Seems some chinese character have a weird behavior into SQL SERVER 2014. Indeed, even if the character is different, is considered as the same by SQL SERVER.
Here is a small SQL Script to execute by yourself to express the issue :
DECLARE @T AS TABLE
(
ID INT IDENTITY PRIMARY KEY,
Value NVARCHAR(256)
);
INSERT INTO @T(Value) VALUES (N'〇'), (N'㐄');
SELECT * FROM @T;
SELECT * FROM @T WHERE Value = N'〇';
Expected behavior for the second select : get only one result, instead of get two results.
I use SQL_Latin1_General_CP1_CI_AS but I try with
Chinese_PRC_CI_AS and no success.
Thanks for your help.
Complete issue detail : http://stackoverflow.com/questions/35981876/sql-server-and-chinese-character-on-insert-select-query
I found the reason : I have to use the collation because theses characters are traditional.
Chinese_Traditional_Pinyin_100_CI_AS