gh-yzou commented on code in PR #1488: URL: https://github.com/apache/polaris/pull/1488#discussion_r2067006226
########## plugins/spark/README.md: ########## @@ -30,11 +30,63 @@ and depends on iceberg-spark-runtime 1.8.1. # Build Plugin Jar A task createPolarisSparkJar is added to build a jar for the Polaris Spark plugin, the jar is named as: -"polaris-iceberg-<iceberg_version>-spark-runtime-<spark_major_version>_<scala_version>.jar" +The result jar is located at plugins/spark/v3.5/build/<scala_version>/libs after the build. -Building the Polaris project produces client jars for both Scala 2.12 and 2.13, and CI runs the Spark -client tests for both Scala versions as well. +# Start Spark with Local Polaris Service using built Jar +Once the jar is built, we can manually test it with Spark and a local Polaris service. -The Jar can also be built alone with a specific version using target `:polaris-spark-3.5_<scala_version>`. For example: -- `./gradlew :polaris-spark-3.5_2.12:createPolarisSparkJar` - Build a jar for the Polaris Spark plugin with scala version 2.12. -The result jar is located at plugins/spark/build/<scala_version>/libs after the build. +The following command starts a Polaris server for local testing, it runs on localhost:8181 with default +realm `POLARIS` and root credentials `root:secret`: +```shell +./gradlew run +``` + +Once the local server is running, the following command can be used to start the spark-shell with the built Spark client +jar, and to use the local Polaris server as a Catalog. + +```shell +bin/spark-shell \ +--jars <path-to-spark-client-jar> \ +--packages org.apache.hadoop:hadoop-aws:3.4.0,io.delta:delta-spark_2.12:3.3.1 \ +--conf spark.sql.extensions=org.apache.iceberg.spark.extensions.IcebergSparkSessionExtensions,io.delta.sql.DeltaSparkSessionExtension \ +--conf spark.sql.catalog.spark_catalog=org.apache.spark.sql.delta.catalog.DeltaCatalog \ +--conf spark.sql.catalog.<catalog-name>.warehouse=<catalog-name> \ +--conf spark.sql.catalog.<catalog-name>.header.X-Iceberg-Access-Delegation=true \ +--conf spark.sql.catalog.<catalog-name>=org.apache.polaris.spark.SparkCatalog \ +--conf spark.sql.catalog.<catalog-name>.uri=http://localhost:8181/api/catalog \ +--conf spark.sql.catalog.<catalog-name>.credential="root:secret" \ +--conf spark.sql.catalog.<catalog-name>.scope='PRINCIPAL_ROLE:ALL' \ +--conf spark.sql.catalog.<catalog-name>.token-refresh-enabled=true \ +--conf spark.sql.catalog.<catalog-name>.type=rest \ +--conf spark.sql.sources.useV1SourceList='' +``` + +Assume the path to the built Spark client jar is +`/polaris/plugins/spark/v3.5/spark/build/2.12/libs/polaris-iceberg-1.8.1-spark-runtime-3.5_2.12-0.10.0-beta-incubating-SNAPSHOT.jar` +and the name of the catalog is `polaris`. The cli command will look like following: + +```shell +bin/spark-shell \ +--jars /polaris/plugins/spark/v3.5/spark/build/2.12/libs/polaris-iceberg-1.8.1-spark-runtime-3.5_2.12-0.10.0-beta-incubating-SNAPSHOT.jar \ +--packages org.apache.hadoop:hadoop-aws:3.4.0,io.delta:delta-spark_2.12:3.3.1 \ +--conf spark.sql.extensions=org.apache.iceberg.spark.extensions.IcebergSparkSessionExtensions,io.delta.sql.DeltaSparkSessionExtension \ +--conf spark.sql.catalog.spark_catalog=org.apache.spark.sql.delta.catalog.DeltaCatalog \ +--conf spark.sql.catalog.polaris.warehouse=<catalog-name> \ +--conf spark.sql.catalog.polaris.header.X-Iceberg-Access-Delegation=true \ +--conf spark.sql.catalog.polaris=org.apache.polaris.spark.SparkCatalog \ +--conf spark.sql.catalog.polaris.uri=http://localhost:8181/api/catalog \ +--conf spark.sql.catalog.polaris.credential="root:secret" \ +--conf spark.sql.catalog.polaris.scope='PRINCIPAL_ROLE:ALL' \ +--conf spark.sql.catalog.polaris.token-refresh-enabled=true \ +--conf spark.sql.catalog.polaris.type=rest \ +--conf spark.sql.sources.useV1SourceList='' +``` + +# Limitations +The Spark client does have limitations as follows: +1) Create table as select (CTAS) is not supported for Delta tables. +2) Dataframe saveAsTable is not supported for Delta tables, which requires CTAS support. +3) Create non-iceberg table without explicit location is not supported. +4) Rename table for non-iceberg table is not supported. +5) ALTER TABLE ... SET LOCATION/SET FILEFORMAT/ADD PARTITION is not supported for DELTA table. +6) No specific guarantee is provided for table types that are neither ICEBERG or DELTA at this moment. Review Comment: I haven't do detailed testing for other table types like csv yet, so no guarantee can be provided for those table types, if user run into problems for other table types, that will be a feature request (that is all I want to say :). ``` I'm guessing it doesn't support Iceberg-as-generic-tables ``` yes, the client route all iceberg table related requests to the IRC endpoints, so they will all be Iceberg Entities. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
