CODEWITHSUNDEEP
sql-serverunicodeencodingutf-8sas
TJB
TJB

Reputation: 877

Reading unicode characters in sessions with wlatin1 encoding in SAS

I have a library which connects to a SQL Server 2012 instance. One of the registered tables contains international addresses so can contain non-Latin characters (Chinese, Russian etc) in nvarchar columns. When connecting to the table via Enterprise Guide, the non-Latin characters are missing.

I have tried changing the SAS Foundation sasv9.cfg to point at the u8 nls folder which has a default encoding of utf-8 instead of the en folder, which is wlatin1.

However, my users then complain their datasets written with the wlatin1 encoding are inaccessible. Changing the sasv9.cfg file back to the 'en' nls folder resolves the issue.

Is there a way I can get users to game this by starting their individual sessions as utf-8?

Thanks

Upvotes: 2

Views: 2144

Answers (1)

Allan Bowe
Allan Bowe

Reputation: 12691

You have a few approaches available. One is to use the inencoding=utf8 library option.

If this fails, you could then try using the encoding=utf8 dataset option.

As a last resort, you can use the kcvt function to translate the specific variable, eg :

kcvt(YourVariable,'utf8','wlatin1')

You can't modify the encoding of workspace session without updating the config file, although it is possible to 'game' a stored process session by setting the context of your client - see the locale gotcha.

Upvotes: 3

Related Questions