Search Results p_language

Overview

APPS.HZ_TRANS_PKG is the seeded transformation procedure package within Oracle Trading Community Architecture (TCA) Data Quality Management (DQM). It supplies the library of transformation functions that cleanse, standardize, and match party, person, organization, address, and contact data held in the HZ schema. The package is a public API (AUTHID CURRENT_USER) declared in ARHDQTRS.pls and documented as "DQM Transformations" in the Oracle Trading Community Architecture Technical Implementation Guide (120hztig.pdf). Its reported lifecycle status is active, and it is classified against the business entities HZ_PARTY, HZ_PERSON, HZ_ORGANIZATION, HZ_ADDRESS, and HZ_CONTACT.

Each transformation function accepts a source attribute value and returns a transformed result, allowing DQM rules and match rules to be assembled declaratively. In both Oracle EBS 12.1.1 and 12.2.2, the package is present under the APPS schema and is referenced by numerous other packages (43 by ETRM count), reflecting its role as a shared utility layer for entity resolution and data quality enforcement.

Key Procedures and Functions

The documented interface comprises 47 procedures and functions. The principal transformation functions share a common calling pattern, receiving the original attribute value, a language identifier, the attribute name, and the entity name; some accept additional context. The p_language parameter signals the linguistic context for transformations that are language-sensitive, such as phonetic and word-clustering routines.

• EXACT — returns the input value unchanged for exact-match comparison, forming the baseline transformation.
• EXACT_PADDED — a padded variant of EXACT used where fixed-width comparison semantics are required.
• RM_SPLCHAR — removes special characters from the value; accepts an optional context argument.
• RM_BLANKS — strips embedded and redundant blank characters.
• CLEANSE — applies configured cleansing rules to standardize the value.
• CLUSTER_WORD — performs word clustering, typically for fuzzy matching support.
• SOUNDX — generates a phonetic (Soundex-style) representation for approximate name matching.
• ACRONYM — derives an acronym representation from multi-word values.
• REVERSE_NAME — reverses name component order to support alternate name matching.
• EXACT_DATE, EXACT_NUMBER — comparison transformations for date and numeric attributes.
• EXACT_EMAIL, CLEANSED_EMAIL, EXACT_URL, CLEANSED_URL — validation and normalization transformations for electronic address attributes.
• WRPERSON_EXACT, WRPERSON_CLEANSE, WRPERSON_CLUSTER — person-specific wrap transformations.
• WRORG_EXACT, WRORG_CLEANSE — organization-specific wrap transformations.

Tables Accessed

The package reads configuration and reference data through APPS synonyms:

• HZ_WORD_LISTS — supplies the word lists used by clustering and cleansing transformations.
• HZ_WORD_REPLACEMENTS — provides substitution mappings applied during cleansing.
• HZ_DQM_STAGE_LOG — supports staging and logging of DQM transformation activity.
• PLITBLM — the PL/SQL internal table used for character-set and linguistic processing in multilingual transformations.

Usage Notes

HZ_TRANS_PKG functions are invoked by DQM transformation rules configured in the Trading Community Architecture, which in turn execute during batch data quality and matching concurrent programs and during real-time party, customer, and supplier processing. Because the package is a documented public API under APPS, custom code may call these functions directly — typically to preview cleansing or matching output, or to embed the same standardization logic used by seeded DQM rules. When invoking any function that exposes p_language, callers should pass the appropriate language code, since phonetic and word-based transformations depend on it for correct results in multilingual deployments.