Data Model Metadata (Preview)

oracle_analytics-describe_data

Input Schema

{
  "$schema": "https://json-schema.org/draft/2020-12/schema",
  "type": "object",
  "properties": {
    "datamodelName": {
      "type": "string"
    },
    "tableName": {
      "type": "string"
    },
    "tableNames": {
      "type": "array",
      "items": {
        "type": "string"
      }
    },
    "tablesOnly": {
      "type": "boolean"
    }
  }
}

Retrieves table and column metadata for OAC subject areas and datasets. This is the primary tool for validating column names, data types, measure classifications, and aggregation behavior before writing Logical SQL or generating workbooks.

Import Parameters

Parameter	Type	Notes
datamodelName	string	Subject area name, such as Sales History Subject Area, or an XSA expression such as XSA('owner'.'dataset').
tablesOnly	boolean	Returns only available table names. Use this before requesting specific tables.
tableName	string	Returns columns for one table. Use the table name only, not the fully qualified name.
tableNames	string[]	Returns columns for multiple tables in one call, when supported.

Example - List Tables Only

{
  "jsonrpc": "2.0",
  "id": 30,
  "method": "tools/call",
  "params": {
    "name": "oracle_analytics-describe_data",
    "arguments": {
      "datamodelName": "Sales History Subject Area",
      "tablesOnly": true
    }
  }
}

Example - Describe One Table

{
  "jsonrpc": "2.0",
  "id": 31,
  "method": "tools/call",
  "params": {
    "name": "oracle_analytics-describe_data",
    "arguments": {
      "datamodelName": "Sales History Subject Area",
      "tableName": "SALES"
    }
  }
}

Example - Describe Multiple Tables in One Request

{
  "jsonrpc": "2.0",
  "id": 32,
  "method": "tools/call",
  "params": {
    "name": "oracle_analytics-describe_data",
    "arguments": {
      "datamodelName": "Sales History Subject Area",
      "tableNames": ["SALES", "TIMES", "CHANNELS"]
    }
  }
}

Typical Response Fields

Field	Notes
subjectArea	Data model name.
tables[].tableName	Table name.
tables[].fullQualifiedName	Fully quoted subject-area/table reference.
tables[].columns[]	Column metadata.
columns[].name	Column name.
columns[].displayName	Display name.
columns[].fullyQualifiedName	Fully quoted column reference for Logical SQL.
columns[].columnType	Usually attribute or measure.
columns[].dataType	Data type such as VARCHAR, NUMERIC, TIMESTAMP, DATE.
columns[].aggregation	Default aggregation where provided.
columns[].nullable	Nullability flag.
columns[].precision, columns[].scale, columns[].length	Numeric/string metadata where available.

Practical Guidance

• Start with tablesOnly: true to avoid very large responses.
• Prefer tableName or tableNames once the relevant tables are known.
• Use fullyQualifiedName directly in Logical SQL.
• Check columnType before aggregating. Measures often already have model-defined aggregation.
• Don't assume measures should be wrapped in SUM; inspect metadata first.