Results for “doc_size”

AI-generated from documented ETRM metadata — verify critical details on the linked pages.

Overview

ADI_WEB_FILES_VL is a reporting view owned by the APPS schema in Oracle E-Business Suite, cataloged under the FRM (Report Manager) product family. It is a virtual, non-materialized view whose rows are assembled at runtime from documents stored in the FND document repository and filtered by two specific document categories. Its purpose is to expose binary document payloads — kiosk images and default logos used by the Oracle Application Data Installer (ADI) branding layer — as a flat result set with a synthetic file name, a MIME type, a size indicator, and the raw content. Although the object name carries no explicit reference to FND_DOCUMENTS_VL, that view is its primary driver, which explains why a search for "fnd_documents_vl" surfaces this object in the ETRM catalog.

Underlying Base Objects

The documented dependency list comprises three objects:

• FND_DOCUMENTS_VL (VIEW) — the base document entity. The view aliases it as D and draws DOCUMENT_ID, FILE_NAME, DESCRIPTION, PROGRAM_ID, MEDIA_ID, and CATEGORY_ID from it. FND_DOCUMENTS_VL is the language-enabled (_VL) variant of the FND_DOCUMENTS entity, meaning it resolves translatable description columns against the session language.
• FND_DOCUMENT_CATEGORIES_VL (VIEW) — aliased CL. It supplies the category name used in the filter predicate and joins to D on CATEGORY_ID.
• FND_DOCUMENTS_LONG_RAW (SYNONYM) — aliased LR. This is the media store that physically holds the LONG RAW image bytes, joined to D on MEDIA_ID.

The join therefore links document metadata, category classification, and media content across three layers of the FND document model. Because every referenced object is itself a view or synonym over underlying tables, ADI_WEB_FILES_VL is a view over views and inherits the security and language behavior of those layers.

Key Columns

• NAME — a derived column produced by concatenating DOCUMENT_ID with an underscore and FILE_NAME. This yields a unique, deterministic identifier suitable for use as a servlet or URL parameter; it is not a stored file name.
• MIME_TYPE — populated from the document DESCRIPTION column. This is a deliberate repurposing of the description attribute to carry the content type (for example, image/gif or image/jpeg) rather than human-readable prose.
• DOC_SIZE — sourced from PROGRAM_ID. Despite the column name, this is the identifier of the concurrent program that uploaded the document, not a byte count. Consumers must not treat it as a length field.
• CONTENT — the LONG_RAW payload pulled from FND_DOCUMENTS_LONG_RAW.LONG_RAW. This is the only column carrying binary image data and is subject to the well-known restrictions on LONG and LONG RAW columns in SQL.

Common Use Cases and Queries

The view serves applications that must render ADI kiosk imagery without direct knowledge of the FND document schema. A typical consumer query retrieves all available branded images:

• SELECT name, mime_type, content FROM apps.adi_web_files_vl; — returns every image in the two kiosk categories.
• SELECT name, mime_type, content FROM apps.adi_web_files_vl WHERE mime_type = 'image/gif'; — filters by the description-backed content type.
• SELECT name FROM apps.adi_web_files_vl ORDER BY name; — enumerates identifiers without retrieving the binary payload.

Because the category filter is embedded in the view definition, all rows returned are restricted to ADI_KIOSK_IMAGE and ADI_KIOSK_DEFAULT_LOGO; no additional category predicate is required. Extraction of CONTENT through JDBC, PL/SQL, or the Oracle Applications Framework proceeds by streaming the LONG RAW value, since it cannot be used in WHERE, GROUP BY, or DISTINCT clauses. For diagnostics, joining or aggregating on the view should be limited to NAME, MIME_TYPE, and DOC_SIZE; the content column should be selected only when the binary payload is genuinely required.