Search Results insert_expression_key

Overview

GMD_EXPRESSION_MIG_UTIL is an Oracle EBS Process Manufacturing (OPM/GMD) utility package owned by APPS and classified as a UTIL object. Its business function is to support the migration, parsing, tokenization, storage, and evaluation of technical parameter expressions used in product development and quality specifications. Within Oracle Process Manufacturing, technical parameters — such as assay targets, specification limits, and calculated quality attributes — are frequently defined using algebraic expressions that combine numeric literals, operators, and named parameters. This package provides the underlying engine that decomposes such expression strings into structured rows stored in the parsed-expression table, assigns expression keys, and evaluates the resulting values against a given entity or line context. It is a helper infrastructure package rather than a business-facing API, and it is designed to be called by higher-level OPM specification and formula programs that need to interpret user-defined expressions during data migration or runtime processing.

Key Procedures and Functions

The package exposes nine documented subprograms that together perform expression tokenization, classification, persistence, and evaluation.

• PARSE_EXPRESSION — Parses an incoming expression string for a given organization and technical parameter, decomposing it into tokens and driving the subsequent classification and storage steps.
• INSERT_EXPRESSION_KEY — Inserts an expression key for a specified organization and technical parameter, anchoring the parsed token rows to a unique expression identifier. This is the subprogram users most commonly search for when debugging expression-key insertion behavior.
• IS_OPERATOR — Boolean classifier that determines whether a supplied token represents a supported arithmetic or logical operator.
• IS_PARAMETER — Boolean classifier that determines whether a supplied token corresponds to a valid parameter for the organization, returning the resolved parameter identifier and data type.
• IS_NUMBER — Boolean classifier that determines whether a supplied token is a numeric literal.
• ADD_EXPRESSION_ROW — Persists an individual parsed expression row, capturing its key, type, data type, and associated expression parameter identifier.
• EVALUATE_EXPRESSION_VALUE — Evaluates a single expression line against an in-memory expression table and returns the computed value.
• GET_VALUE — Retrieves the value of a named parameter for a given line, supporting the evaluation of referenced operands.
• EVALUATE_EXPRESSION — Evaluates a complete expression for an entity and line, returning both the computed numeric value and the resolved expression text.

Tables Accessed

The package interacts with several OPM tables through APPS synonyms. GMD_PARSED_EXPRESSION is the primary persistent store for tokenized expression rows written by ADD_EXPRESSION_ROW and read during evaluation. GMD_TECH_PARAMETERS_B supplies the technical parameter definitions against which token parameters are validated by IS_PARAMETER. GMD_TECH_EXP_SEQ_ID_S provides the sequence used to generate expression identifiers. GMD_MATERIAL_DETAILS_GTMP and GMD_TECHNICAL_DATA_GTMP are global temporary tables used as transient staging areas during parsing and evaluation, and PLITBLM is the standard PL/SQL integer table type used internally for line collections.

Usage Notes

GMD_EXPRESSION_MIG_UTIL is typically invoked indirectly. It is referenced by at least one other package in the EBS schema, which orchestrates expression migration and evaluation during specification setup, formula migration, or quality data conversion. Direct calls are usually made from custom migration scripts or concurrent programs that need to re-parse or re-evaluate stored technical parameter expressions after data loads into 12.1.1 or 12.2.2. Because it is an internal utility and the header indicates a noship status, its interface is not guaranteed to be stable across releases; developers should treat the public subprogram signatures as implementation detail and avoid building long-term custom dependencies on them without regression testing after patching.