날짜 참조 테이블 만들기

 

  • Version : SQL Server 2005, 2008, 2008R2, 2012, 2014

 

데이터를 검색 할 때 가장 많이 사용되는 조건 중 하나가 날짜 일 것이다. 다양한 통계 쿼리를 만들다 보면 날짜 범위에 관한 조건문이 많이 사용되는데 이때 해당 기간을 검색하기 위한 참조 테이블을 만들어 사용하면 편리하다. (흔히 날짜 차원 테이블이라고 부른다.)

 

날짜 참조 테이블을 생성한다.

USE TEMPDB

 

IF OBJECT_ID('dbo.#t') is not null

DROP TABLE dbo.#t;

CREATE TABLE #t (

[Date] datetime

, [Year] smallint

, [Quarter] tinyint

, [Month] tinyint

, [Day] smallint -- from 1 to 366 = 1st to 366th day in a year

, [Week] tinyint -- from 1 to 54 = the 1st to 54th week in a year;

, [Monthly_week] tinyint -- 1/2/3/4/5=1st/2nd/3rd/4th/5th week in a month

, [Week_day] tinyint -- 1=Mon, 2=Tue, 3=Wed, 4=Thu, 5=Fri, 6=Sat, 7=Sun

);

GO

 

날짜 정보를 생성 한다. 아래 스크립트의 경우 Week 컬럼의 시작 요일은 월요일이다. 시작 요일을 일요일로 바꾸고 싶으면 -1을 한다.

-- populate the table D_Date, and the day of week is defined as

-- 1=Mon, 2=Tue, 3=Wed, 4=Thu,5=Fri, 6=Sat, 7=Sun

;WITH C0 AS (SELECT c FROM (VALUES(1),(1)) AS D(c)),

C1 AS (SELECT 1 AS c FROM C0 AS A CROSS JOIN C0 AS B),

C2 AS (SELECT 1 AS c FROM C1 AS A CROSS JOIN C1 AS B),

C3 AS (SELECT 1 AS c FROM C2 AS A CROSS JOIN C2 AS B),

C4 AS (SELECT 1 AS c FROM C3 AS A CROSS JOIN C3 AS B),

C5 AS (SELECT 1 AS c FROM C4 AS A CROSS JOIN C3 AS B),

C6 AS (select rn=row_number() over (order by c) from C5),

C7 as (select [date]=dateadd(day, rn-1, '19000101') FROM C6 WHERE rn <= datediff(day, '19000101', '99991231')+1)

INSERT INTO #t ([year], [quarter], [month], [week], [day], [monthly_week], [week_day], [date])

SELECT datepart(yy, [DATE]), datepart(qq, [date]), datepart(mm, [date]), datepart(wk, [date])

, datediff(day, dateadd(year, datediff(year, 0, [date]), 0), [date])+1

, datepart(week, [date]) -datepart(week, dateadd(month, datediff(month, 0, [date]) , 0))+1

, CASE WHEN datepart(dw, [date])+@@datefirst-1 > 7 THEN (datepart(dw, [date])+@@datefirst-1)%7

ELSE datepart(dw, [date])+@@datefirst-1 END

, [date]

FROM C7

--where [date] between '19900101' and '20990101'; -- if you want to populate a range of dates

GO

 

날짜 정보가 1900-01-01부터 9999-12-31일 까지 데이터가 생성되었다. (자그마치 2958464개의 행이다.)

 

 

[월별 첫째 날짜와 마지막 날짜 찾기]

-- find first day of prev/curr/next month with current date

declare @curr_dt datetime = '2015-02-11';

 

select distinct First_Mth_Day=first_value([Date]) over (partition by [month] order by [Date] asc)

, Last_Mth_Day=last_value([Date]) over (partition by [month] order by [Date] asc ROWS BETWEEN Current Row AND UNBOUNDED FOLLOWING )

from #t

where datediff(month, @curr_dt, [Date]) between -1 and 1

and datediff(year, @curr_dt, [Date]) between -1 and 1

order by 1

go

 

 

[토, 일요일을 제외한 첫 날짜와 마지막 날짜 구하기]

-- find the first/last weekday of prev/curr/next month

declare @curr_dt datetime = '2015-02-11';

select distinct First_Week_Day=first_value([Date]) over (partition by [month] order by [Date] asc)

, Last_Week_Day=last_value([Date]) over (partition by [month] order by [Date] asc ROWS BETWEEN Current Row AND 31 FOLLOWING )

from #t

where datediff(month, @curr_dt, [Date]) between -1 and 1

and datediff(year, @curr_dt, [Date]) between -1 and 1

and Week_Day between 1 and 5

order by 1

go

 

 

[주말(토요일 또는 일요일)의 첫 날짜와 마지막 날짜 구하기]

-- find the first / last weekend day of prev/curr/next month

declare @curr_dt datetime = '2015-02-11';

with c as (

select distinct First_Wknd_Day=first_value([Date]) over (partition by [month] order by [Date] asc)

, Last_Wknd_Day=last_value([Date]) over (partition by [month] order by [Date] asc ROWS BETWEEN Current Row AND UNBOUNDED FOLLOWING )

from #t

where datediff(month, @curr_dt, [Date]) between -1 and 1

and datediff(year, @curr_dt, [Date]) between -1 and 1

and Week_Day between 6 and 7

)

select [Month]=choose((row_number() over (order by First_wknd_day asc)), 'Prev_Mth', 'Curr_Mth', 'Next_Mth')

, First_Wknd_Day, Last_Wknd_day

from c

go

 

 

[매월 주 시작 날짜와 주 마지막 날짜 구하기]

-- find the first/last weekday in each month of all years

select distinct [Year], [Month], First_Week_Day=first_value([Date]) over (partition by [year], [month] order by [Date] asc)

, Last_Week_Day=last_value([Date]) over (partition by [year], [month] order by [Date] asc ROWS BETWEEN Current Row AND 32 FOLLOWING )

from #t

where Week_Day between 1 and 5 -- for weekend, change to: between 6 and 7

order by 1, 2

go

 

 

[검색 기간동안 토,일요일이 몇번있는지 구하기]

-- how many weekend days or weekdays between two dates

declare @start_day datetime ='2015-01-11', @End_day datetime ='2015-02-03'

select [Start_Date] = @Start_day, [End_Date]=@End_day, Total_weekend_days = count(*)

from #t

where (Week_Day between 6 and 7) -- for weekdays, use "between 1 and 5"

and ([Date] between @start_day and @end_day)

go

 

 

[매월 2번째 주 시작날짜 구하기]

-- find nth week/weekend day of each prev/curr/next month

-- eg. find the 2nd Monday of each prev/curr/next month

declare @curr_dt datetime = getdate();

; with c as (select rn=RANK() over (partition by [month] order by [date] ASC), [Date] -- attention to 'ASC'

from #t

where datediff(month, @curr_dt, [Date]) between -1 and 1

and datediff(year, @curr_dt, [Date]) between -1 and 1

and Week_Day = 1 -- 1=Mon, 2=Tue,... 7=Sun

)

select [Month]=choose((row_number() over (order by (select null))), 'Prev_Mth', 'Curr_Mth', 'Next_Mth')

,[2nd_Monday] = [Date] from c

where rn=2 -- nth, for example if finding the 3rd Monday, set rn=3

order by 2

go

 

 

[매월 마지막 2번쨰 주 마지막 날짜 구하기]

-- find the nth last week/weekend day of prve/curr/next month

-- eg. find the 2nd last Sat of prev/curr/next month

declare @curr_dt datetime = getdate();

; with c as (select rn=RANK() over (partition by [month] order by [date] DESC), [Date] -- attention to 'DESC'

from #t

where datediff(month, @curr_dt, [Date]) between -1 and 1

and datediff(year, @curr_dt, [Date]) between -1 and 1

and Week_Day = 6 -- 1=Mon, 2=Tus, ... 7=Sun

)

select [Month]=choose((row_number() over (order by (select null))), 'Prev_Mth', 'Curr_Mth', 'Next_Mth')

, [2nd_Last_Sat]=[Date] from c

where rn=2

order by 2

go

 

 

 

[참고자료]

http://www.mssqltips.com/sqlservertip/3508/sql-server-date-time-calculation-examples/

 

 


강성욱 / jevida@naver.com
Microsoft SQL Server MVP
Blog : http://sqlmvp.kr
Facebook : http://facebook.com/sqlmvp





profile

강성욱 / jevida@naver.com
Microsoft SQL Server MVP
Blog : http://sqlmvp.kr
Facebook : http://facebook.com/sqlmvp

Kakao Talk : SQLMVP

Line : jevida


현재 LA에 거주하고 있으며 SQL에 관심있는 분이면 언제든 친추 환영합니다.