Closed bruno-f-cruz closed 2 months ago
Not sure what is going on with the CD/CI but tests are passing on my local machine with a clean environment. Can someone else test it?
It looks like the acq datetime is off a little:
ManifestConfig(name='test', processor_full_name='na', subject_id=7, acquisition_datetime=datetime.datetime(2024, 9, 3, 9, 18, 29, 186489), schedule_time=None, force_cloud_sync=False, transfer_endpoint='http://aind-data-transfer-service/api/v1/submit_jobs', platform='behavior', capsule_id=None, mount=None, s3_bucket=<BucketType.PRIVATE: 'private'>, project_name='no project', destination='path', modalities={'behavior': ['path']}, schemas=['path'], script={}) != ManifestConfig(name='test', processor_full_name='na', subject_id=7, acquisition_datetime=datetime.datetime(2024, 9, 3, 9, 18, 29, 186436), schedule_time=None, force_cloud_sync=False, transfer_endpoint='http://aind-data-transfer-service/api/v1/submit_jobs', platform='behavior', capsule_id=None, mount=None, s3_bucket=<BucketType.PRIVATE: 'private'>, project_name='no project', destination='path', modalities={'behavior': ['path']}, schemas=['path'], script={})
Expected :ManifestConfig(name='test', processor_full_name='na', subject_id=7, acquisition_datetime=datetime.datetime(2024, 9, 3, 9, 18, 29, 186436), schedule_time=None, force_cloud_sync=False, transfer_endpoint='http://aind-data-transfer-service/api/v1/submit_ ...
Actual :ManifestConfig(name='test', processor_full_name='na', subject_id=7, acquisition_datetime=datetime.datetime(2024, 9, 3, 9, 18, 29, 186489), schedule_time=None, force_cloud_sync=False, transfer_endpoint='http://aind-data-transfer-service/api/v1/submit_ ...
Yep that's what I said on the PR. I just can't see a wait to fix it without fixing it upstream like I have been asking for. The tests I added with this PR showcases the use case. You can basically enter it as an AIND-data-schema-model object.
Got it, I misunderstood that.
Closes #58
Unfortunately, because these classes are implemented upstream it's hard to provide proper type hints and still ensure proper schema compilation. This is the best I could think of for now.