Add new grpc endpoint to get genome id from accession id #63
Conversation
app/api/resources/grpc_client.py
Outdated
| @@ -148,3 +149,12 @@ def get_dataset_attributes( | |||
| genome_uuid=genome_uuid, dataset_type=dataset_type | |||
| ) | |||
| return self.stub.GetAttributesValuesByUUID(dataset_attributes) | |||
|
|
|||
| def get_genome_by_specific_keyword(self, assembly_accession_id: str): | |||
There was a problem hiding this comment.
It seems that you are looking up genome id specifically by assembly accession id. If so, you probably shouldn't say by_specific_keyword in the function name, and in the class name in models/genome.py.
Both thoas and the grpc service allow various fields to be used as a "keyword"; but here you are only using a specific one.
app/api/resources/metadata.py
Outdated
| @@ -238,3 +239,19 @@ async def get_genome_dataset_attributes( | |||
| except Exception as ex: | |||
| logging.error(ex) | |||
| return response_error_handler({"status": 500}) | |||
|
|
|||
| @router.get("/genomeid/{assembly_accession_id}", name="genome_id") | |||
There was a problem hiding this comment.
This does not look like a great pathname. Something like /assembly/{assembly_accession_id}/genome_id/ would be closer to what is happening here. We may want to discuss this collectively.
app/api/resources/metadata.py
Outdated
| @@ -238,3 +239,21 @@ async def get_genome_dataset_attributes( | |||
| except Exception as ex: | |||
| logging.error(ex) | |||
| return response_error_handler({"status": 500}) | |||
|
|
|||
| @router.get("/genome") | |||
There was a problem hiding this comment.
Didn't we say that since genome in other routes has a different shape, this route should be called something other than genome?
app/api/resources/metadata.py
Outdated
|
|
||
| @router.get("/genomeid") | ||
| async def get_genome_by_keyword(request: Request, assembly_accession_id: str): | ||
| if (assembly_accession_id == ""): |
There was a problem hiding this comment.
What if assembly accession ID doesn’t exist in the response?
app/api/resources/metadata.py
Outdated
| if (genome_by_keyword_object.release_version > latest_genome_by_keyword_object.release_version): | ||
| latest_genome_by_keyword_object = genome_by_keyword_object | ||
| if (latest_genome_by_keyword_object.genome_uuid): | ||
| return latest_genome_by_keyword_object |
There was a problem hiding this comment.
Shouldn't it return a response (e.g. responses.JSONResponse) instead of a pydantic model instance?
There was a problem hiding this comment.
I was doing it at the resolver end. Makes sense to do it here. Will update.
APISpecification.yaml
Outdated
| @@ -283,10 +283,41 @@ paths: | |||
| example: 'Could not find details for genome 0cbc227a-7179-43c2-8061-a50d94423c4d and dataset homologies.' | |||
| '500': | |||
| $ref: '#/components/responses/500InternalServerError' | |||
| /api/metadata/genomeid/: | |||
| get: | |||
| summary: Returns a list of genome uuids and its release version | |||
There was a problem hiding this comment.
The code returns a single GenomeIDObject, OpenAPI spec expects a list of objects. Which one is correct?
There was a problem hiding this comment.
Ah true. It sorts and returns one. Good point. Thanks
Description
To resolve rapid urls with GCA accession, we need an endpoint to fetch this detail from grpc. After consulting with Bilal, we decided to use
GenomeBySpecificKeywordRequest. This grpc endpoint returns an array of results from which we are picking up genome id and the release version. Release version is used in resolver to sort and find the latest genome idRelated JIRA Issue(s)
https://www.ebi.ac.uk/panda/jira/browse/ENSWBSITES-2830
Review App URL(s)
NA
Knowledge Base
NA
Checklist
Dependencies
No