Understanding Planning Unicode Conversions

This section describes considerations to take into account when planning to convert to a Unicode database.

Unicode data storage is currently supported on the following PeopleSoft database platforms:

  • Oracle

  • Microsoft SQL Server

  • IBM DB2/UDB

For specific versions or required patches for these database platforms, see the Certifications tab on My Oracle Support for hardware and software requirements for your PeopleTools release.

When planning to convert to a Unicode database consider the impact on the physical size of your data.

Depending on the transformation of Unicode that your database platform uses and the mix of languages you plan to use in the PeopleSoft database, you can expect to see the database grow.

PeopleSoft supports the coexistence of Unicode and non-Unicode databases in a single implementation. Should you need to maintain databases in both Unicode and non-Unicode character sets, most PeopleSoft utilities and commands can operate across both databases seamlessly.

Depending on the character set of the non-Unicode database, some character data may be lost when transferring data between databases. For example, if you load data from a Unicode Japanese and English database into a Latin-1 non-Unicode database, the English data will load correctly, but the Japanese characters will be lost and converted into replacement characters such as question marks.

Although it is easy to have Unicode and non-Unicode databases co-exist, you will need to maintain two separate PeopleSoft file server and batch server environments for each database type, because the COBOL code supplied with some PeopleSoft applications differs significantly between Unicode and non-Unicode databases. Therefore, if you will be maintaining Unicode and non-Unicode databases and you require COBOL, you will need to install the PeopleTools and application CDs twice, and answer the question about Unicode during each installation appropriately

Data Mover

DAT files created by Data Mover are always encoded in Unicode. Data Mover can read these Unicode DAT files and load their contents into either Unicode or non-Unicode databases. When creating a new Unicode database, ensure that the Data Mover command SET UNICODE ON is issued before the IMPORT * command.

Upgrade Compare/Copy

Upgrade Compare and Upgrade Copy supports copying objects between Unicode and non-Unicode databases. There are no restrictions on the source and target database character sets.

Cache Files

Cache files on the PeopleSoft application server are always stored in Unicode format. If you rebuild your non-Unicode database as a Unicode database, you can still use the cache files you may have pre-loaded on the application server.

SQR

On database platforms supporting Unicode, SQR always connects to the database using a Unicode character set via the database vendor’s API. Therefore, you can use SQR to copy data to and from Unicode and non-Unicode databases. When writing and reading files with SQR, you can select the character set that should be used to encode the file using the ENCODING parameter to the SQR OPEN command.

SQR may require some settings to be defined in the PSSQR.INI (or PSSQR.UNX) file for SQR to correctly read and write files in Unicode.