Naming Rules on Oracle

The following rules apply when naming schema objects:

  • The maximum length for identifiers in Oracle is 30 bytes. When Uniface constructs object names, it automatically truncates entity and field names from the application model to comply with the Oracle limit. For example, index number 11 of the table named LONG_NAME_OF_THIRTY_CHARACTERS, is named by Uniface as LONG_NAME_OF_THIRTY_CHARACTI11. Uniface uses this truncation method when it constructs names for the following objects:

    • Tables

    • Columns

    • Primary and Unique key constraints

    • Indexes

    • Package specifications and package bodies

    • Procedure parameters

    • Placeholders

    Ensure that the first 26 bytes of all the entity and field names are unique, because they will be truncated somewhere after that point to comply with the Oracle restriction.

  • Unicode character can be used in table names, user names, and passwords.

  • Letters, numbers, and underscores are all allowed in entity and field names in the application model, however, the first character must be a letter. If you want to use an object which has a mixed-case name, you must create a synonym for that object in Oracle.

    For example, if you want to access an Oracle table called Dept, use the name DEPT in the application model and create a synonym by issuing the following statement in the SQL Workbench: 

    CREATE SYNONYM DEPT FOR "Dept"

  • Uniface encloses identifiers in SQL and PL/SQL statements in double quotation marks (""). This allows you to use Oracle reserved words for entity and field names in the application model, and it improves the portability of Uniface applications across DBMSs.

  • It is recommended that you avoid using Oracle reserved words wherever possible.