Search Results pji_extraction_util




Overview
Explore in AI Chat

Overview

APPS.PJI_EXTRACTION_UTIL is a utility package within the Oracle E-Business Suite Project Intelligence (PJI) extraction subsystem. It supports the ETRM (Enterprise Technical Reference Manual) extraction architecture by managing the scope, organization-level configuration, and run-time parameters that govern how project data is extracted from Oracle Projects transactional tables into the PJI staging and reporting structures. The package exposes a mix of configuration seed procedures, extraction-scope maintenance routines, and control functions that determine parallelization and batch sizing for extraction batch jobs. Its API classification is UTIL, indicating it is an internal helper package rather than a business-facing API, and it is referenced by four other packages in the ETRM tree. The header comment ($Header: PJIUT02S.pls 120.2) shows the object has been stable since 2005, predating the 12.1.1 and 12.2.2 releases, and is expected to behave identically across both releases.

Key Procedures and Functions

  • UPDATE_EXTR_SCOPE — Refreshes the extraction scope, determining which projects and organizational units are eligible for PJI extraction based on the configured system parameters and project attributes.
  • POPULATE_ORG_EXTR_INFO — Populates organization-level extraction information, seeding the PJI_ORG_EXTR_INFO table for organizations that have not yet been initialized.
  • UPDATE_ORG_EXTR_INFO — Updates existing organization-level extraction records, typically after changes to organization setup or extraction configuration.
  • SEED_PJI_FM_STATS — Seeds the PJI fact-management statistics used for reporting and monitoring of the extraction process.
  • TRUNCATE_PJI_TABLES — Truncates PJI and/or PJP staging tables as part of an extraction refresh cycle. Accepts several control parameters (error buffer, return code, check flag, table-group truncation flags, and an FPM upgrade flag) that allow selective truncation and optional execution of fact-plan-manager upgrade logic.
  • GET_PARALLEL_PROCESSES — A function returning the number of parallel processes to be used during extraction, derived from database parameters (typically V$PARAMETER CPU count) and PJI system configuration.
  • GET_BATCH_SIZE — A function returning the batch size used to partition extraction work across parallel workers. Together with GET_PARALLEL_PROCESSES, it controls the chunking of records processed per worker and is commonly referenced in concurrent-program driver logic.

Tables Accessed

The package reads and writes two categories of tables. Setup and configuration tables include FND_PROFILE_OPTIONS and FND_PROFILE_OPTION_VALUES (profile-option lookups that drive extraction behavior and the GET_BATCH_SIZE / GET_PARALLEL_PROCESSES functions), PJI_SYSTEM_PARAMETERS and PJI_SYSTEM_CONFIG_HIST (PJI system configuration and change history), and V$PARAMETER (used to detect CPU count for parallel process determination). Projects-related tables include PA_IMPLEMENTATIONS_ALL, PA_PROJECTS_ALL, PA_PROJECT_STATUSES, PA_PROJECT_TYPES_ALL, PA_TIME_CAL_NAME, GL_PERIODS, and the PJI extraction-scope table PA_PJI_PROJ_EVENTS_LOG; these provide the project, status, type, calendar, and period context needed to compute extraction scope. PJI staging tables PJI_ORG_EXTR_INFO and PJI_PROJ_EXTR_STATUS store the resulting organization extraction configuration and per-project extraction status. DUAL is used for single-row function evaluation.

Usage Notes

PJI_EXTRACTION_UTIL is not intended for direct end-user invocation. It is called from PJI concurrent programs and from the driver packages that orchestrate the Project Intelligence extraction cycle, including initialization (POPULATE_ORG_EXTR_INFO, SEED_PJI_FM_STATS), scope maintenance (UPDATE_EXTR_SCOPE, UPDATE_ORG_EXTR_INFO), and the truncation/refresh step (TRUNCATE_PJI_TABLES). The control functions GET_PARALLEL_PROCESSES and GET_BATCH_SIZE are typically queried by extraction driver code to size worker pools and record batches before submitting parallel jobs; custom code that needs to mirror PJI batch behavior should call these functions rather than hard-code values. As an APPS-owned utility, it should be invoked from within the APPS schema context. Because the header is stable at version 120.2, behavior is consistent between 12.1.1 and 12.2.2; no known differences apply.