March 31, 2008 at 1:27 pm
In order to properly support multiple languages, in particular Japanese, should we just use the default collation installed on the SQL Server for NCHAR and NVARCHAR columns rather than specifying a collation?
Currently, our scripts which create the database schema specify a US English collation (SQL_Latin1_General_CP1_CI_AS) for NCHAR and NVARCHAR columns. (All string columns are either NCHAR or NVARCHAR data types in our database.) I think someone just generated the scripts from SSMS and it has just always been that way.
We are running into problems inserting data in to the table--go figure--and I need to provide a definitive solution. I have worked with SQL Server for several years but never had to deal with a language other than US English.
I appreciate any input you can provide. I did review other posts on this topic but did not find anything meeting this inqury. Thank you in advance.
March 31, 2008 at 9:48 pm
Storage and collation are distinct - for storage purposes you already have what you need - the unicode types that allow you to store the Japanese character set. Collation determines sorting, case sensitivity etc - you will need to dig into this to find exactly what you need. The app that I work with uses US default collation as the non-English fields are generally reference type information and we aren't really concerned with any sort issues.
What insert problems are you experiencing?
April 1, 2008 at 9:00 am
Thank you for your reply.
The problem being reported, I believe, was string truncation--which I know has to do with the length of data attempting to be inserted into the column. Unfortunately, I did not see the error, the insert statement, or the data being inserted. I was originally told, 'it does not work' with no information as to what 'it' referred or what action was taking place. From one email this morning, I believe the problem they were experiencing is string truncation. They have left for the day and I will not be able to get any additional information, until late tonight.
Last night, the individuals performing the test setup a Japanese Windows Server 2003 and installed SQL Server 2005 on it. They state that the insertion of both English and Japanese works fine with no alteration to the database creation scripts. Again, I do not know if the same insert statement was performed or one which may have contained less data which would fit into the column or if it was a completely different table. Sigh.
April 1, 2008 at 10:05 am
When is it appropriate to use code page 850? I cannot seem to find any details on this collation. At my previous employer, it was a company standard to install SQL Server with collation CP850 dictionary order, case-insensitive, accent-insensitive.
I am probably derailing the train, here...
Would using CP850 have any benefit in a database which could potentially store different language data (English, German, Japanese)?
Viewing 4 posts - 1 through 3 (of 3 total)
You must be logged in to reply to this topic. Login to reply