Get first row for one group [closed]

问题

It's difficult to tell what is being asked here. This question is ambiguous, vague, incomplete, overly broad, or rhetorical and cannot be reasonably answered in its current form. For help clarifying this question so that it can be reopened, visit the help center.

Closed 7 years ago.

I want to get the first row for every GV(TeacherID)

GV |   Class|  SUM 
GV1|   L001 |  5000 
GV1|   L002 |  5000 
GV1|   L003 |  5000 
GV2|   L002 |  7000 
GV2|   L003 |  7000 
GV2|   L001 |  7000 
GV3|   L001 |  8000 
GV3|   L002 |  8000 
GV3|   L003 |  8000 

Help me.

回答1:

Edit: ok, now you've edited the question such that this answer looks completely irrelevant...sigh...I'll leave it in case it helps you get started. Cheers.

---

The simplest solution given your specs:

  select teacherid 
    from mytable 
group by teacherid;

If you need other information in addition to the teacherid:

  select teacherid, ...other cols...
    from (select teacherid, ...other cols...
                 row_number() over (
                     partition by teacherid
                     order by classid /* or class as per edit */) as row_num
            from mytable) my_derived_table
   where my_derived_table.row_num = 1;

Caveat emptor: I don't have an installation of SQL-Server handy to test on, so syntax may not be exactly correct; but it's close.

回答2:

WITH T AS (
    SELECT yourTable.*, ROW_NUMBER() OVER(PARTITION BY TeacherID 
      ORDER BY ClassID) AS RN
    FROM yourTable 
)
SELECT * 
FROM T
WHERE RN = 1

回答3:

DDL

create table #t
(
    GV  varchar(4),
    Class varchar(4),
    [SUM] int
)

---

Sample Records

insert into #t(GV, Class, [SUM])values('GV1', 'L001', 5000)
insert into #t(GV, Class, [SUM])values('GV1', 'L002', 5000)
insert into #t(GV, Class, [SUM])values('GV1', 'L003', 5000)
insert into #t(GV, Class, [SUM])values('GV2', 'L002', 7000)
insert into #t(GV, Class, [SUM])values('GV2', 'L003', 7000)
insert into #t(GV, Class, [SUM])values('GV2', 'L001', 7000)
insert into #t(GV, Class, [SUM])values('GV3', 'L001', 8000)
insert into #t(GV, Class, [SUM])values('GV3', 'L002', 8000)
insert into #t(GV, Class, [SUM])values('GV3', 'L003', 8000)

---

Query

Select GV, class, [sum] from
(
    Select Row_Number() Over(Partition by GV Order by GV ) as RowId, * from #t
)K
Where RowId = 1

drop table #t

---

Resultset

GV   class sum
---- ----- ----
GV1  L001  5000
GV2  L002  7000
GV3  L001  8000

来源：https://stackoverflow.com/questions/9287119/get-first-row-for-one-group