1. Building Semantic Models in Oracle Analytics Cloud
  2. Reference
  3. Miscellaneous Reference Information
  4. Model Binary Large Object (BLOB) Data and Character Large Object (CLOB) Data

Model Binary Large Object (BLOB) Data and Character Large Object (CLOB) Data

Learn how to model binary large object (BLOB) data and character large object (CLOB) data in a semantic model.

CLOB data is a large plain text document in any character set. The supported BLOB image types are: GIF, PNG, TIFF, JPEG, and BMP. BLOB formats not supported are: PDF, audio, or video.

The default data type for BLOB columns after the import is LongVarBinary, while for CLOB columns it's LongVarChar. The column for the BLOB or CLOB can't exceed the MaxFieldSize limit of 32 KB.

When configuring the physical joins, create a physical join between the tables using the primary key when the primary key is used as a join in the other table.

  1. On the Home page, click Navigator and then click Semantic Models.
  2. In the Semantic Models page, click a semantic model to open it.
  3. Import the physical table containing the BLOB or CLOB data from the data source into the physical layer.
  4. After import, open the physical column for the BLOB or CLOB column, and change the Length field.
  5. Configure physical joins.
  6. Drag the BLOB or CLOB column to the logical layer to generate a logical column.
  7. Configure a physical lookup for the logical column to ensure that the Oracle Analytics query engine doesn't generate a group by or order by on the logical column.
  8. In the logical column's General tab, configure the Descriptor ID column to ensure that Presentation Services uses the correct column when generating filters.
  9. Configure the Sort order column, configure the sort order column to ensure that the Oracle Analytics query engine orders column as expected.
  10. Save the changes.