feat(tests): support custom product versions (#760)

dervoeti · web-flow · commit 6b83a297fdf5 · 2025-06-24T06:43:11.000Z
* feat: support custom product versions

* chore: changelog

* chore: python formatting
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -31,6 +31,7 @@ All notable changes to this project will be documented in this file.
   - The defaults from the docker images itself will now apply, which will be different from 1000/0 going forward
   - This is marked as breaking because tools and policies might exist, which require these fields to be set
 - Deprecate Trino 470 ([#755]).
+- test: support custom versions ([#760]).
 
 ### Fixed
 
@@ -52,6 +53,7 @@ All notable changes to this project will be documented in this file.
 [#748]: https://github.com/stackabletech/trino-operator/pull/748
 [#752]: https://github.com/stackabletech/trino-operator/pull/752
 [#755]: https://github.com/stackabletech/trino-operator/pull/755
+[#760]: https://github.com/stackabletech/trino-operator/pull/760
 
 ## [25.3.0] - 2025-03-21
 
diff --git a/tests/templates/kuttl/smoke/check-s3.py b/tests/templates/kuttl/smoke/check-s3.py
@@ -2,14 +2,19 @@
 import trino
 import argparse
 import sys
+import re
 
 if not sys.warnoptions:
     import warnings
 warnings.simplefilter("ignore")
 
 
 def get_connection(username, password, namespace):
-    host = 'trino-coordinator-default-0.trino-coordinator-default.' + namespace + '.svc.cluster.local'
+    host = (
+        "trino-coordinator-default-0.trino-coordinator-default."
+        + namespace
+        + ".svc.cluster.local"
+    )
     # If you want to debug this locally use
     # kubectl -n kuttl-test-XXX port-forward svc/trino-coordinator-default 8443
     # host = '127.0.0.1'
@@ -18,7 +23,7 @@ def get_connection(username, password, namespace):
         host=host,
         port=8443,
         user=username,
-        http_scheme='https',
+        http_scheme="https",
         auth=trino.auth.BasicAuthentication(username, password),
         session_properties={"query_max_execution_time": "60s"},
     )
@@ -33,34 +38,47 @@ def run_query(connection, query):
     return cursor.fetchall()
 
 
-if __name__ == '__main__':
+if __name__ == "__main__":
     # Construct an argument parser
     all_args = argparse.ArgumentParser()
     # Add arguments to the parser
-    all_args.add_argument("-n", "--namespace", required=True, help="Namespace the test is running in")
+    all_args.add_argument(
+        "-n", "--namespace", required=True, help="Namespace the test is running in"
+    )
 
     args = vars(all_args.parse_args())
     namespace = args["namespace"]
 
     print("Starting S3 tests...")
     connection = get_connection("admin", "admin", namespace)
 
-    trino_version = run_query(connection, "select node_version from system.runtime.nodes where coordinator = true and state = 'active'")[0][0]
-    print(f"[INFO] Testing against Trino version \"{trino_version}\"")
+    trino_version = run_query(
+        connection,
+        "select node_version from system.runtime.nodes where coordinator = true and state = 'active'",
+    )[0][0]
+    print(f'[INFO] Testing against Trino version "{trino_version}"')
 
-    assert len(trino_version) >= 3
-    assert trino_version.isnumeric()
+    # Strip SDP release suffix from the version string
+    trino_product_version = re.split(r"-stackable", trino_version, maxsplit=1)[0]
+
+    assert len(trino_product_version) >= 3
+    assert trino_product_version.isnumeric()
     assert trino_version == run_query(connection, "select version()")[0][0]
 
-    run_query(connection, "CREATE SCHEMA IF NOT EXISTS hive.minio WITH (location = 's3a://trino/')")
+    run_query(
+        connection,
+        "CREATE SCHEMA IF NOT EXISTS hive.minio WITH (location = 's3a://trino/')",
+    )
 
     run_query(connection, "DROP TABLE IF EXISTS hive.minio.taxi_data")
     run_query(connection, "DROP TABLE IF EXISTS hive.minio.taxi_data_copy")
     run_query(connection, "DROP TABLE IF EXISTS hive.minio.taxi_data_transformed")
     run_query(connection, "DROP TABLE IF EXISTS hive.hdfs.taxi_data_copy")
     run_query(connection, "DROP TABLE IF EXISTS iceberg.minio.taxi_data_copy_iceberg")
 
-    run_query(connection, """
+    run_query(
+        connection,
+        """
 CREATE TABLE IF NOT EXISTS hive.minio.taxi_data (
     vendor_id VARCHAR,
     tpep_pickup_datetime VARCHAR,
@@ -73,13 +91,24 @@ def run_query(connection, query):
     format = 'csv',
     skip_header_line_count = 1
 )
-    """)
-    assert run_query(connection, "SELECT COUNT(*) FROM hive.minio.taxi_data")[0][0] == 5000
-    rows_written = run_query(connection, "CREATE TABLE IF NOT EXISTS hive.minio.taxi_data_copy AS SELECT * FROM hive.minio.taxi_data")[0][0]
+    """,
+    )
+    assert (
+        run_query(connection, "SELECT COUNT(*) FROM hive.minio.taxi_data")[0][0] == 5000
+    )
+    rows_written = run_query(
+        connection,
+        "CREATE TABLE IF NOT EXISTS hive.minio.taxi_data_copy AS SELECT * FROM hive.minio.taxi_data",
+    )[0][0]
     assert rows_written == 5000 or rows_written == 0
-    assert run_query(connection, "SELECT COUNT(*) FROM hive.minio.taxi_data_copy")[0][0] == 5000
+    assert (
+        run_query(connection, "SELECT COUNT(*) FROM hive.minio.taxi_data_copy")[0][0]
+        == 5000
+    )
 
-    rows_written = run_query(connection, """
+    rows_written = run_query(
+        connection,
+        """
 CREATE TABLE IF NOT EXISTS hive.minio.taxi_data_transformed AS
 SELECT
     CAST(vendor_id as BIGINT) as vendor_id,
@@ -89,61 +118,178 @@ def run_query(connection, query):
     CAST(trip_distance as DOUBLE) as trip_distance,
     CAST(ratecode_id as BIGINT) as ratecode_id
 FROM hive.minio.taxi_data
-""")[0][0]
+""",
+    )[0][0]
     assert rows_written == 5000 or rows_written == 0
-    assert run_query(connection, "SELECT COUNT(*) FROM hive.minio.taxi_data_transformed")[0][0] == 5000
+    assert (
+        run_query(connection, "SELECT COUNT(*) FROM hive.minio.taxi_data_transformed")[
+            0
+        ][0]
+        == 5000
+    )
 
     print("[INFO] Testing HDFS")
 
-    run_query(connection, "CREATE SCHEMA IF NOT EXISTS hive.hdfs WITH (location = 'hdfs://hdfs/trino/')")
-    rows_written = run_query(connection, "CREATE TABLE IF NOT EXISTS hive.hdfs.taxi_data_copy AS SELECT * FROM hive.minio.taxi_data")[0][0]
+    run_query(
+        connection,
+        "CREATE SCHEMA IF NOT EXISTS hive.hdfs WITH (location = 'hdfs://hdfs/trino/')",
+    )
+    rows_written = run_query(
+        connection,
+        "CREATE TABLE IF NOT EXISTS hive.hdfs.taxi_data_copy AS SELECT * FROM hive.minio.taxi_data",
+    )[0][0]
     assert rows_written == 5000 or rows_written == 0
-    assert run_query(connection, "SELECT COUNT(*) FROM hive.hdfs.taxi_data_copy")[0][0] == 5000
+    assert (
+        run_query(connection, "SELECT COUNT(*) FROM hive.hdfs.taxi_data_copy")[0][0]
+        == 5000
+    )
 
     print("[INFO] Testing Iceberg")
-    run_query(connection, "DROP TABLE IF EXISTS iceberg.minio.taxi_data_copy_iceberg")  # Clean up table to don't fail an second run
-    assert run_query(connection, """
+    run_query(
+        connection, "DROP TABLE IF EXISTS iceberg.minio.taxi_data_copy_iceberg"
+    )  # Clean up table to don't fail an second run
+    assert (
+        run_query(
+            connection,
+            """
 CREATE TABLE IF NOT EXISTS iceberg.minio.taxi_data_copy_iceberg
 WITH (partitioning = ARRAY['vendor_id', 'passenger_count'], format = 'parquet')
 AS SELECT * FROM hive.minio.taxi_data
-""")[0][0] == 5000
+""",
+        )[0][0]
+        == 5000
+    )
     # Check current count
-    assert run_query(connection, "SELECT COUNT(*) FROM iceberg.minio.taxi_data_copy_iceberg")[0][0] == 5000
-    assert run_query(connection, 'SELECT COUNT(*) FROM iceberg.minio."taxi_data_copy_iceberg$snapshots"')[0][0] == 1
-    assert run_query(connection, 'SELECT COUNT(*) FROM iceberg.minio."taxi_data_copy_iceberg$partitions"')[0][0] == 12
-    assert run_query(connection, 'SELECT COUNT(*) FROM iceberg.minio."taxi_data_copy_iceberg$files"')[0][0] == 12
+    assert (
+        run_query(
+            connection, "SELECT COUNT(*) FROM iceberg.minio.taxi_data_copy_iceberg"
+        )[0][0]
+        == 5000
+    )
+    assert (
+        run_query(
+            connection,
+            'SELECT COUNT(*) FROM iceberg.minio."taxi_data_copy_iceberg$snapshots"',
+        )[0][0]
+        == 1
+    )
+    assert (
+        run_query(
+            connection,
+            'SELECT COUNT(*) FROM iceberg.minio."taxi_data_copy_iceberg$partitions"',
+        )[0][0]
+        == 12
+    )
+    assert (
+        run_query(
+            connection,
+            'SELECT COUNT(*) FROM iceberg.minio."taxi_data_copy_iceberg$files"',
+        )[0][0]
+        == 12
+    )
 
-    assert run_query(connection, "INSERT INTO iceberg.minio.taxi_data_copy_iceberg SELECT * FROM hive.minio.taxi_data")[0][0] == 5000
+    assert (
+        run_query(
+            connection,
+            "INSERT INTO iceberg.minio.taxi_data_copy_iceberg SELECT * FROM hive.minio.taxi_data",
+        )[0][0]
+        == 5000
+    )
 
     # Check current count
-    assert run_query(connection, "SELECT COUNT(*) FROM iceberg.minio.taxi_data_copy_iceberg")[0][0] == 10000
-    assert run_query(connection, 'SELECT COUNT(*) FROM iceberg.minio."taxi_data_copy_iceberg$snapshots"')[0][0] == 2
-    assert run_query(connection, 'SELECT COUNT(*) FROM iceberg.minio."taxi_data_copy_iceberg$partitions"')[0][0] == 12
-    assert run_query(connection, 'SELECT COUNT(*) FROM iceberg.minio."taxi_data_copy_iceberg$files"')[0][0] == 24
+    assert (
+        run_query(
+            connection, "SELECT COUNT(*) FROM iceberg.minio.taxi_data_copy_iceberg"
+        )[0][0]
+        == 10000
+    )
+    assert (
+        run_query(
+            connection,
+            'SELECT COUNT(*) FROM iceberg.minio."taxi_data_copy_iceberg$snapshots"',
+        )[0][0]
+        == 2
+    )
+    assert (
+        run_query(
+            connection,
+            'SELECT COUNT(*) FROM iceberg.minio."taxi_data_copy_iceberg$partitions"',
+        )[0][0]
+        == 12
+    )
+    assert (
+        run_query(
+            connection,
+            'SELECT COUNT(*) FROM iceberg.minio."taxi_data_copy_iceberg$files"',
+        )[0][0]
+        == 24
+    )
 
-    if trino_version == '377':
+    if trino_version == "377":
         # io.trino.spi.TrinoException: This connector [iceberg] does not support versioned tables
-        print("[INFO] Skipping the Iceberg tests reading versioned tables for trino version 377 as it does not support versioned tables")
+        print(
+            "[INFO] Skipping the Iceberg tests reading versioned tables for trino version 377 as it does not support versioned tables"
+        )
     else:
         # Check count for first snapshot
-        first_snapshot = run_query(connection, 'select snapshot_id from iceberg.minio."taxi_data_copy_iceberg$snapshots" order by committed_at limit 1')[0][0]
-        assert run_query(connection, f"SELECT COUNT(*) FROM iceberg.minio.taxi_data_copy_iceberg FOR VERSION AS OF {first_snapshot}")[0][0] == 5000
+        first_snapshot = run_query(
+            connection,
+            'select snapshot_id from iceberg.minio."taxi_data_copy_iceberg$snapshots" order by committed_at limit 1',
+        )[0][0]
+        assert (
+            run_query(
+                connection,
+                f"SELECT COUNT(*) FROM iceberg.minio.taxi_data_copy_iceberg FOR VERSION AS OF {first_snapshot}",
+            )[0][0]
+            == 5000
+        )
 
     # Compact files
-    run_query(connection, "ALTER TABLE iceberg.minio.taxi_data_copy_iceberg EXECUTE optimize")
+    run_query(
+        connection, "ALTER TABLE iceberg.minio.taxi_data_copy_iceberg EXECUTE optimize"
+    )
 
     # Check current count
-    assert run_query(connection, "SELECT COUNT(*) FROM iceberg.minio.taxi_data_copy_iceberg")[0][0] == 10000
-    assert run_query(connection, 'SELECT COUNT(*) FROM iceberg.minio."taxi_data_copy_iceberg$snapshots"')[0][0] == 3
-    assert run_query(connection, 'SELECT COUNT(*) FROM iceberg.minio."taxi_data_copy_iceberg$partitions"')[0][0] == 12
-    assert run_query(connection, 'SELECT COUNT(*) FROM iceberg.minio."taxi_data_copy_iceberg$files"')[0][0] == 12  # Compaction yeah :)
+    assert (
+        run_query(
+            connection, "SELECT COUNT(*) FROM iceberg.minio.taxi_data_copy_iceberg"
+        )[0][0]
+        == 10000
+    )
+    assert (
+        run_query(
+            connection,
+            'SELECT COUNT(*) FROM iceberg.minio."taxi_data_copy_iceberg$snapshots"',
+        )[0][0]
+        == 3
+    )
+    assert (
+        run_query(
+            connection,
+            'SELECT COUNT(*) FROM iceberg.minio."taxi_data_copy_iceberg$partitions"',
+        )[0][0]
+        == 12
+    )
+    assert (
+        run_query(
+            connection,
+            'SELECT COUNT(*) FROM iceberg.minio."taxi_data_copy_iceberg$files"',
+        )[0][0]
+        == 12
+    )  # Compaction yeah :)
 
     # Test could be improved by also testing update and deletes
 
     # Test postgres connection
-    run_query(connection, 'SHOW SCHEMAS IN postgresgeneric')
-    run_query(connection, 'CREATE SCHEMA IF NOT EXISTS postgresgeneric.tpch')
-    run_query(connection, 'CREATE TABLE IF NOT EXISTS postgresgeneric.tpch.nation AS SELECT * FROM tpch.tiny.nation')
-    assert run_query(connection, "SELECT COUNT(*) FROM postgresgeneric.tpch.nation")[0][0] == 25
+    run_query(connection, "SHOW SCHEMAS IN postgresgeneric")
+    run_query(connection, "CREATE SCHEMA IF NOT EXISTS postgresgeneric.tpch")
+    run_query(
+        connection,
+        "CREATE TABLE IF NOT EXISTS postgresgeneric.tpch.nation AS SELECT * FROM tpch.tiny.nation",
+    )
+    assert (
+        run_query(connection, "SELECT COUNT(*) FROM postgresgeneric.tpch.nation")[0][0]
+        == 25
+    )
 
     print("[SUCCESS] All tests in check-s3.py succeeded!")
diff --git a/tests/templates/kuttl/smoke_aws/check-s3.py b/tests/templates/kuttl/smoke_aws/check-s3.py
@@ -2,6 +2,7 @@
 import trino
 import argparse
 import sys
+import re
 
 if not sys.warnoptions:
     import warnings
@@ -57,8 +58,11 @@ def run_query(connection, query):
     )[0][0]
     print(f'[INFO] Testing against Trino version "{trino_version}"')
 
-    assert len(trino_version) >= 3
-    assert trino_version.isnumeric()
+    # Strip SDP release suffix from the version string
+    trino_product_version = re.split(r"-stackable", trino_version, maxsplit=1)[0]
+
+    assert len(trino_product_version) >= 3
+    assert trino_product_version.isnumeric()
     assert trino_version == run_query(connection, "select version()")[0][0]
 
     # WARNING (@NickLarsenNZ): Hard-coded bucket
