[AL-4418] [AL-3976] Make dataset tests parallelizable #831
Conversation
apollonin
left a comment
apollonin
left a comment
There was a problem hiding this comment.
this is a great start! Do you want to merge these changes right away?
| dataset.update(description=description) | ||
| assert dataset.description == description | ||
|
|
||
| dataset.delete() |
There was a problem hiding this comment.
It would be good to add a check for the deleted dataset here by querying for the dataset
with pytest.raises(SomeException):
client.get_dataset(dataset.uid)
There was a problem hiding this comment.
This already exists on line 44.
| from copy import copy | ||
|
|
||
|
|
||
| def test_get_one_and_many_dataset_order(client): |
There was a problem hiding this comment.
Nice work adding coverage for pagination!
There was a problem hiding this comment.
I felt that part of the test_dataset was really testing pagination so I moved that here.
whistler
changed the title
whistler
changed the title
| def test_timeout_error(client, project): | ||
| with pytest.raises(labelbox.exceptions.TimeoutError) as excinfo: | ||
| query_str = """query getOntology { | ||
| project (where: {id: $%s}) { |
There was a problem hiding this comment.
maybe it timed out before this got evaluated.
| CAPTURE_DT_SCHEMA_ID = "cko8sdzv70006h2dk8jg64zvb" | ||
| PRE_COMPUTED_EMBEDDINGS_ID = 'ckrzang79000008l6hb5s6za1' | ||
| CUSTOM_TEXT_SCHEMA_NAME = 'custom_text' | ||
| CUSTOM_TEXT_SCHEMA_NAME = 'datarow_metadata_custom_text' |
There was a problem hiding this comment.
I wonder if we want to use random schema name here for each test?
There was a problem hiding this comment.
I think this change is not needed anymore so I'll revert it, before we were creating and deleting this for each test, but that doesn't work with parallel tests.
| assert project.organization() == org | ||
| assert set(user.projects()) == user_projects.union({project}) | ||
| assert set(org.projects()) == org_projects.union({project}) | ||
| assert project in user_projects |
There was a problem hiding this comment.
great catch! It has been intermittently failing recently on staging and with your change now should be fine!
whistler
force-pushed
the
imuhammad/AL-3976-parallelizable-dataset-tests
branch
from
1f6c81d to
6177b59
Compare
Remove the assumption that only the current test is modifying data stores for dataset integration tests. Tested on staging.
