December 12, 2007 at 8:07 am
We're changing our main CRM db to unicode on our SQL Server 2000 SP4 server because we need to support multiple character sets including russian.
This works fine for most applications but we're having trouble with applications connecting through ODBC.
A simple select of an nvarchar column containing russian characters returns question marks. QA or EM show the data correctly.
All documents I can find indicate that the SQL Server ODBC driver supports unicode from MDAC 2.1 onwards so I'm at a loss as to why this is happening.
We're using version 3.52 of the ODBC driver. The SQLSRV32.DLL is version 0.3.85.1117
Hans
December 12, 2007 at 8:45 am
I'm not a localization expert, however I've seen that question marks often appear in clients when the local code page doesn't support the character.
I've seen the Word quotes in emails here (or articles) causing problems in browsers or email clients where they appear.
December 12, 2007 at 1:44 pm
I thought about the code page as well, but on the same machine it displays correctly in QA and EM which would indicate that the codepage is fine.
I tried multiple ODBC clients to make sure it wasn't the application.
December 14, 2007 at 2:41 am
Have you already converted your (var)char columns to n(var)char?
Wildly guessing:
UTF-8 isn't implemented in sql server 2000
http://support.microsoft.com/kb/232580
Working of unicode in ODBC
December 14, 2007 at 8:19 am
it looks like the applications we used for testing (MS Query and ODBC Query Tool) were not able to render the russian charachters. I found a tool which specifies that it does support Unicode (Aglowsoft) and that one did rendered the characters perfectly without any change to db or DSN.
case closed.
thanks for all your thoughts on this matter
Hans
Viewing 5 posts - 1 through 5 (of 5 total)
You must be logged in to reply to this topic. Login to reply