July 8, 2007 at 9:38 am
Guys - I've read through mots of the threads on this section and they contain some good information.
What I'm interested most in is people's experiences on this? We are considering entering into the Japanese market place.
We have over 1,200 tables, 10,000 stored proc/functions. No NVARCHAR tables defined. Our application is written in C++.
Does anyone have a sense of the typical stumbling blocks that came up? From what I've read (and it makes sense) every field we need to store Unicode data in requires a change to the core type and as a result the associated stored procedures.
Any guidance as to where to start or plan for a task like this would be appreciated.
July 9, 2007 at 2:36 am
I'm not educated in this arena....but I'd go looking for items within your application that won't convert/upgrade....ie any char fields over 4000, or text/binary items that need special treatment. This will enable you to re-structure, and get working, your current app before the upgrade.
After that I'd invent a dummy 3 table database with linked app, including cascade relationships under your existing style of operation and see what hurdles you hit getting it work under Japanese. Code change timings and data conversion exercises + localisation issues for your app need to be logged and 100% cleared.
After that it should become a matter of scale....
It sounds a lot of work...so good luck!
July 24, 2007 at 3:47 pm
If your application runs in 2005 you can use Nvarchar and Nvarchar (max) and no you cannot store Japanese, Korean, Chinese and I think Vietnamese in Varchar because their alphabet is not the 26 characters long Latin alphabet, their alphabets are more than 2000 characters. Japanese I don't see you running an application with 1,200 tables, 10,000 stored proc/functions in varchar.
Kind regards,
Gift Peddie
July 25, 2007 at 11:40 am
Knowing nothing about your application - it's quite possible that you could require the users to enter all content in English as far as the database is concerned, and the C# programming would need to adapt for the user's GUI, with a mix of Japanese and English on the screens. That would be the happiest solution from a technical point of view, but only a limited solution for the Japanese clients.
July 25, 2007 at 11:48 am
Mixing Asian language with no accent Latin alphabet language like English is a recipe for character conversion issues.
Kind regards,
Gift Peddie
Viewing 5 posts - 1 through 4 (of 4 total)
You must be logged in to reply to this topic. Login to reply