Question

我需要为每个company_name和node_name获取表T_TBL_DATA中的第一个文本（具有最低row_id的行）。类似的东西：

SELECT company_name, node_name, [text]
FROM T_TBL_DATA
GROUP BY company_name, node_name
WHERE the selected [text] is from the row with the smallest row_id within each group

有什么想法吗？

Answer 1

select company_id, node_id, [text]
from t_tbl_data
where row_id in
(
    select min(row_id)
    from t_tbl_data
    group by company_id, node_id
)

Answer 2

假设id是唯一的。数据库无关的解决方案在某种程度上：

select
    company_id, node_id, (select txt from T_TBL_DATA t2 where t2.id =  min(t1.id))
from T_TBL_DATA t1
group by company_id, node_id

对于MS SQL Server 2005及更高版本更简单一点：

-- =============
-- sample data
-- =============
declare @t table
(
    id int,
    company_id int,
    node_id int,
    txt varchar(50)
)

insert into @t values (1, 1, 1, 'abc1')
insert into @t values (2, 1, 1, 'abc2')
insert into @t values (3, 1, 2, 'abc3')
insert into @t values (4, 1, 2, 'abc4')
insert into @t values (5, 2, 1, 'abc5')
insert into @t values (6, 2, 1, 'abc6')
insert into @t values (7, 2, 2, 'abc7')
insert into @t values (8, 2, 2, 'abc8')

-- =============
-- solution
-- =============
select
    company_id, node_id, txt
from
(
    select 
        *,
        id2 = min(id) over(partition by company_id, node_id)
    from @t
) t
where id = id2

查询选择组中的特定行

2 个答案: