I am using SQL server 2012
create table t(dt1 date,dt2 date,dt3 date,dt4 date)
insert into t values('1970-01-01','2008-10-10',NULL,NULL),(NULL,'2008-10-10','2017-10-12',NULL),('1970-01-01','2008-10-10',NULL,'2018-10-09')
I need to get the minimum date from these columns, if the column value ='1970-01-01' then I need the second minimum date.
Below is what I tried which is not resulting correct result.
select *,case when (dt1='1970-01-01' or dt2='1970-01-01' or dt3='1970-01-01' or dt4='1970-01-01' )and dt1<=dt2 then dt1 else dt2
end as DDt
from t
Expected output result:
Edit - I need the second minimum date, added more cases here.
Use outer apply
select *
from t
outer apply (select ddt = min(v)
from (values (dt1), (dt2), (dt3), (dt4)) q(v)
where v > '19700101'
) q
The following query works for the scenario : -
SELECT *, MinValue
FROM t
CROSS APPLY (SELECT MIN(d) MinValue FROM (VALUES (dt1), (dt2), (dt3),(dt4)) AS
a(d) WHERE d <> '01-01-1970') A
You could use subquery with values constructors
select *, (select min(dates)
from (values (dt1), (dt2), (dt3), (dt4))a(dates)
where a.dates > '1970-01-01') as DDt
from t;
Try this solution :
SELECT * , (SELECT MIN(Dates) FROM (VALUES (dt1), (dt2), (dt3), (dt4)) AS Fields(Dates) WHERE Fields.Dates > '1970-01-01') AS DDT
FROM [dbo].[t]
Related
Hello I am trying to calculate the time difference of 2 consecutive rows for Date (either in hours or Days), as attached in the image
Highlighted in Yellow is the result I want which is basically the difference of the date in that row and 1 above.
How can we achieve it in the SQL? Attached is my complex code which has the rest of the fields in it
with cte
as
(
select m.voucher_no, CONVERT(VARCHAR(30),CONVERT(datetime, f.action_Date, 109),100) as action_date,f.col1_Value,f.col3_value,f.col4_value,f.comments,f.distr_user,f.wf_status,f.action_code,f.wf_user_id
from attdetailmap m
LEFT JOIN awftaskfin f ON f.oid = m.oid and f.client ='PC'
where f.action_Date !='' and action_date between '$?datef' and '$?datet'
),
.*select *, ROW_NUMBER() OVER(PARTITION BY action_Date,distr_user,wf_Status,wf_user_id order by action_Date,distr_user,wf_Status,wf_user_id ) as row_no_1 from cte
cte2 as
(
select *, ROW_NUMBER() OVER(PARTITION BY voucher_no,action_Date,distr_user,wf_Status,wf_user_id order by voucher_no ) as row_no_1 from cte
)
select distinct(v.dim_value) as resid,c.voucher_no,CONVERT(datetime, c.action_Date, 109) as action_Date,c.col4_value,c.comments,c.distr_user,v.description,c.wf_status,c.action_code, c.wf_user_id,v1.description as name,r.rel_value as pay_office,r1.rel_value as site
from cte2 c
LEFT OUTER JOIN aagviuserdetail v ON v.user_id = c.distr_user
LEFT OUTER JOIN aagviuserdetail v1 ON v1.user_id = c.wf_user_id
LEFT OUTER JOIN ahsrelvalue r ON r.resource_id = v.dim_Value and r.rel_Attr_id = 'P1' and r.period_to = '209912'
LEFT OUTER JOIN ahsrelvalue r1 ON r1.resource_id = v.dim_Value and r1.rel_Attr_id = 'Z1' and r1.period_to = '209912'
where c.row_no_1 = '1' and r.rel_value like '$?site1' and voucher_no like '$?trans'
order by voucher_no,action_Date
The key idea is lag(). However, date/time functions vary among databases. So, the idea is:
select t.*,
(date - lag(date) over (partition by transaction_no order by date)) as diff
from t;
I should note that this exact syntax might not work in your database -- because - may not even be defined on date/time values. However, lag() is a standard function and should be available.
For instance, in SQL Server, this would look like:
select t.*,
datediff(second, lag(date) over (partition by transaction_no order by date), date) / (24.0 * 60 * 60) as diff_days
from t;
I am using the following in a WITH clause to create a FULL JOIN in Big Query:
WITH
a AS(
SELECT
date AS Date,
SUM(Val1 / (1 - (Val2 + Val3))) AS Calc1,
FROM `project.dataset.table1`
GROUP BY Date
),
b as (SELECT
date AS Date,
FROM `project.dataset.table2`
GROUP BY Date
)
SELECT a.Date, SUM(Calc1)
FULL JOIN a on b.Date = a.Date
GROUP BY b.Date
Calc1 is creating a 'division by zero: 1 / 0' error, and I can't seem to work out how to restructure this so it doesn't occur. The query works fine outside of the WITH clause, as I can simply not include the GROUP BY so have no need to SUM Calc1?
Below is for BigQuery Standard SQL
Use
SUM(SAFE_DIVIDE(Val1, 1 - (Val2 + Val3))) AS Calc1
instead of
SUM(Val1 / (1 - (Val2 + Val3))) AS Calc1
Use NULLIF :
WITH
a AS(
SELECT
date AS Date,
SUM(Val1 / NULLIF((1 - (Val2 + Val3)),0)) AS Calc1,
FROM `project.dataset.table1`
GROUP BY Date, SUM(Calc1)
)
Have you analysed your data within "table1" to ensure that Val1, Val2 & Val3 are consistently populated, or do you have NULL values?
This could be the issue with your subtraction from 1.
WITH
a AS(
SELECT
date AS Date,
SUM(Val1 / (1 - (isnull(Val2,0.00) + isnull(Val3,0.00)))) AS Calc1
FROM `project.dataset.table1`
GROUP BY Date
),
b as (
SELECT
date AS Date,
FROM `project.dataset.table2`
GROUP BY Date
)
SELECT a.Date, SUM(a.Calc1)
FULL JOIN a on b.Date = a.Date`enter code here`
GROUP BY b.Date
I have a MS SQL table that contains stock data with the following columns: Id, Symbol, Date, Open, High, Low, Close.
I would like to self-join the table, so I can get a day-to-day % change for Close.
I must create a query that will join the table with itself in a way that every record contains also the data from the previous session (be aware, that I cannot use yesterday's date).
My idea is to do something like this:
select * from quotes t1
inner join quotes t2
on t1.symbol = t2.symbol and
t2.date = (select max(date) from quotes where symbol = t1.symbol and date < t1.date)
However I do not know if that's the correct/fastest way. What should I take into account when thinking about performance? (E.g. will putting UNIQUE index on a (Symbol, Date) pair improve performance?)
There will be around 100,000 new records every year in this table. I am using MS SQL Server 2008
One option is to use a recursive cte (if I'm understanding your requirements correctly):
WITH RNCTE AS (
SELECT *, ROW_NUMBER() OVER (PARTITION BY symbol ORDER BY date) rn
FROM quotes
),
CTE AS (
SELECT symbol, date, rn, cast(0 as decimal(10,2)) perc, closed
FROM RNCTE
WHERE rn = 1
UNION ALL
SELECT r.symbol, r.date, r.rn, cast(c.closed/r.closed as decimal(10,2)) perc, r.closed
FROM CTE c
JOIN RNCTE r on c.symbol = r.symbol AND c.rn+1 = r.rn
)
SELECT * FROM CTE
ORDER BY symbol, date
SQL Fiddle Demo
If you need a running total for each symbol to use as the percentage change, then easy enough to add an additional column for that amount -- wasn't completely sure what your intentions were, so the above just divides the current closed amount by the previous closed amount.
Something like this w'd work in SQLite:
SELECT ..
FROM quotes t1, quotes t2
WHERE t1.symbol = t2.symbol
AND t1.date < t2.date
GROUP BY t2.ID
HAVING t2.date = MIN(t2.date)
Given SQLite is a simplest of a kind, maybe in MSSQL this will also work with minimal changes.
Index on (symbol, date)
SELECT *
FROM quotes q_curr
CROSS APPLY (
SELECT TOP(1) *
FROM quotes
WHERE symbol = q_curr.symbol
AND date < q_curr.date
ORDER BY date DESC
) q_prev
You do something like this:
with OrderedQuotes as
(
select
row_number() over(order by Symbol, Date) RowNum,
ID,
Symbol,
Date,
Open,
High,
Low,
Close
from Quotes
)
select
a.Symbol,
a.Date,
a.Open,
a.High,
a.Low,
a.Close,
a.Date PrevDate,
a.Open PrevOpen,
a.High PrevHigh,
a.Low PrevLow,
a.Close PrevClose,
b.Close-a.Close/a.Close PctChange
from OrderedQuotes a
join OrderedQuotes b on a.Symbol = b.Symbol and a.RowNum = b.RowNum + 1
If you change the last join to a left join you get a row for the first date for each symbol, not sure if you need that.
You can use option with CTE and ROW_NUMBER ranking function
;WITH cte AS
(
SELECT symbol, date, [Open], [High], [Low], [Close],
ROW_NUMBER() OVER(PARTITION BY symbol ORDER BY date) AS Id
FROM quotes
)
SELECT c1.Id, c1.symbol, c1.date, c1.[Open], c1.[High], c1.[Low], c1.[Close],
ISNULL(c2.[Close] / c1.[Close], 0) AS perc
FROM cte c1 LEFT JOIN cte c2 ON c1.symbol = c2.symbol AND c1.Id = c2.Id + 1
ORDER BY c1.symbol, c1.date
For improving performance(avoiding sorting and RID Lookup) use this index
CREATE INDEX ix_symbol$date_quotes ON quotes(symbol, date) INCLUDE([Open], [High], [Low], [Close])
Simple demo on SQLFiddle
What you had is fine. I don't know if translating the sub-query into the join will help. However, you asked for it, so the way to do it might be to join the table to itself once more.
select *
from quotes t1
inner join quotes t2
on t1.symbol = t2.symbol and t1.date > t2.date
left outer join quotes t3
on t2.symbol = t3.symbol and t2.date > t3.date
where t3.date is null
You could do something like this:
DECLARE #Today DATETIME
SELECT #Today = DATEADD(DAY, 0, DATEDIFF(DAY, 0, CURRENT_TIMESTAMP))
;WITH today AS
(
SELECT Id ,
Symbol ,
Date ,
[OPEN] ,
High ,
LOW ,
[CLOSE],
DATEADD(DAY, -1, Date) AS yesterday
FROM quotes
WHERE date = #today
)
SELECT *
FROM today
LEFT JOIN quotes yesterday ON today.Symbol = yesterday.Symbol
AND today.yesterday = yesterday.Date
That way you limit your "today" results, if that's an option.
EDIT: The CTEs listed as other questions may work well, but I tend to be hesitant to use ROW_NUMBER when dealing with 100K rows or more. If the previous day may not always be yesterday, I tend to prefer to pull out the check for the previous day in its own query then use it for reference:
DECLARE #Today DATETIME, #PreviousDay DATETIME
SELECT #Today = DATEADD(DAY, 0, DATEDIFF(DAY, 0, CURRENT_TIMESTAMP));
SELECT #PreviousDay = MAX(Date) FROM quotes WHERE Date < #Today;
WITH today AS
(
SELECT Id ,
Symbol ,
Date ,
[OPEN] ,
High ,
LOW ,
[CLOSE]
FROM quotes
WHERE date = #today
)
SELECT *
FROM today
LEFT JOIN quotes AS previousday
ON today.Symbol = previousday.Symbol
AND previousday.Date = #PreviousDay
Following code I added to the SQL Server query and now have to do the same in Oracle. I need to do grouping in the view rather than in the C#. I get this error message:
ORA-01747 Invalid user.table.column or column specification.
How must I code this to work in Oracle?
SELECT CTE.FACILITY_KEY, CTE.DATE, CTE.PATIENT_STATUS, COUNT(*) AS [COUNT]
FROM CTE
GROUP BY CTE.FACILITY_KEY, CTE.DATE, CTE.PATIENT_STATUS;
at the beginning of query I have this full code here:
CREATE OR REPLACE VIEW DBD_V_CDL_CHANGES AS
WITH CTE AS
(
SELECT TR.FACILITY_KEY
, MV.VALUE_CODE
, CAST(COUNT(*) AS NUMERIC(9, 0)) COUNT
FROM OPTC.THS_T_TRANSACTIONS1 TR
JOIN OPTC.THS_M_MENU2 M
ON M.MENU_ID = TR.MENU_ID
JOIN OPTC.THS_M_VALUES MV
ON MV.MENU_ID = TR.MENU_ID_VALUE
JOIN OPTC.THS_M_VALUES MV2
ON MV2.MENU_ID = TR.PREVIOUS_MENU_ID_VALUE
JOIN OGEN.GEN_M_PATIENT_MAST PM
ON PM.PAT_NUMBER = TR.PAT_NUMBER
WHERE TR.TR_DATETIME BETWEEN TRUNC(SYSDATE)
AND TRUNC(SYSDATE) + 86399 / 86400
AND TR.EDIT_NO < 0
AND MV.VALUE_TYPE IS NULL
AND MV2.VALUE_TYPE IS NULL
AND MV.VALUE_CODE >= 0
AND MV2.VALUE_CODE >= 0
AND M.SUB_SYS_EXT = 'G1'
AND ABS(MV.VALUE_CODE - MV2.VALUE_CODE) > 1
AND (PM.DISCHARGE_DATE IS NULL OR PM.DISCHARGE_DATE < SYSDATE)
GROUP BY TR.FACILITY_KEY, MV.VALUE_CODE)
SELECT CTE.FACILITY_KEY, CTE.DATE, CTE.PATIENT_STATUS, COUNT(*) AS [COUNT] FROM CTE
GROUP BY CTE.FACILITY_KEY, CTE.DATE, CTE.PATIENT_STATUS;
I see a few things wrong with your code.
First, you are selecting the following three columns FACILITY_KEY, VALUE_CODE and the count in the CTE:
SELECT TR.FACILITY_KEY ,
MV.VALUE_CODE ,
COUNT(*) as Count -- note there is no need to CAST(COUNT(*) AS NUMERIC(9, 0)) this
FROM OPTC.THS_T_TRANSACTIONS1 TR
But then when you select from the CTE you are selecting columns that you are not returning in the CTE:
with cte as
(
-- your query here does not return DATE or PATIENT_STATUS
)
SELECT CTE.FACILITY_KEY,
CTE.DATE,
CTE.PATIENT_STATUS,
COUNT(*) AS COUNT
FROM CTE
GROUP BY CTE.FACILITY_KEY, CTE.DATE, CTE.PATIENT_STATUS;
Where do PATIENT_STATUS and Date come from since you are not including them in your CTE? So these do not exist when you are trying to select them.
I replicated your error by including columns in the list that were not select in the CTE query.
The second issue is the CTE.DATE column. DATE is a reserved word, place that is double quotes CTE."DATE"
...AS [COUNT], ...AS NUMERIC(9, 0)) is not Oracle syntax and will never work. Simply remove [ ] and use NUMBER instead of NUMERIC. There is no need to CAST Count(). The Count() function will always return number, e.g. 0-zero or some number.
This is valid syntax in Oracle:
SELECT deptno, count(*) total_count_by_dept -- no need to cast or AS --
FROM scott.emp
GROUP BY deptno
/
Try not to use reserved words as COUNT for aliases:
SELECT CTE.FACILITY_KEY, CTE.DATE, CTE.PATIENT_STATUS, COUNT(*) AS total_cnt -- 'AS' is for clarity only, not required
FROM CTE
GROUP BY CTE.FACILITY_KEY, CTE.DATE, CTE.PATIENT_STATUS
/
My initial query looks like this:
select process_date, count(*) batchCount
from T1.log_comments
order by process_date asc;
I need to be able to do some quick analysis for weekends that are missing, but wanted to know if there was a quick way to fill in the missing dates not present in process_date.
I've seen the solution here but am curious if there's any magic hidden in db2 that could do this with only a minor modification to my original query.
Note: Not tested, framed it based on my exposure to SQL Server/Oracle. I guess this gives you the idea though:
*now amended and tested on DB2*
WITH MaxDateQry(MaxDate) AS
(
SELECT MAX(process_date) FROM T1.log_comments
),
MinDateQry(MinDate) AS
(
SELECT MIN(process_date) FROM T1.log_comments
),
DatesData(ProcessDate) AS
(
SELECT MinDate from MinDateQry
UNION ALL
SELECT (ProcessDate + 1 DAY) FROM DatesData WHERE ProcessDate < (SELECT MaxDate FROM MaxDateQry)
)
SELECT a.ProcessDate, b.batchCount
FROM DatesData a LEFT JOIN
(
SELECT process_date, COUNT(*) batchCount
FROM T1.log_comments
) b
ON a.ProcessDate = b.process_date
ORDER BY a.ProcessDate ASC;