Dataservices 4.2 Character Set Problem

AMM
Padawan

Hi guys,

 

we have SAP dataservices 4.1 in use, also with an EXASOL 7 database.

Now we want to upgrade to SAP dataservice 4.2, and got a character set problem in connection with EXASOL.

The problem is that the records read from an Oracle database in UTF-8 are converted incorrectly ("ü" => "ü" ).

I.e. Dataservice generally converts the record from the Oracle database (UTF-8) as Windows-1252.
The conversion back to UTF-8 for Exasol then goes wrong.

To solve this problem, we have made the setting: "ANSIARGENCDING=65001;ANSIDATAENCODING=65001" in the ODBC driver for EXASOL. It works with that, but the conversion takes time. It is, after all, a double conversion.
Is there another solution?

Does anyone have any advice?

2 REPLIES 2

exa-MathiasHo
Community Manager
Community Manager

Hi @AMM,

Thank you for your question. Did the advice from @exa-Kristof help you? If so, please let the community know in case somebody else has this kind of issue. 

exa-Kristof
Team Exasol
Team Exasol

Hi @AMM ,

I don't have much Dataservice experience, but here are some thoughts:

First of all: If I get you right, you were able to solve your problem, you "just" want to get rid of the double conversion. Is that correct? 

There was no issue in 4.1, correct? So something has changed in 4.2. Is there a parameter to configure Dataservices so that it does no conversion? I found an article in the SAP knowledge base, but can't fully access it. Maybe you can: https://userapps.support.sap.com/sap/support/knowledge/en/2371743

An alternative way of loading data from Oracle to Exasol would be to use the IMPORT statement. Setup instructions for Oracle can be found here: https://docs.exasol.com/loading_data/connect_databases/oracle.htm  

I wouldn't expect any codepage issues using that way and from my experience you might even experience runtime improvements (but that depends on a lot of things, so I can't guarantee it).