Search Results dbms_hadoop_internal




Overview
Explore in AI Chat

Overview

SYS.DBMS_HADOOP_INTERNAL is an internal Oracle-supplied PL/SQL package that provides the low-level infrastructure supporting Oracle's Big Data and Hadoop integration stack. Declared with AUTHID DEFINER and intended to be accessible only by the public-facing DBMS_HADOOP package, it is not a documented EBS application API and is not part of the Oracle E-Business Suite application schema. In the EBS 12.1.1 and 12.2.2 environments it appears in the database data dictionary because Oracle Database 12c and later ship it as a standard component of the RDBMS, and EBS instances inherit it through the underlying database home.

The package serves two broad business functions. First, it encapsulates security and privilege verification logic that determines whether the current database user may read from or write to configuration and debug directories associated with Hadoop clusters. Second, it provides data conversion and pipelined table functions that relay rows from external C procedures and Hive metastores into PL/SQL. The user search term "get_ddl" corresponds to the GET_DDL function, which returns DDL text for Hadoop-related objects managed through this integration layer.

Key Procedures and Functions

Nine documented program units make up the package interface:

  • UNIX_TS_TO_DATE — Ancillary conversion function that transforms a Julian date/time value into a calendar date.
  • DEBUG_USER_PRIVILEGED — Determines whether a named user holds WRITE privileges on the ORACLE_BIGDATA_DEBUG directory; returns 1 when privileged, 0 otherwise.
  • GETNUMBEROFITEMS — Parses a CLOB using a supplied boundary key and returns the count of items found, supporting structured parsing of external payloads.
  • USER_PRIVILEGED — Verifies READ privilege on a specified Hadoop cluster for a given user, requiring READ permission on the ORACLE_BIGDATA_CONFIG directory; returns 1 or 0.
  • GET_DEBUG_DIR — Returns the debug directory path applicable to the current user.
  • GET_CONFIG_DIR — Returns the configuration directory path used by the Big Data integration.
  • GETHIVETABLE — A pipelined table function that returns rows from C external procedures via ODCI into PL/SQL, with rows originating from Hive metastores.
  • GET_DDL — Returns DDL text for Hadoop or Hive objects registered in the Big Data metadata layer.
  • GET_NAME — Returns the associated object name used during metadata resolution.

Tables Accessed

The ETRM metadata records no tables accessed through APPS synonyms for this package. This is consistent with its nature: DBMS_HADOOP_INTERNAL operates on directory objects, external procedures, and Hive metastore data rather than on EBS application tables. Its GET_DDL and GETHIVETABLE functions obtain metadata from external sources rather than from EBS base tables.

Usage Notes

DBMS_HADOOP_INTERNAL is invoked indirectly. Application code and the public DBMS_HADOOP package call its functions; direct invocation from EBS forms, concurrent programs, or custom code is neither supported nor advisable given the ACCESSIBLE BY restriction. In EBS 12.1.1 and 12.2.2, the package is relevant only where Hadoop or Big Data integrations are configured. Ten other packages are documented as referencing it, indicating it sits beneath a layered API surface. Its primary practical relevance to EBS administrators is diagnostic, particularly through GET_DDL, GET_CONFIG_DIR, and the privilege-check functions.