DBCS coding considerations

If the application will be used in a DBCS environment, ensure that it is DBCS-enabled. Here are some suggestions to consider when you develop the general product design.

• Reserve more expansion space for DBCS textual data translation than you reserve for SBCS textual data translation. (It is possible, however, that the number of bytes used may be reduced when a SBCS sentence is being translated into DBCS.)
• Ensure programs can understand shift-out and shift-in delimiters. Otherwise, EBCDIC mixed-byte character strings cannot be handled.
• Do not enable short responses for DBCS. For short responses, it is difficult to shift in and out of DBCS. The yes and no are examples of short responses.
• Remember to use the graphic data type G where appropriate.
• Remember that the 5494 remote controller supports the graphic data type.
• Be careful when converting mixed data between DBCS-host code and DBCS-PC code, because the transition may change the data length. Losing and gaining SO and SI character pairs can upset field-length calculations.
• Make sure the double-byte data is always processed in a double-byte unit. Do not split a double-byte character.
• Design the display as well as the print format to avoid the problem of truncation of a double-byte character into two single-byte units.

• Create physical files
  When you create a physical file, display file, and printer file for a DBCS environment, consider the IGCDTA parameter in these commands.
• Target physical files
  When you specify OUTPUT(*OUTFILE) with the Copy Spool File (CPYSPLF), Display Spooled File (DSPSPLF), or Work with Spooled Files (WRKSPLF) command under the DBCS version of the operating system, the target physical file must be DBCS-enabled.
• Use DBCS CCSIDs
  Use DBCS CCSIDs for DBCS languages. When you design an application to be used in the DBCS environment, consider this information.
• Use DDS keywords
  With DDS keywords, you can specify alternative ways to enter data through display files, change input- and output-capable alphanumeric data fields to DBCS data fields, or specify the special features of the DBCS printer output.
• DBCS file data types
  This table summarizes the changes of the data type of a field in a physical file when the data type is being referred to in a logical file.
• The Katakana code page (00290)
  The Katakana code page (code page 00290) of Japan supports uppercase English and single-byte Katakana (phonetics) characters.
• UCS-2 level-1 support and IBM DBCS displays
  i5/OS™ supports ISO/IEC 10646 Universal Coded Character Set 2, Level 1 (UCS-2, Level-1). IBM^® DBCS-capable display stations, however, do not support UCS-2 Level-1 data.