Character conversion with CLI?

Hi, we are using Informix CLI (2.0 16bit, 2.5 32bit) to access our Online 5.07 data from VisualBasic 4.0 applications. While setup and performance are satisfactory, we have a problem with the different codesets on the platforms involved. The CLI drivers came with no documentation except an installation guide, so I have to turn to the net for help. The databases in question are accessed both from 7bit-ASCII-Terminals and from Windows-PCs, so the CHAR data are 7bit-ASCII. To display German Umlaut characters, the Terminals have a "CH.CODE" key that displays e.g. the "|" character as the German umlaut "=F6" (<-- o with double dots on top). A terminal user enters a string with "o umlaut" and sees that, but what's actually stored in the database is a "|", and that's also what the Windows user sees. If a Windows user enters an umlaut character, this will be ASCII > 127, and the terminal users can't see that. What I need is a translation module in the ODBC driver that translates the Windows "o umlaut" to the ASCII "|" character etc. and vice versa. There is an option in the "ODBC INFORMIX 7.2 Advanced Driver Setup" that lets me choose a translation table. I tried that and picked "MS Codepage Translation Program" with the option "Windows Latin 1 (1252)/437 (English)", but that gave me an error message when I first tried to receive data that the translation DLL could not be found. I installed CLI as directed, so I am a little confused. Is the described translation table really what I need? If so, how can I use it? If not, what can I do? Regards, Richard -- = +--------------------------+------------------------------------------+ | Dr. Richard Spitz | INTERNET: spitz@ana.med.uni-muenchen.de | | EDV-Gruppe Anaesthesie | Tel : +49-89-7095-3413 | | Klinikum Grosshadern | FAX : +49-89-7095-8886 | | 81366 Munich, Germany | | +--------------------------+------------------------------------------+