From ASCII into MSSQL database .... UTF8 ?

[Former Member]

Hi all,

We are using SAP ID Mgmt 7.1 SP5 on Windows x64 server and MSSQL2005 database.

Process

We create a csv file from Microsoft Excel and Saved as CSV (default encoded ANSI)

We upload the file on the server

We read the file from an administration job with Pass FromASCII

Issue

Accentuate caracters are not readed correctly and are replaced by "?"

Ex CSV file:

Firstname;Lastname;

Cécile;DeFrance;

in temp table:

Firstname Lastname

C?cile DeFrance

If I convert the csv file into UTF8 without BOM (using Notepad++), it works !!! But it's not acceptable to convert file each time I need mass import.

We never met this issue when running on Oracle. Do you guys have some tricks for me ?

Thanks,

Benjamin

Edited by: Benjamin Gourdon on Jun 9, 2011 6:47 PM

[Former Member]

Hi Benjamin,

It sounds strange that you should have this problem only with MSSQL and not with Oracle.

The FromASCII pass will use the default character set of the server where it is running. If this is ANSI, you should be OK.

One solution to your problem would be find a utility to convert the file from ANSI to UTF-8, and then call this utility from a ToShell pass, before reading the file.

Best regards

John Erik Setsaas

Development Architect IdM

[Former Member]

Hi Erik,

My MSSQL database is set with "SQL_Latin1_General_CP850_BIN2". I think my issue can be solve by changing this value. Do you have tips for me in order to choose the correct one ?

Thanks for your help,

Benjamin

[former_member2987]

That's the correct database collation for IDM/Netweaver. You should not change it.