Convert python_wheel_test.go to acceptance test#2471
Merged
Conversation
denik
added a commit
that referenced
this pull request
Mar 11, 2025
## Changes New bool config setting CloudSlow in acceptance tests. If set, it enables this test on Cloud but skips it in -short setting there. It does not affect local runs, "Slow" is only applied to Cloud. The reason is that we won't need to wait for cluster with mocked testserver. Additionally, this setting enables -tail if -v is already enabled. ## Why Certain tests that use "bundle run" are too long due to starting a cluster. ## Tests Using this option in #2471
add repl for TEST_INSTANCE_POOL_ID add uuid1 env/repl with dashes replaced with underscore - WIP - use randomString rm out.requests.txt update test fix the test mark my test as CloudLong update test.toml disable local runs split test in two for parallel run rename
31c7252 to
2b5daea
Compare
This was referenced Mar 11, 2025
denik
commented
Mar 12, 2025
| } | ||
|
|
||
| func TestPythonWheelTaskDeployAndRunWithWrapper(t *testing.T) { | ||
| runPythonWheelTest(t, "python_wheel_task", "12.2.x-scala2.12", true) |
Contributor
Author
There was a problem hiding this comment.
@andrewnester why was this Spark version used in this test?
Contributor
There was a problem hiding this comment.
Installing wheels from Workspace file system is only supported starting from DBR 13.1+. But before users used older DBRs and python wheel tasks but installed it from DBFS. We still want to support older DBRs and did the trampoline workaround. Hence this is to test that python wheel tasks in DABs are working for older DBRs
Contributor
Author
There was a problem hiding this comment.
Thanks! Added you comment into the test 060b8f5
denik
commented
Mar 12, 2025
| import ( | ||
| "bufio" | ||
| "context" | ||
| "encoding/base32" |
Contributor
Author
There was a problem hiding this comment.
all changes in acceptance_test.go will disappear after linked PRs are merged and this one is rebased.
andrewnester
approved these changes
Mar 12, 2025
github-merge-queue Bot
pushed a commit
that referenced
this pull request
Mar 12, 2025
Useful for integration tests where project names must be unique. I'm using it in #2471
denik
added a commit
that referenced
this pull request
Mar 12, 2025
Follow #2471 to where incorrect config is used. Modify all tests in integration_whl to print rendered databricks.yml Add $NODE_TYPE_ID env var.
denik
added a commit
that referenced
this pull request
Mar 12, 2025
Follow #2471 to where incorrect config is used. Modify all tests in integration_whl to print rendered databricks.yml Add $NODE_TYPE_ID env var.
github-merge-queue Bot
pushed a commit
that referenced
this pull request
Mar 12, 2025
…E_ID (#2478) This is a follow up to #2471 where incorrect config was used in interactive_cluster test. ## Changes - Fixed interactive_cluster to use proper config, it was accidentally referring to config from ../base - Add $NODE_TYPE_ID env var and replacement to acceptance tests, this is necessary for interactive_cluster test. - Disable acceptance/bundle/override on cloud. This started failing because it has real node type that gets replaced with NODE_TYPE_ID but only in AWS env. Since the test is focussed on config merging, there is no need to run it against real workspaces. - Modify all tests in integration_whl to print rendered databricks.yml, to prevent this kind of error.
denik
added a commit
that referenced
this pull request
Mar 18, 2025
When this tests were integration tests, they did not do run() part in -short mode. After converting them to acceptance I kept them on always in #2471 Disabling them in short mode because it seems they do expand PR integration test time.
github-merge-queue Bot
pushed a commit
that referenced
this pull request
Mar 18, 2025
## Why When this tests were integration tests, they did not do run() part in -short mode. After converting them to acceptance I kept them on always in #2471 This PR disables them in short mode (they will still run on main) because it seems they do expand PR integration test time.
denik
added a commit
that referenced
this pull request
May 20, 2026
## Changes New bool config setting CloudSlow in acceptance tests. If set, it enables this test on Cloud but skips it in -short setting there. It does not affect local runs, "Slow" is only applied to Cloud. The reason is that we won't need to wait for cluster with mocked testserver. Additionally, this setting enables -tail if -v is already enabled. ## Why Certain tests that use "bundle run" are too long due to starting a cluster. ## Tests Using this option in #2471
denik
added a commit
that referenced
this pull request
May 20, 2026
Discovered that 'trap' does not work with $CLI on Windows when working #2471 This example shows a workaround.
denik
added a commit
that referenced
this pull request
May 20, 2026
denik
added a commit
that referenced
this pull request
May 20, 2026
Useful for integration tests where project names must be unique. I'm using it in #2471
denik
added a commit
that referenced
this pull request
May 20, 2026
Convert integration/bundle/integration/bundle/python_wheel_test.go to acceptance tests. I plan to expand these tests to check patchwheel functionality. Inside each test there were two runs - with params and without, I've expanded each run into separate test to reduce total time as this runs can be done in parallel. Also add new env var DEFAULT_SPARK_VERSION that matches the one in integration tests. The tests are currently enabled on every PR (`CloudLong=true` is commented out), this can be changed after landing.
denik
added a commit
that referenced
this pull request
May 20, 2026
…E_ID (#2478) This is a follow up to #2471 where incorrect config was used in interactive_cluster test. ## Changes - Fixed interactive_cluster to use proper config, it was accidentally referring to config from ../base - Add $NODE_TYPE_ID env var and replacement to acceptance tests, this is necessary for interactive_cluster test. - Disable acceptance/bundle/override on cloud. This started failing because it has real node type that gets replaced with NODE_TYPE_ID but only in AWS env. Since the test is focussed on config merging, there is no need to run it against real workspaces. - Modify all tests in integration_whl to print rendered databricks.yml, to prevent this kind of error.
denik
added a commit
that referenced
this pull request
May 20, 2026
## Why When this tests were integration tests, they did not do run() part in -short mode. After converting them to acceptance I kept them on always in #2471 This PR disables them in short mode (they will still run on main) because it seems they do expand PR integration test time.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Convert integration/bundle/integration/bundle/python_wheel_test.go to acceptance tests. I plan to expand these tests to check patchwheel functionality.
Inside each test there were two runs - with params and without, I've expanded each run into separate test to reduce total time as this runs can be done in parallel.
Also add new env var DEFAULT_SPARK_VERSION that matches the one in integration tests.
The tests are currently enabled on every PR (
CloudLong=trueis commented out), this can be changed after landing.