Article Directory
Knowledge point 07: Shell scheduling test
-
Goal : Realize the scheduling test of Shell commands
-
implement
-
Requirement : Use BashOperator to schedule and execute a Linux command
-
the code
-
create
# 默认的Airflow自动检测工作流程序的文件的目录 mkdir -p /root/airflow/dags cd /root/airflow/dags vim first_bash_operator.py
-
to develop
# import from airflow import DAG from airflow.operators.bash import BashOperator from airflow.utils.dates import days_ago from datetime import timedelta # define args default_args = { 'owner': 'airflow', 'email': ['[email protected]'], 'email_on_failure': True, 'email_on_retry': True, 'retries': 1, 'retry_delay': timedelta(minutes=1), } # define dag dag = DAG( 'first_airflow_dag', default_args=default_args, description='first airflow task DAG', schedule_interval=timedelta(days=1), start_date=days_ago(1), tags=['itcast_bash'], ) # define task1 run_bash_task = BashOperator( task_id='first_bashoperator_task', bash_command='echo "hello airflow"', dag=dag, ) # run the task run_bash_task
-
Using bashOperator at work
bash_command='sh xxxx.sh'
-
xxxx.sh: as required
- Linux commands
- hive -f
- spark-sql -f
- spark-submit python | jar
-
-
submit
python first_bash_operator.py
-
Check
-
implement
-
-
summary
- Realize the scheduling test of Shell command
Knowledge point 08: Dependency scheduling test
-
Goal : Realize AirFlow's dependency scheduling test
-
implement
-
Requirements : Use BashOperator to schedule and execute multiple Tasks and build dependencies
-
the code
-
create
cd /root/airflow/dags vim second_bash_operator.py
-
to develop
# import from datetime import timedelta from airflow import DAG from airflow.operators.bash import BashOperator from airflow.utils.dates import days_ago # define args default_args = { 'owner': 'airflow', 'email': ['[email protected]'], 'email_on_failure': True, 'email_on_retry': True, 'retries': 1, 'retry_delay': timedelta(minutes=1), } # define dag dag = DAG( 'second_airflow_dag', default_args=default_args, description='first airflow task DAG', schedule_interval=timedelta(days=1), start_date=days_ago(1), tags=['itcast_bash'], ) # define task1 say_hello_task = BashOperator( task_id='say_hello_task', bash_command='echo "start task"', dag=dag, ) # define task2 print_date_format_task2 = BashOperator( task_id='print_date_format_task2', bash_command='date +"%F %T"', dag=dag, ) # define task3 print_date_format_task3 = BashOperator( task_id='print_date_format_task3', bash_command='date +"%F %T"', dag=dag, ) # define task4 end_task4 = BashOperator( task_id='end_task', bash_command='echo "end task"', dag=dag, ) say_hello_task >> [print_date_format_task2,print_date_format_task3] >> end_task4
-
-
submit
python second_bash_operator.py
-
Check
-
-
summary
- Implement AirFlow's dependency scheduling test
Knowledge point 09: Python scheduling test
-
Goal : Implement scheduled testing of Python code
-
implement
-
Requirement : Scheduling the running of Python code Task
-
the code
-
create
cd /root/airflow/dags vim python_etl_airflow.py
-
to develop
# import package from airflow import DAG from airflow.operators.python import PythonOperator from airflow.utils.dates import days_ago import json # define args default_args = { 'owner': 'airflow', } # define the dag with DAG( 'python_etl_dag', default_args=default_args, description='DATA ETL DAG', schedule_interval=None, start_date=days_ago(2), tags=['itcast'], ) as dag: # function1 def extract(**kwargs): ti = kwargs['ti'] data_string = '{"1001": 301.27, "1002": 433.21, "1003": 502.22, "1004": 606.65, "1005": 777.03}' ti.xcom_push('order_data', data_string) # function2 def transform(**kwargs): ti = kwargs['ti'] extract_data_string = ti.xcom_pull(task_ids='extract', key='order_data') order_data = json.loads(extract_data_string) total_order_value = 0 for value in order_data.values(): total_order_value += value total_value = { "total_order_value": total_order_value} total_value_json_string = json.dumps(total_value) ti.xcom_push('total_order_value', total_value_json_string) # function3 def load(**kwargs): ti = kwargs['ti'] total_value_string = ti.xcom_pull(task_ids='transform', key='total_order_value') total_order_value = json.loads(total_value_string) print(total_order_value) # task1 extract_task = PythonOperator( task_id='extract', python_callable=extract, ) extract_task.doc_md = """\ #### Extract task A simple Extract task to get data ready for the rest of the data pipeline. In this case, getting data is simulated by reading from a hardcoded JSON string. This data is then put into xcom, so that it can be processed by the next task. """ # task2 transform_task = PythonOperator( task_id='transform', python_callable=transform, ) transform_task.doc_md = """\ #### Transform task A simple Transform task which takes in the collection of order data from xcom and computes the total order value. This computed value is then put into xcom, so that it can be processed by the next task. """ # task3 load_task = PythonOperator( task_id='load', python_callable=load, ) load_task.doc_md = """\ #### Load task A simple Load task which takes in the result of the Transform task, by reading it from xcom and instead of saving it to end user review, just prints it out. """ # run extract_task >> transform_task >> load_task
-
-
submit
python python_etl_airflow.py
-
Check
-
-
summary
- Implement scheduled testing of Python code
Knowledge point 10: Oracle and MySQL scheduling methods
-
Goal : Understand the scheduling methods of Oracle and MySQL
-
implement
-
Oracle scheduling : refer to "oracle task scheduling detailed operation document.md"
-
step1: Install the Oracle client locally
-
step2: Install the AirFlow integrated Oracle library
-
step3: Create an Oracle connection
-
step4: Development and testing
query_oracle_task = OracleOperator( task_id = 'oracle_operator_task', sql = 'select * from ciss4.ciss_base_areas', oracle_conn_id = 'oracle-airflow-connection', autocommit = True, dag=dag )
-
-
MySQL scheduling : "MySQL task scheduling detailed operation document.md"
-
step1: Install the MySQL client locally
-
step2: Install the AirFlow integrated MySQL library
-
step3: Create a MySQL connection
-
step4: Development and testing
-
Method 1: specify the SQL statement
query_table_mysql_task = MySqlOperator( task_id='query_table_mysql', mysql_conn_id='mysql_airflow_connection', sql=r"""select * from test.test_airflow_mysql_task;""", dag=dag )
-
Method 2: specify the SQL file
query_table_mysql_task = MySqlOperator( task_id='query_table_mysql_second', mysql_conn_id='mysql-airflow-connection', sql='test_airflow_mysql_task.sql', dag=dag )
-
-
Method 3: specify variables
insert_sql = r""" INSERT INTO `test`.`test_airflow_mysql_task`(`task_name`) VALUES ( 'test airflow mysql task3'); INSERT INTO `test`.`test_airflow_mysql_task`(`task_name`) VALUES ( 'test airflow mysql task4'); INSERT INTO `test`.`test_airflow_mysql_task`(`task_name`) VALUES ( 'test airflow mysql task5'); """ insert_table_mysql_task = MySqlOperator( task_id='mysql_operator_insert_task', mysql_conn_id='mysql-airflow-connection', sql=insert_sql, dag=dag )
-
-
-
-
summary
- Understand the scheduling methods of Oracle and MySQL
Knowledge point 11: big data component scheduling method
-
Goal : Understand big data component scheduling methods
-
implement
-
Types supported by AirFlow
- HiveOperator
- PrestoOperator
- SparkSqlOperator
-
Requirements : Sqoop, MR, Hive, Spark, Flink
-
Solution : use BashOperator or PythonOperator uniformly, and encapsulate the corresponding program in the script
-
Sqoop
run_sqoop_task = BashOperator( task_id='sqoop_task', bash_command='sqoop --options-file xxxx.sqoop', dag=dag, )
-
Hive
run_hive_task = BashOperator( task_id='hive_task', bash_command='hive -f xxxx.sql', dag=dag, )
-
Spark
run_spark_task = BashOperator( task_id='spark_task', bash_command='spark-sql -f xxxx.sql', dag=dag, )
-
Considerable
run_flink_task = BashOperator( task_id='flink_task', bash_command='flink run /opt/flink-1.12.2/examples/batch/WordCount.jar', dag=dag, )
-
-
-
summary
- Understand Big Data Component Scheduling Methods