I am trying to load a SQL Server table with text fields that contain diacritical characters such as è (e+grave accent), ö(o+umlaut), etc.
The source coude of MsSQLDataWriter.java shows that UTF-8 encoded data is sent to BCP. However BCP does not appear to support UTF-8; normal characters are fine, but diacritics show up as multiple garbage characters.
Experiments with specifying various codepages in the BCP-command were unsuccesful.
When I tweak the java-code of MsSQLDataWriter to use charset ISO-8859-1 instead of UTF-8 that works fine for me.
Is there a way to load the diacritical characters with MsSQLDataWriter without tweaking the code?
it should be possible to specify a code page for BCP. on the wiki documentation of the MSSQl loader (http://wiki.clovergui.net/doku.php?id=c … ata_writer),
you can see the “codePageSpecifier” parameter of the component. Try to use that with a valid windows code page number.
use the UniversalWriter to save the data into flat file and doing the code page conversion (e.g. UTF-8 into WINDOWS-1250 for Central European). After that use System Execute component to execute bcp. That will load the data (in correct encoding) into database.