Charset in Adabas

, , , ,
1

Dear All,

As I know there is Variable type U in Natural program to support Unicode8 or 16.

Do I need to set anything in Adabas like charset in Oracle?

The question sound stupid, but I am not familiar with Adabas

Thanks

Ray

2

This maps to format “W” in the Adabas FDT (the field or “column” definition, to put it in SQL terms).

More information can be found here:

[url]http://techcommunity.softwareag.com/ecosystem/documentation/adabas/ada614unw/admin/dbuessup.htm[/url]

3

Hello,

Natural internally works with UTF-16 and Adabas Open Systems with UTF-8.
On Adabas Mainframe, what the application sees depends on the user encoding defined with UWCODE (default UTF-16).

I suggest to always use the OPRB parameter to ensure that the application gets Unicode fields in the correct encoding.

OPRB:(DBID=1,WCHARSET='UTF-16LE',DBID=8,WCODE=4095)

The above example shows the essential parameters:

  • access to open systems database 1, application characterset is UTF-16 little endian (e.g. for Intel386 box)

  • access to mainframe database 8, application characterset is 4095 (= UTF-16), endianness of application determined by nucleus

When planning new applications with Unicode fields the following should be considered:

  1. Adabas measures field length in bytes, Natural U fields in 2 bytes per character (e.g. WW,40,W = U20)

  2. conversion between UTF-8 and UTF-16 encodings means differing length requirements and that data stored in UTF-8 might not fit in UTF-16 field and vice versa. Suggestion: define fields lengths bigger in database and make sure application does not exceed a certain limit.

Kind regards,
Marbod Mueller