parsed. See:
    # https://airflow.apache.org/faq.html#what-s-the-deal-with-start-date
    'start_date': datetime.datetime(2018, 1, 1),
}

# Define a DAG (directed acyclic graph) of tasks.
# Any task you create within the context manager is automatically added to the
# DAG object.
with models.DAG(
        'composer_sample_simple_greeting',
        schedule_interval=datetime.timedelta(days=1),
        default_args=default_dag_args) as dag:
    def greeting():
        import logging
        logging.info('Hello World!')

    # An instance of an operator is called a task. In this case, the
    # hello_python task calls the "greeting" Python function.
    hello_python = python_operator.PythonOperator(
        task_id='hello',
        python_callable=greeting)

    # Likewise, the goodbye_bash task calls a Bash script.
    goodbye_bash = bash_operator.BashOperator(
        task_id='bye',
        bash_command='echo Goodbye.')

    # Define the order in which the tasks complete by using the >> and <<
    # operators. In this example, hello_python executes before goodbye_bash.
    hello_python >> goodbye_bash

Airflow-Anleitungen und -Konzepte

Weitere Informationen zum Definieren von Airflow-DAGs finden Sie in der Airflow-Anleitung und in den Airflow-Konzepten.

Airflow-Operatoren

Die folgenden Beispiele enthalten einige beliebte Airflow-Operatoren. Informationen zu weiteren Airflow-Operatoren finden Sie in der Apache Airflow API-Referenz und im Quellcode der Operatoren core, contrib und providers.

BashOperator

Mit dem BashOperator können Sie Befehlszeilenprogramme ausführen.

Airflow 2

Auf GitHub ansehen Feedback

from airflow.operators import bash

    # Create BigQuery output dataset.
    make_bq_dataset = bash.BashOperator(
        task_id="make_bq_dataset",
        # Executing 'bq' command requires Google Cloud SDK which comes
        # preinstalled in Cloud Composer.
        bash_command=f"bq ls {bq_dataset_name} || bq mk {bq_dataset_name}",
    )

Airflow 1

Auf GitHub ansehen Feedback

from airflow.operators import bash_operator

    # Create BigQuery output dataset.
    make_bq_dataset = bash_operator.BashOperator(
        task_id="make_bq_dataset",
        # Executing 'bq' command requires Google Cloud SDK which comes
        # preinstalled in Cloud Composer.
        bash_command="bq ls {} || bq mk {}".format(bq_dataset_name, bq_dataset_name),
    )

Cloud Composer führt die bereitgestellten Befehle in einem Bash-Skript auf einem Worker aus. Der Worker ist ein Docker-Container auf Basis von Debian und enthält mehrere Pakete.

PythonOperator

Verwenden Sie den PythonOperator, um beliebigen Python-Code auszuführen.

Cloud Composer führt den Python-Code in einem Container aus, der Pakete für die in Ihrer Umgebung verwendete Cloud Composer-Image-Version enthält.

Informationen zum Installieren weiterer Python-Pakete finden Sie unter Python-Abhängigkeiten installieren.

Google Cloud-Operatoren

Mit den Google Cloud-Airflow-Operatoren können Sie Aufgaben ausführen, die Google Cloud-Produkte nutzen. Cloud Composer konfiguriert automatisch eine Airflow-Verbindung zum Projekt der Umgebung.

BigQuery-Operatoren fragen Daten in BigQuery ab und verarbeiten sie.

Airflow 2

Auf GitHub ansehen Feedback

from airflow.providers.google.cloud.operators import bigquery
from airflow.providers.google.cloud.transfers import bigquery_to_gcs

    bq_recent_questions_query = bigquery.BigQueryInsertJobOperator(
        task_id="bq_recent_questions_query",
        configuration={
            "query": {
                "query": RECENT_QUESTIONS_QUERY,
                "useLegacySql": False,
                "destinationTable": {
                    "projectId": project_id,
                    "datasetId": bq_dataset_name,
                    "tableId": bq_recent_questions_table_id,
                },
            }
        },
        location=location,
    )

Airflow 1

Auf GitHub ansehen Feedback

from airflow.contrib.operators import bigquery_operator

    # Query recent StackOverflow questions.
    bq_recent_questions_query = bigquery_operator.BigQueryOperator(
        task_id="bq_recent_questions_query",
        sql="""
        SELECT owner_display_name, title, view_count
        FROM `bigquery-public-data.stackoverflow.posts_questions`
        WHERE creation_date < CAST('{max_date}' AS TIMESTAMP)
            AND creation_date >= CAST('{min_date}' AS TIMESTAMP)
        ORDER BY view_count DESC
        LIMIT 100
        """.format(
            max_date=max_query_date, min_date=min_query_date
        ),
        use_legacy_sql=False,
        destination_dataset_table=bq_recent_questions_table_id,
    )

Mit Cloud Dataflow-Operatoren werden Apache Beam-Jobs in Dataflow ausgeführt.
Mit Cloud Data Fusion-Operatoren werden Cloud Data Fusion-Pipelines verwaltet und ausgeführt.
Mit Dataproc-Operatoren werden Hadoop- und Spark-Jobs in Dataproc ausgeführt.
Mit Datastore-Operatoren werden Daten in Datastore gelesen und geschrieben.
Mit AI Platform-Operatoren werden Trainings- und Vorhersagejobs in AI Platform ausgeführt.
Mit Cloud Storage-Operatoren werden Daten in Cloud Storage gelesen und geschrieben.

EmailOperator

Verwenden Sie den EmailOperator, um E-Mails von einem DAG zu senden. Wenn Sie E-Mails aus einer Cloud Composer-Umgebung senden möchten, müssen Sie die Verwendung von SendGrid in der Umgebung konfigurieren.

Airflow 2

Auf GitHub ansehen Feedback

from airflow.operators import email

    # Send email confirmation (you will need to set up the email operator
    # See https://cloud.google.com/composer/docs/how-to/managing/creating#notification
    # for more info on configuring the email operator in Cloud Composer)
    email_summary = email.EmailOperator(
        task_id="email_summary",
        to="{{var.value.email}}",
        subject="Sample BigQuery notify data ready",
        html_content="""
        Analyzed Stack Overflow posts data from {min_date} 12AM to {max_date}
        12AM. The most popular question was '{question_title}' with
        {view_count} views. Top 100 questions asked are now available at:
        {export_location}.
        """.format(
            min_date=min_query_date,
            max_date=max_query_date,
            question_title=(
                "{{ ti.xcom_pull(task_ids='bq_read_most_popular', "
                "key='return_value')[0][0] }}"
            ),
            view_count=(
                "{{ ti.xcom_pull(task_ids='bq_read_most_popular', "
                "key='return_value')[0][1] }}"
            ),
            export_location=output_file,
        ),
    )

Airflow 1

Auf GitHub ansehen Feedback

from airflow.operators import email_operator

    # Send email confirmation
    email_summary = email_operator.EmailOperator(
        task_id="email_summary",
        to="{{var.value.email}}",
        subject="Sample BigQuery notify data ready",
        html_content="""
        Analyzed Stack Overflow posts data from {min_date} 12AM to {max_date}
        12AM. The most popular question was '{question_title}' with
        {view_count} views. Top 100 questions asked are now available at:
        {export_location}.
        """.format(
            min_date=min_query_date,
            max_date=max_query_date,
            question_title=(
                "{{ ti.xcom_pull(task_ids='bq_read_most_popular', "
                "key='return_value')[0][0] }}"
            ),
            view_count=(
                "{{ ti.xcom_pull(task_ids='bq_read_most_popular', "
                "key='return_value')[0][1] }}"
            ),
            export_location=output_file,
        ),
    )

Benachrichtigungen

Zum Senden einer E-Mail-Benachrichtigung, wenn ein Operator im DAG fehlerhaft ist, legen Sie für email_on_failure den Wert True fest. Zum Senden von E-Mails aus einer Cloud Composer-Umgebung müssen Sie Ihre Umgebung für die Verwendung von SendGrid konfigurieren.

Airflow 2

Auf GitHub ansehen Feedback

from airflow import models

default_dag_args = {
    "start_date": yesterday,
    # Email whenever an Operator in the DAG fails.
    "email": "{{var.value.email}}",
    "email_on_failure": True,
    "email_on_retry": False,
    "retries": 1,
    "retry_delay": datetime.timedelta(minutes=5),
    "project_id": project_id,
}

with models.DAG(
    "composer_sample_bq_notify",
    schedule_interval=datetime.timedelta(weeks=4),
    default_args=default_dag_args,
) as dag:

Airflow 1