Collation

  • Hi,

    We are planning to store multi language (mostly asian language Chinese, Korean, Thai & English) data in a single database. All the necessary character fields are set to nchar or nvarchar. No issues in storing or retrieving

    I am facing problem in sorting order. I am using the default collation.

    What collation do I need to use? I cannot change the collation for the entire SQL Server (many other databases are there). Many of my stored procedures and functions are using temp table so i used COLLATE database_default wherever nchar/nvarchar is used.

    Reply,

    rgds,

    Aru.

  • Run a search for collation in the BOL(books online)  and you will get code to do per column collation for your langauges.  I  would also save the stored procs T-SQL as Unicode before executing it.  Try the link below for a localization tool kit.  Hope this helps.

    http://www.microsoft.com/downloads/details.aspx?FamilyID=6b6fb09f-f25c-48e9-9e26-b55144600da1&DisplayLang=en

    Kind regards,

    Gift Peddie

    Kind regards,
    Gift Peddie

  • Hello Arumugan,

    What is your problem(s) exactly?

    You could use the COLLATE keyword in the ORDER BY clause.

    You could also (i don't know if it is your need) add a column stating which langage is used to store the values for this row and use it in the WHERE clause to restrict thet data you want to retreive and sort.

    Best regards,

    Carl

  • Aru,

    We have a similar situation here. Our solution was to apply SP4 to SQL 2000 and use the SQL_Latin1_General_CP850_BIN2 collation setting on the database. We discovered some issues using other collations with different languages (Czech was one of them). The drawback is your database is now Case Sensitive, but for us that was a "benefit"

    good luck,

    Darrell

  • Hi,

    My requirement is based on user selected culture, datagrid (ASP.NET) need to be sorted. If Chinese (PRC) is selected, datagird need to be sorted based on stroke.

    My problem is those chinese characters are not properly sorted.

    I am using default collation. What need to be done at UI/DB level. I am using sp to fetch recors.

    Thanks & rgds,

    Aru.

  • I think the default SQL Server 2000 collation will collate CJK characters based on Unicode codepoint, and that is not a stroke order collation.

    So you'll want to collate in the grid, or anyway, at some point along the way.

Viewing 6 posts - 1 through 5 (of 5 total)

You must be logged in to reply to this topic. Login to reply