Hi! Some folks around our org have documented var...
# ingestion
g
Hi! Some folks around our org have documented various tables and columns in spreadsheets that can be read as CSV. Each line has enough info to recreate the table and column URNs. What's the best way to load these to DataHub?
Here's what I'm considering for each table description:
Copy code
curl --location --request POST '<http://mydatahubhost.com:8080/aspects?action=ingestProposal>' \
--header 'X-RestLi-Protocol-Version: 2.0.0' \
--header 'Content-Type: application/json' \
--data '{
  "proposal" : {
    "entityType": "dataset",
    "entityUrn" : "urn:li:dataset:(urn:li:dataPlatform:bigquery,my-gcp-project.my_dataset.some_table,PROD)",
    "changeType" : "UPSERT",
    "aspectName" : "editableDatasetProperties",
    "aspect" : {
      "value": "{\"description\":\"Foo bar baz\"}",
      "contentType": "application/json"
    }
  }
}'
Is this along the right lines? Is there a similar POST request to update column descriptions?
g
That looks correct @gorgeous-diamond-82312 ^ For column descriptions, you can refer to the
updateDescription
mutation. https://github.com/acryldata/datahub-fork/blob/acryl-main/datahub-graphql-core/src/main/resources/entity.graphql#L206
🙇 1
we have not yet made
updateDescription
support entity-level description updates