MS ODBC w/UTF8 & Oracle 10g database NULLS char fields > 1024 char
From: jhdatabase1 (jhdatabase1_at_discussions.microsoft.com)
Date: 12/02/04
- Previous message: mkamper_at_-NOSPAM-businessprotection.com: "Re: ODBC failure - strange behavior"
- Messages sorted by: [ date ] [ thread ]
Date: Thu, 2 Dec 2004 13:19:04 -0800
We've found the following:
IF fields over 1024 characters in size are edited using a record set:
- With an Oracle 10g database
- AND that Oracle 10g database has been set to use UTF8 encoding
- AND the database connection is made using the Microsoft ODBC driver
- AND the edits are performed using a record set
THEN your edits will NULL out the character field's value.
The above problem does NOT occur with:
- Oracle 8
- Oracle 9
- If the database has regular Latin 1 encoding
- If you use the Oracle ODBC driver rather than the MS ODBC driver.
- Using SQL statements to update the data instead of a record set
This issue requires a fix to the Microsoft ODBC driver that Microsoft
publishes. The latest Microsoft ODBC driver (version 2.575.1117.0), as of Dec
2, 2004, does not resolve this problem.
Has anyone else experienced this?
How can we get Microsoft to fix this bug?
- Previous message: mkamper_at_-NOSPAM-businessprotection.com: "Re: ODBC failure - strange behavior"
- Messages sorted by: [ date ] [ thread ]
Relevant Pages
|
