The undefined code page

A special code-page name, undefined, tells OpenEdge not to do any conversions when reading or writing data to or from this code page. You can specify this name wherever you can specify a code-page name. While declaring data to be undefined is useful in certain situations, do not declare all of your data to be undefined, even if your environment is homogeneous. By identifying the code page of your data, you prepare OpenEdge to handle correctly any future extensions to your environment.

For example, the code pages of the demo and sports databases are set to undefined so that you can use these databases with any character set. This is possible because they contain only ASCII characters, which are included in most other character sets. These databases also have their own character sorting for text within the database. If you elect to customize these databases to your code-page environment, convert the databases to your own code page and collation. For instructions on how to convert the databases to your own code pages, see Using Databases

If you set -cpinternal to undefined for a double-byte or UTF-8 client, multi-byte characters will not be treated properly. They will be treated as single-byte values, and data corruption is likely. The lead and trail bytes might convert as single-byte characters, the double-byte characters might be split, and phrases might be separated incorrectly.