berkeley: API endpoint /data_objects/study/{study_id} not returning correct results

[aclum]

Describe the bug
This endpoint is returning an empty array b/c alldocs doesn't exist in berkeley

To Reproduce
Steps to reproduce the behavior:

1. curl -X 'GET'
   'https://api-berkeley.microbiomedata.org/data_objects/study/nmdc%3Asty-11-33fbta56'
   -H 'accept: application/json'
2. compare empty return body to results on production

Expected behavior
Response json body should contain the same records as production.

Example user story template:
AS A {user|persona|system},
[INSTEAD OF {current condition}]
I EXPECTED {result}
[SO THAT {value or justification}]
[NO LATER THAN {best by date}]

Screenshots
If applicable, add screenshots to help explain your problem.

Acceptance Criteria
Response body is ~51K in size.

Example scenario-based template:
Given (some given context or precondition), when (I take this action), then (this will be the specific outcome).

Additional context
Consider adding test coverage for this endpoint.

cc @sujaypatil96 because ncbi export code also needs alldocs

[eecavanna]

Once #694 has been merged in, I expect this bug to also be squashed.

[eecavanna]

On second thought, I think—even with the alldocs collection present—there is still a bug here specific to when using the Berkeley schema.

The endpoint code uses the Biosample's part_of field to identify biosamplies associated with the specified Study. In the Berkeley schema, there is no part_of slot on the Biosample class. There is an associated_studies field, though. CC: @sujaypatil96

[eecavanna]

Locally, updating the endpoint's code to use associated_studies made the difference between a result of [] and a result whose content-length is 42452.

-     biosamples = mdb.biosample_set.find({"part_of": study["id"]}, ["id"])
+     biosamples = mdb.biosample_set.find({"associated_studies": study["id"]}, ["id"])

Note: This is with dump /global/cfs/projectdirs/m3408/nmdc-mongodumps/dump_nmdc-berkeley_2024-09-19_20-20-01 loaded in my local Mongo server.

[eecavanna]

Now that #694 has been merged in, I'll open a PR having the above code change in it.

[ssarrafan]

@sujaypatil96 @eecavanna who needs to review this?

[eecavanna]

I merged it in without review (other than self-review) just now in the interest of time.

I would still like for @sujaypatil96 to review it when he gets a chance—and he can open a ticket (or message me on Slack) in case he has concerns about its contents. It is not urgent, from my perspective.

[eecavanna]

With the endpoint fix in place, and with the alldocs collection in place, the endpoint now returns a non-empty response:

The response (when downloaded as a file via Swagger UI) is 55 KB, which (in this context) I consider to be "in the ballpark" of the 51 KB @aclum put in the issue description. For example, maybe some workflows have added more data objects. 🤷

[aclum]

Confirmed fixed in berkeley