I would like to conduct a bit of a survey from all those Teradata customers out there that pull from Unicode sources (i.e. SAP).
When setting up the EDW, which of the following scenarios would you choose, with a short explanation as to why?:
1) All character columns in the EDW are always defined as Unicode
2) Character columns that are identified as potentially carrying Unicode characters are defined as Unicode. Others are defined as Latin.
3) All character columns over a certain length (say varchar(10)) are defined as Unicode, the others are defined as Latin.
4) Other... feel free to share!
I just want to see how other Teradata systems, sourcing from UTF8 sources, handle Unicode columns in the EDW.
Thanks to those who wish to participate!