Added E2E test to run Python wheels on interactive cluster created in bundle (#1864)

## Changes
Added E2E test to run python wheels on interactive cluster created in
bundle.

We had a gap in testing wheel on all purpose clusters, so this PR
addresses the gap
This commit is contained in:
Andrew Nester 2024-11-01 15:22:47 +01:00 committed by GitHub
parent f3bf33da27
commit 71cf426755
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
6 changed files with 102 additions and 4 deletions

View File

@ -0,0 +1,25 @@
{
"properties": {
"project_name": {
"type": "string",
"default": "my_test_code",
"description": "Unique name for this project"
},
"spark_version": {
"type": "string",
"description": "Spark version used for job cluster"
},
"node_type_id": {
"type": "string",
"description": "Node type id for job cluster"
},
"unique_id": {
"type": "string",
"description": "Unique ID for job name"
},
"instance_pool_id": {
"type": "string",
"description": "Instance pool id for job cluster"
}
}
}

View File

@ -0,0 +1,29 @@
bundle:
name: wheel-task
workspace:
root_path: "~/.bundle/{{.unique_id}}"
resources:
clusters:
test_cluster:
cluster_name: "test-cluster-{{.unique_id}}"
spark_version: "{{.spark_version}}"
node_type_id: "{{.node_type_id}}"
num_workers: 1
data_security_mode: USER_ISOLATION
jobs:
some_other_job:
name: "[${bundle.target}] Test Wheel Job {{.unique_id}}"
tasks:
- task_key: TestTask
existing_cluster_id: "${resources.clusters.test_cluster.cluster_id}"
python_wheel_task:
package_name: my_test_code
entry_point: run
parameters:
- "one"
- "two"
libraries:
- whl: ./dist/*.whl

View File

@ -0,0 +1,15 @@
from setuptools import setup, find_packages
import {{.project_name}}
setup(
name="{{.project_name}}",
version={{.project_name}}.__version__,
author={{.project_name}}.__author__,
url="https://databricks.com",
author_email="john.doe@databricks.com",
description="my example wheel",
packages=find_packages(include=["{{.project_name}}"]),
entry_points={"group1": "run={{.project_name}}.__main__:main"},
install_requires=["setuptools"],
)

View File

@ -0,0 +1,2 @@
__version__ = "0.0.1"
__author__ = "Databricks"

View File

@ -0,0 +1,16 @@
"""
The entry point of the Python Wheel
"""
import sys
def main():
# This method will print the provided arguments
print("Hello from my func")
print("Got arguments:")
print(sys.argv)
if __name__ == "__main__":
main()

View File

@ -5,17 +5,18 @@ import (
"github.com/databricks/cli/internal"
"github.com/databricks/cli/internal/acc"
"github.com/databricks/cli/internal/testutil"
"github.com/databricks/cli/libs/env"
"github.com/google/uuid"
"github.com/stretchr/testify/require"
)
func runPythonWheelTest(t *testing.T, sparkVersion string, pythonWheelWrapper bool) {
func runPythonWheelTest(t *testing.T, templateName string, sparkVersion string, pythonWheelWrapper bool) {
ctx, _ := acc.WorkspaceTest(t)
nodeTypeId := internal.GetNodeTypeId(env.Get(ctx, "CLOUD_ENV"))
instancePoolId := env.Get(ctx, "TEST_INSTANCE_POOL_ID")
bundleRoot, err := initTestTemplate(t, ctx, "python_wheel_task", map[string]any{
bundleRoot, err := initTestTemplate(t, ctx, templateName, map[string]any{
"node_type_id": nodeTypeId,
"unique_id": uuid.New().String(),
"spark_version": sparkVersion,
@ -45,9 +46,19 @@ func runPythonWheelTest(t *testing.T, sparkVersion string, pythonWheelWrapper bo
}
func TestAccPythonWheelTaskDeployAndRunWithoutWrapper(t *testing.T) {
runPythonWheelTest(t, "13.3.x-snapshot-scala2.12", false)
runPythonWheelTest(t, "python_wheel_task", "13.3.x-snapshot-scala2.12", false)
}
func TestAccPythonWheelTaskDeployAndRunWithWrapper(t *testing.T) {
runPythonWheelTest(t, "12.2.x-scala2.12", true)
runPythonWheelTest(t, "python_wheel_task", "12.2.x-scala2.12", true)
}
func TestAccPythonWheelTaskDeployAndRunOnInteractiveCluster(t *testing.T) {
_, wt := acc.WorkspaceTest(t)
if testutil.IsAWSCloud(wt.T) {
t.Skip("Skipping test for AWS cloud because it is not permitted to create clusters")
}
runPythonWheelTest(t, "python_wheel_task_with_cluster", defaultSparkVersion, false)
}