Having fun loading data in Unicode format. I'm generating a flat file ("|" delimited) from SQL Server Integration services in Unicode format. This is in order to accomodate chinese characters mixed in with english characters.
According to the information from Microsoft help and a little hexploration I discovered that the flat file is in Unicode Little Endian format.
But I cannot get this to process using MLoad when I use the -c UTF16 character set. The teradata database is on a Windows 2003 server. The destination table has a mixture of column character sets with the standard varchar for only english character columns, and the CHARACTER SET UNICODE for those columns that handle both english and chinese.
A further frustrating part is that when I convert the file to UTF-8 format (using Notepad save as) and use -c UTF8 it loads without error.
I have also tried different combinations of the -c -i -u runtime flags with UTF16 and UTF16-LE values with no success.
The most common error that is occurring is UTY3410 - semi colon not found to terminate statement. This would seem to indicate that MLOAD script file is in the wrong format, but I still have the problem when I save this in Unicode (using Notepad save as) format.
Any help would be greatly appreciated as I would prefer not to have to reformat my source data into UTF8 each time I want to load.