chore(doc): add some examples for SparkSqlOnK8SOperator operator (#112)

examples/airflow/iceberg_query.sql

@@ -0,0 +1,14 @@
+CREATE TABLE IF NOT EXISTS prod.db.sample (
+    id bigint,
+    data string,
+    category string
+)
+USING iceberg
+PARTITIONED BY (category);
+
+MERGE INTO prod.db.sample t
+USING (SELECT * FROM prod.db.another_sample WHERE category = 'foo') s
+ON t.id = s.id
+WHEN MATCHED THEN UPDATE SET *
+WHEN NOT MATCHED THEN INSERT *
+WHEN NOT MATCHED BY SOURCE THEN DELETE;

examples/airflow/query.sql

@@ -0,0 +1,14 @@
+CREATE TABLE students (
+    name VARCHAR(64),
+    address VARCHAR(64),
+    added_at TIMESTAMP
+)
+USING PARQUET
+PARTITIONED BY (student_id INT);
+
+INSERT INTO students VALUES
+    ('Amy Smith', '123 Park Ave, San Jose', {{ ts }}, 111111);
+
+INSERT INTO students VALUES
+    ('Bob Brown', '456 Taylor St, Cupertino', {{ ts }}, 222222);,
+    ('Cathy Johnson', '789 Race Ave, Palo Alto', {{ ts}}, 333333);

examples/airflow/spark_sql_dag.py

@@ -0,0 +1,87 @@
+from __future__ import annotations
+
+from datetime import datetime
+
+from spark_on_k8s.airflow.operators import SparkSqlOnK8SOperator
+from spark_on_k8s.client import ExecutorInstances, PodResources
+
+from airflow.models.dag import DAG
+
+with DAG(
+    dag_id="spark_sql_on_k8s",
+    schedule=None,
+    start_date=datetime(2024, 1, 1),
+) as dag:
+    """
+    This DAG executes Spark SQL queries on Kubernetes using the SparkSqlOnK8SOperator.
+    """
+
+    SparkSqlOnK8SOperator(
+        task_id="spark_sql_application",
+        sql="CREATE TABLE test_table (key INT, value STRING) AS SELECT 1 AS key, 'value' AS value",
+        namespace="spark",
+        image="pyspark-job",
+        image_pull_policy="Never",
+        app_name="spark-sql-job-example",
+        service_account="spark",
+        app_waiter="log",
+        driver_resources=PodResources(cpu=1, memory=1024, memory_overhead=512),
+        executor_resources=PodResources(cpu=1, memory=1024, memory_overhead=512),
+        executor_instances=ExecutorInstances(min=0, max=5, initial=5),
+        ui_reverse_proxy=True,
+    )
+
+    SparkSqlOnK8SOperator(
+        task_id="spark_multiple_sql_application",
+        sql="SELECT 1; SELECT 2; SELECT {{ ds }}",
+        namespace="spark",
+        image="pyspark-job",
+        image_pull_policy="Never",
+        app_name="spark-sql-job-example",
+        service_account="spark",
+        app_waiter="log",
+        driver_resources=PodResources(cpu=1, memory=1024, memory_overhead=512),
+        executor_resources=PodResources(cpu=1, memory=1024, memory_overhead=512),
+        executor_instances=ExecutorInstances(min=0, max=5, initial=5),
+        ui_reverse_proxy=True,
+    )
+
+    SparkSqlOnK8SOperator(
+        task_id="spark_sql_from_file_application",
+        sql="query.sql",
+        namespace="spark",
+        image="pyspark-job",
+        image_pull_policy="Never",
+        app_name="spark-sql-job-example",
+        service_account="spark",
+        app_waiter="log",
+        driver_resources=PodResources(cpu=1, memory=1024, memory_overhead=512),
+        executor_resources=PodResources(cpu=1, memory=1024, memory_overhead=512),
+        executor_instances=ExecutorInstances(min=0, max=5, initial=5),
+        ui_reverse_proxy=True,
+    )
+
+    SparkSqlOnK8SOperator(
+        task_id="spark_iceberg_sql_application",
+        sql="iceberg_query.sql",
+        # it's better to use a custom image with the iceberg-spark-runtime package installed
+        packages=["org.apache.iceberg:iceberg-spark-runtime-3.5_2.12:1.7.0"],
+        spark_conf={
+            "spark.sql.extensions": "org.apache.iceberg.spark.extensions.IcebergSparkSessionExtensions",
+            "spark.sql.catalog.spark_catalog": "org.apache.iceberg.spark.SparkSessionCatalog",
+            "spark.sql.catalog.spark_catalog.type": "hive",
+            "spark.sql.catalog.prod": "org.apache.iceberg.spark.SparkCatalog",
+            "spark.sql.catalog.prod.type": "rest",
+            "spark.sql.catalog.prod.uri": "http://localhost:8080",
+        },
+        namespace="spark",
+        image="pyspark-job",
+        image_pull_policy="Never",
+        app_name="spark-sql-job-example",
+        service_account="spark",
+        app_waiter="log",
+        driver_resources=PodResources(cpu=1, memory=1024, memory_overhead=512),
+        executor_resources=PodResources(cpu=1, memory=1024, memory_overhead=512),
+        executor_instances=ExecutorInstances(min=0, max=5, initial=5),
+        ui_reverse_proxy=True,
+    )