API Reference

Appendix A. Naming Conventions

This section provides information about the conventions that apply when naming database manager objects, such as databases and tables, and authentication IDs.

• Character strings that represent names of database manager objects can contain any of the following: a-z, A-Z, 0-9, @, #, and $.

• The first character in the string must be an alphabetic character, @, #, or $; it cannot be a number or the letter sequences SYS, DBM, or IBM.

• Unless otherwise noted, names can be entered in lowercase letters; however, the database manager processes them as if they were uppercase.

  The exception to this is character strings that represent names under the systems network architecture (SNA). Many values, such as logical unit names (partner_lu and local_lu), are case sensitive. The name must be entered exactly as it appears in the SNA definitions that correspond to those terms.

• A database name or database alias is a unique character string containing from one to eight letters, numbers, or keyboard characters from the set described above.

  Databases are cataloged in the system and local database directories by their aliases in one field, and their original name in another. For most functions, the database manager uses the name entered in the alias field of the database directories. (The exceptions are CHANGE DATABASE COMMENT and CREATE DATABASE, where a directory path must be specified.)

• The long identifier or alias for a database table or view, and the name of a column within a table or a view, are unique character strings 1 to 18 characters in length.

  A fully qualified table name consists of the schema.tablename. The schema is the unique user ID under which the table was created.

• Authentication IDs (both user IDs and group IDs) cannot exceed eight characters in length.

For more information about naming conventions, see the Administration Guide.

[ DB2 List of Books | Search the DB2 Books ]