db-torque-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Tobias Hilka" <thi...@vps.de>
Subject Unicode Problem using Torque
Date Mon, 05 Feb 2007 10:35:32 GMT
Hello everyone,

Some weeks ago I posted a new issue (see
http://issues.apache.org/jira/browse/TORQUE-70) concerning oracle and the
escape character used there. (column was NVARCHAR, so escape character has
to be N'\')

Now a similar problem also occured using MSSQL Server. We have a column
which is of type NVARCHAR. In this column, we enter some czech letters (for
example č, a c with this arrow pointing downwards above it). The query
executed is not case sensitive, so torque builds up something like 

where lastname like UPPER('%č%') 

which returns all entries having a regular "c", not the special czech "č" in
it.

Here, the same fix applies: if you write 

where lastname like UPPER(N'%č%') 

all the desired result are retrieved.

As far as I tested it with my local system here, everything works fine if
you add this N in front of every string literal. 
So, what do you suggest for me to do? I could patch the class SQLExpression
(build, buildLike, ...), but this is not a very nice solution.

Thanks for your help.

Tobias

Mime
View raw message