Connect DataEase and preview SQL datasets

Start with the shared setup and dataset. The versioned observations below use the MatrixOne build identified in the overview.

Create a MySQL datasource

The tested version was DataEase 2.10.17 with its bundled MySQL Connector/J 8.2.0.

  1. In datasource management, create a MySQL datasource using host/port mode.

  2. Enter MO_HOST, port 6001, database bi_tutorial, username and password.

  3. For the isolated local non-TLS test, the additional JDBC options were:

characterEncoding=UTF-8&connectTimeout=5000&useSSL=false&allowPublicKeyRetrieval=true&zeroDateTimeBehavior=convertToNull

Configure certificate verification for remote use. allowPublicKeyRetrieval=true permits authentication key retrieval and should only be used when appropriate for the trusted deployment; it does not replace TLS. zeroDateTimeBehavior=convertToNull is a driver conversion policy, not a guarantee that source dates are valid.

  1. Validate and save the datasource, then inspect its table list.

  2. Create a SQL dataset and preview the shared total and category queries. Net amount must be 410.000000 overall and 130.000000/280.000000 by category.

  3. Use the daily SQL as a separate dataset for a time-series visualization. Recheck totals before building a chart; the chart/dashboard workflow is a next validation step and was not part of the API coverage below.

API checks and limits

The unmodified desktop-mode server was tested through its actual /de2api routes:

Operation

Route

Important detail

Validate / save datasource

/datasource/validate, /datasource/save

configuration is base64-encoded JSON

Discover tables/views

/datasource/getTables

Supply the saved datasourceId

Preview SQL

/datasetData/previewSql

Supply datasourceId, base64 SQL and isCross: false

For example, the preview request body is:

{
  "datasourceId": "SAVED_DATASOURCE_ID",
  "sql": "BASE64_OF_THE_SQL_QUERY",
  "isCross": false
}

These route paths are relative to /de2api. The preview response can contain only a prefix of the complete result set. Use a full JDBC query or an independently validated export to check complete rows; a matching preview is not proof of completeness.

The saved datasource, table discovery and SQL preview API were verified; the bundled JDBC path also returned complete expected results. Chart dashboards, cross-source execution and browser rendering were not covered. A separate datasource aimed at another test database validated but returned an unexplained invalid connection during preview; fully qualified SQL through the primary datasource worked. Treat that observation as unresolved, not a confirmed engine defect or universal database support claim. Desktop-mode metadata/cache//opt path problems are deployment issues and should be diagnosed separately.