sql 多个字段分组，删除重复记录，保留ID最小的一条

IF OBJECT_ID('cardDetail') IS NOT NULL
  DROP TABLE cardDetail
 CREATE TABLE cardDetail
 (
 id INT IDENTITY(1,1) PRIMARY KEY,
 cardNO VARCHAR(8),
 NAME VARCHAR(30),
 idCard VARCHAR(18),
 cardState CHAR(1)--卡片状态 1正常 2注销
 )
  INSERT cardDetail(cardNO,NAME,idCard,cardState)
 VALUES
 ('','张三','',1),
 ('','张三','',1),
 ('','李四','',1),
 ('','王五','',1),
 ('','张三','',0),
 ('','王五','',1),
 ('','欧阳','',1),
 ('','欧阳','',1) 

--SQL编写要求 
 --根据idcard,cardstate分组后删除重复的,使其保留id最小的记录
 
 --最终结果为：
 (1，'001','张三','31010',1),
 (4，'003','王五','31012',1),
 (7，'007','欧阳','31013',1),

解决

with t as
(
   select id,cardNO,NAME,idCard,cardState,
      row_number() over (partition by idcard,cardstate order by id) as od
   from cardDetail
)
select * from t where od = 1;