Preview the rows a data source config would return
Fetches the source described by the request without persisting anything and returns a sample of rows with inferred columns, the JSON path or sheet that was used, and a warning when the source is empty. Requires `data_sources:write` rather than read, because a preview exercises stored connection secrets against a caller-supplied URL; HTTP sources are restricted to public addresses, and fetch failures come back as validation errors with secret values redacted. Sample size is capped at 50 rows (default 20).
Authorization
apiKey Workspace API key created in Workspace settings → API keys. Scopes on the key bound what it can read and write.
In: header
Path Parameters
Request Body
application/json
TypeScript Definitions
Use the request body type in TypeScript.
Response Body
application/json
application/json
application/json
curl -X POST "https://example.com/v1/workspaces/string/data-sources/preview" \ -H "Content-Type: application/json" \ -d '{ "kind": "http" }'{ "data": { "rows": [ {} ], "columns": [ { "sourcePath": "string", "suggestedLabel": "string", "suggestedKind": "text", "samples": [] } ], "totalRowsSampled": 0, "rowsPathUsed": "string", "sheetNames": [ "string" ], "sheetNameUsed": "string", "warning": "string" }}{ "_tag": "ApiNotFound", "error": "not_found", "message": "That item does not exist, or this key cannot see it.", "status": 404}{ "_tag": "ApiNotFound", "error": "not_found", "message": "That item does not exist, or this key cannot see it.", "status": 404}Create a data source
Registers a data source, provisions the database it syncs into, and files a workspace item for it under the requested parent. Requires `data_sources:write` (plus edit access to the parent item when one is given); a label outside 1 to 120 characters, an invalid config for the kind, an out-of-range refresh interval, or an unknown `connectionId` are rejected before anything persists. Send an `Idempotency-Key` header to make retries safe; a replay within the window returns the original response.
Import a spreadsheet into a new database
Parses an uploaded or provider-hosted spreadsheet once and creates a database with the mapped fields and rows, returning the created database with its field, record, and skipped-row counts. Requires `databases:write`, plus `files:read` when importing an uploaded file. Malformed or oversized workbooks are rejected as validation errors; passing a `clientMutationId` makes the import replay-safe, and reusing one while its import is still running, or with different input, is reported as an in-progress or conflicting-mutation error. Send an `Idempotency-Key` header to make retries safe; a replay within the window returns the original response.