generated: '2026-07-18'
method: searched
source: https://docs.e6data.com/product-documentation
docs:
  connectivity: https://docs.e6data.com/product-documentation/connectivity
  query_count_api: https://docs.e6data.com/product-documentation/query-history/query-count-api
  authentication: authentication/e6data-authentication.yml
  errors: errors/e6data-error-codes.yml
authentication:
  style: bearer-token
  header: Authorization
  format: "Bearer <ACCESS_TOKEN>"
  detail: Personal Access Token (PAT) or Service Account key; see authentication/e6data-authentication.yml
primary_interface:
  type: SQL
  transports:
    - JDBC (type 4 driver, jdbc:e6data://... , class io.e6.jdbc.driver.E6Driver, aligned to e6data API v1.0)
    - Python connector (thrift-based, PyPI e6data-python-connector)
    - BI/SQL clients (DBeaver, DbVisualizer, Apache Superset, Tableau, Power BI, Metabase, Zeppelin, Jupyter)
  detail: >-
    e6data is a lakehouse SQL compute engine; the main programmatic surface is
    SQL over JDBC / Python connector against a per-cluster endpoint, not a broad
    REST resource API. Connection requires cluster ingress to be enabled and a
    cluster UUID, hostname, port, catalog (Glue/Hive/Unity) and database.
rest_api:
  present: true
  scope: narrow (management/reporting endpoints, e.g. query-history query count)
  base_url_pattern: "https://<subdomain>.e6data.com/endpoints/v1/"
  versioning: uri-path (v1)
  content_type: application/json
  example_endpoint:
    name: Query Count
    method: POST
    path: /endpoints/v1/query-history/get-query-count/
    query_params: [from_date, to_date]
    auth: "Authorization: Bearer <ACCESS_TOKEN>"
    request_fields: [workspace_name, run_by, cluster_name]
    response_envelope: "{ result: [ { status, count } ] }"
pagination:
  documented: partial
  detail: >-
    The Query Count API returns the latest 100 records by default when no
    start/end date is supplied; no cursor/offset scheme is documented for the
    narrow REST surface. SQL result paging is handled by the JDBC/Python
    ResultSet (forward-only, read-only; cursor movement not supported by JDBC).
idempotency:
  documented: false
  detail: No idempotency-key contract is published; queries are the unit of work over SQL.
error_handling:
  rest_envelope: "JSON result payloads"
  coded_registry: errors/e6data-error-codes.yml (QE-prefixed codes)
  client_behavior: JDBC surfaces failures as java.sql.SQLException; connector raises exceptions.
time:
  timezone: UTC
  detail: >-
    REST date params (from_date/to_date) are UTC in 'YYYY-MM-DD HH:MM:SS' form;
    JDBC timestamp query values must be supplied in UTC.
rate_limiting:
  documented: false
  network_controls: [IP allowlisting / IP sets, network firewall, cluster ingress, PrivateLink/VPC peering]