Added a fallback mechanism for the reading metadata

[Copilot]

JDBC drivers for some databases (e.g., Apache Hive, Spark SQL) throw SQLFeatureNotSupportedException for standard ResultSetMetaData methods. This prevents reading data from those sources.

Changes

• Fallback handling in getTableColumnsMetadata() (DbType.kt):

  • getColumnName() → getColumnLabel() → "column_N"
  • getTableName() → extract from qualified name → empty string
  • isNullable() → DatabaseMetaData.getColumns() → true (safe default)
  • getColumnTypeName() → "OTHER"
  • getColumnType() → Types.OTHER
  • getColumnClassName() → "java.lang.Object"

• BIGINT type consistency: Force Types.BIGINT to always map to Long::class regardless of javaClassName reported by driver (fixes MariaDB driver reporting Integer for small BIGINT values)

• Extension point documentation: KDoc explains how to override getTableColumnsMetadata() in subclasses for databases where DatabaseMetaData.getColumns() is slow (Teradata, Oracle)

Usage for unsupported databases

// Create custom DbType for unsupported database
object HiveDb : DbType("hive") {
    override val driverClassName = "org.apache.hive.jdbc.HiveDriver"
    // ... implement required abstract methods
}

// Pass custom dbType to read operations
DataFrame.readSqlQuery(connection, "SELECT * FROM table", dbType = HiveDb)

Fixes #1220

---

💡 You can make Copilot smarter by setting up custom instructions, customizing its development environment and configuring Model Context Protocol (MCP) servers. Learn more Copilot coding agent tips in the docs.