in the below table example - Table A, we have entries for four different ID's 1,2,3,4 with the respective status and its time. I wanted to find the "ID" which took the maximum amount of time to change the "Status" from Started to Completed. In the below example it is ID = 4. I wanted to run a query and find the results, where we currently has approximately million records in a table. It would be really great, if someone provide an effective way to retrieve this data.
Table A
ID Status Date(YYYY-DD-MM HH:MM:SS)
1. Started 2017-01-01 01:00:00
1. Completed 2017-01-01 02:00:00
2. Started 2017-10-02 03:00:00
2. Completed 2017-10-02 05:00:00
3. Started 2017-15-03 06:00:00
3. Completed 2017-15-03 09:00:00
4. Started 2017-22-04 10:00:00
4. Completed 2017-22-04 15:00:00
Thanks!
Bruce
You can query as below:
Select top 1 with ties Id from #yourDate y1
join #yourDate y2
On y1.Id = y2.Id
and y1.[STatus] = 'Started'
and y2.[STatus] = 'Completed'
order by Row_number() over(order by datediff(mi,y1.[Date], y2.[date]) desc)
SELECT
started.ID, timediff(completed.date, started.date) as elapsed_time
FROM TABLE_A as started
INNER JOIN TABLE_A as completed ON (completed.ID=started.ID AND completed.status='Completed')
WHERE started.status='Started'
ORDER BY elapsed_time desc
be sure there's a index on TABLE_A for the columns ID, date
I haven't run this sql but it may solve your problem.
select a.id, max(DATEDIFF(SECOND, a.date, b.date + 1)) from TableA as a
join TableA as b on a.id = b.id
where a.status="started" and b.status="completed"
Here's a way with a correlated sub-query. Just uncomment the TOP 1 to get ID 4 in this case. This is based off your comments that there is only 1 "started" record, but could be multiple "completed" records for each ID.
declare #TableA table (ID int, Status varchar(64), Date datetime)
insert into #TableA
values
(1,'Started','2017-01-01 01:00:00'),
(1,'Completed','2017-01-01 02:00:00'),
(2,'Started','2017-02-10 03:00:00'),
(2,'Completed','2017-02-10 05:00:00'),
(3,'Started','2017-03-15 06:00:00'),
(3,'Completed','2017-03-15 09:00:00'),
(4,'Started','2017-04-22 10:00:00'),
(4,'Completed','2017-04-22 15:00:00')
select --top 1
s.ID
,datediff(minute,s.Date,e.EndDate) as TimeDifference
from #TableA s
inner join(
select
ID
,max(Date) as EndDate
from #TableA
where Status = 'Completed'
group by ID) e on e.ID = s.ID
where
s.Status = 'Started'
order by
datediff(minute,s.Date,e.EndDate) desc
RETURNS
+----+----------------+
| ID | TimeDifference |
+----+----------------+
| 4 | 300 |
| 3 | 180 |
| 2 | 120 |
| 1 | 60 |
+----+----------------+
If you know that 'started' will always be the earliest point in time for each ID and the last 'completed' record you are considering will always be the latest point in time for each ID, the following should have good performance for a large number of records:
SELECT TOP 1
id
, DATEDIFF(s, MIN([Date]), MAX([date])) AS Elapsed
FROM #TableA
GROUP BY ID
ORDER BY DATEDIFF(s, MIN([Date]), MAX([date])) DESC
Related
I am trying to find a solution for the following issue that I have in sql-server:
I have one table t1 of which I want to use each date for each agency and loop it through the query to find out the avg_rate. Here is my table t1:
Table T1:
+--------+-------------+
| agency | end_date |
+--------+-------------+
| 1 | 2017-10-01 |
| 2 | 2018-01-01 |
| 3 | 2018-05-01 |
| 4 | 2012-01-01 |
| 5 | 2018-04-01 |
| 6 | 2017-12-01l |
+--------+-------------+
I literally want to use all values in the column end_date and plug it into the query here (I marked it with ** **):
with averages as (
select a.id as agency
,c.rate
, avg(c.rate) over (partition by a.id order by a.id ) as avg_cost
from table_a as a
join rates c on a.rate_id = c.id
and c.end_date = **here I use all values from t1.end_date**
and c.Start_date = **here I use all values from above minus half a year** = dateadd(month,-6,end_date)
group by a.id
,c.rate
)
select distinct agency, avg_cost from averages
order by 1
The reason why I need two dynamic dates is that the avg_rates vary if you change the timeframe between these dates.
My problem and my question is now:
How can you take the end_date from table t1 plug it into the query where c.end_date is and loop if through all values in t1.end_date?
I appreciate your help!
Do you really need a windowed average? Try this out.
;with timeRanges AS
(
SELECT
T.end_date,
start_date = dateadd(month,-6, T.end_date)
FROM
T1 AS T
)
select
a.id as agency,
c.rate,
T.end_date,
T.start_date,
avg_cost = avg(c.rate)
from
table_a as a
join rates c on a.rate_id = c.id
join timeRanges AS T ON A.DateColumn BETWEEN T.start_date AND T.end_date
group by
a.id ,
c.rate,
T.end_date,
T.start_date
You need a date column to join your data against T1 (I called it DateColumn in this example), otherwise all time ranges would return the same averages.
I can think of several ways to do this - Cursor, StoredProcedure, Joins ...
Given the simplicity of your query, a cartesian product (Cross Join) of Table T1 against the averages CTE should do the magic.
Trying to update one column, from another table with the highest Date.
Table 1 Example:
PartNumber | Cost
1000 | .10
1001 | .20
Table 2 Example:
PartNumber | Cost | Date
1000 | .10 | 2017-01-01
1000 | .50 | 2017-02-01
1001 | .20 | 2017-01-01
1002 | .50 | 2017-02-02
I would like to update table 1 with the most recent values from table2, which would be .50 for each... The query I use to update this has worked just fine until I realized I was not grabbing the correct Cost because there were multiples.. I now want to grab the highest dated revision.
My query:
UPDATE dex_mfgx..insp_master
SET dex_mfgx..insp_master.costperpart = t2.sct_cst_tot
FROM dex_mfgx..insp_master AS t1
INNER JOIN qad_repl..sct_det_sql AS t2
ON t1.partnum = t2.sct_part
WHERE t1.partnum = t2.sct_part and t2.sct_cst_date = MAX(t2.sct_cst_date) ;
My Error:
Msg 147, Level 15, State 1, Line 6
An aggregate may not appear in the WHERE clause unless it is in a subquery contained in a HAVING clause or a select list, and the column being aggregated is an outer reference.
Not having much luck with HAVING or GROUPING, although I havent used them much..
Any have an idea that would help?
I think I understand what you are trying to solve now. Thanks to Lamak for setting me straight as I was way off base originally.
Something like this I think is what you are looking for.
with TotalCosts as
(
SELECT t2.sct_cst_tot
, t1.partnum
, RowNum = ROW_NUMBER() over(partition by t1.partnun order by t2.sct_cst_date desc)
FROM dex_mfgx..insp_master AS t1
INNER JOIN qad_repl..sct_det_sql AS t2 ON t1.partnum = t2.sct_part
)
update t1
set costperpart = tc.sct_cst_tot
from dex_mfgx..insp_master AS t1
join TotalCosts tc on tc.partnum = t1.partnum
where tc.RowNum = 1
In our company, our clients perform various activities that we log in different tables - Interview attendance, Course Attendance, and other general activities.
I have a database view that unions data from all of these tables giving us the ActivityView that looks like this.
As you can see some activities overlap - for example while attending an interview, a client may have been performing a CV update activity.
+----------------------+---------------+---------------------+-------------------+
| activity_client_id | activity_type | activity_start_date | activity_end_date |
+----------------------+---------------+---------------------+-------------------+
| 112 | Interview | 2015-06-01 09:00 | 2015-06-01 11:00 |
| 112 | CV updating | 2015-06-01 09:30 | 2015-06-01 11:30 |
| 112 | Course | 2015-06-02 09:00 | 2015-06-02 16:00 |
| 112 | Interview | 2015-06-03 09:00 | 2015-06-03 10:00 |
+----------------------+---------------+---------------------+-------------------+
Each client has a "Sign Up Date", recorded on the client table, which is when they joined our programme. Here it is for our sample client:
+-----------+---------------------+
| client_id | client_sign_up_date |
+-----------+---------------------+
| 112 | 2015-05-20 |
+-----------+---------------------+
I need to create a report that will show the following columns:
+-----------+---------------------+--------------------------------------------+
| client_id | client_sign_up_date | date_client_completed_5_hours_of_activity |
+-----------+---------------------+--------------------------------------------+
We need this report in order to see how effective our programme is. An important aim of the programme is that we get every client to complete at least 5 hours of activity as quickly as possible.
So this report will tell us how long from sign up does it take each client to achieve this figure.
What makes this even trickier is that when we calculate 5 hours of total activity, we must discount overlapping activities:
In the sample data above the client attended an interview between 09:00 and 11:00.
On the same day they also performed CV updating activity from 09:30 to 11:30.
For our calculation, this would give them total activity for the day of 2.5 hours (150 minutes) - we would only count 30 minutes of the CV updating as the Interview overlaps it up to 11:00.
So the report for our sample client would give the following result:
+-----------+---------------------+--------------------------------------------+
| client_id | client_sign_up_date | date_client_completed_5_hours_of_activity |
+-----------+---------------------+--------------------------------------------+
| 112 | 2015-05-20 | 2015-06-02 |
+-----------+---------------------+--------------------------------------------+
So my question is how can I create the report using a select statement ?
I can work out how to do this by writing a stored procedure that will loop through the view and write the result to a report table.
But I would much prefer to avoid a stored procedure and have a select statement that will give me the report on the fly.
I am using SQL Server 2005.
See SQL Fiddle here.
with tbl as (
-- this will generate daily merged ovelaping time
select distinct
a.id
,(
select min(x.starttime)
from act x
where x.id=a.id and ( x.starttime between a.starttime and a.endtime
or a.starttime between x.starttime and x.endtime )
) start1
,(
select max(x.endtime)
from act x
where x.id=a.id and ( x.endtime between a.starttime and a.endtime
or a.endtime between x.starttime and x.endtime )
) end1
from act a
), tbl2 as
(
-- this will add minute and total minute column
select
*
,datediff(mi,t.start1,t.end1) mi
,(select sum(datediff(mi,x.start1,x.end1)) from tbl x where x.id=t.id and x.end1<=t.end1) totalmi
from tbl t
), tbl3 as
(
-- now final query showing starttime and endtime for 5 hours other wise null in case not completed 5(300 minutes) hours
select
t.id
,min(t.start1) starttime
,min(case when t.totalmi>300 then t.end1 else null end) endtime
from tbl2 t
group by t.id
)
-- final result
select *
from tbl3
where endtime is not null
This is one way to do it:
;WITH CTErn AS (
SELECT activity_client_id, activity_type,
activity_start_date, activity_end_date,
ROW_NUMBER() OVER (PARTITION BY activity_client_id
ORDER BY activity_start_date) AS rn
FROM activities
),
CTEdiff AS (
SELECT c1.activity_client_id, c1.activity_type,
x.activity_start_date, c1.activity_end_date,
DATEDIFF(mi, x.activity_start_date, c1.activity_end_date) AS diff,
ROW_NUMBER() OVER (PARTITION BY c1.activity_client_id
ORDER BY x.activity_start_date) AS seq
FROM CTErn AS c1
LEFT JOIN CTErn AS c2 ON c1.rn = c2.rn + 1
CROSS APPLY (SELECT CASE
WHEN c1.activity_start_date < c2.activity_end_date
THEN c2.activity_end_date
ELSE c1.activity_start_date
END) x(activity_start_date)
)
SELECT TOP 1 client_id, client_sign_up_date, activity_start_date,
hoursOfActivicty
FROM CTEdiff AS c1
INNER JOIN clients AS c2 ON c1.activity_client_id = c2.client_id
CROSS APPLY (SELECT SUM(diff) / 60.0
FROM CTEdiff AS c3
WHERE c3.seq <= c1.seq) x(hoursOfActivicty)
WHERE hoursOfActivicty >= 5
ORDER BY seq
Common Table Expressions and ROW_NUMBER() were introduced with SQL Server 2005, so the above query should work for that version.
Demo here
The first CTE, i.e. CTErn, produces the following output:
client_id activity_type start_date end_date rn
112 Interview 2015-06-01 09:00 2015-06-01 11:00 1
112 CV updating 2015-06-01 09:30 2015-06-01 11:30 2
112 Course 2015-06-02 09:00 2015-06-02 16:00 3
112 Interview 2015-06-03 09:00 2015-06-03 10:00 4
The second CTE, i.e. CTEdiff, uses the above table expression in order to calculate time difference for each record, taking into consideration any overlapps with the previous record:
client_id activity_type start_date end_date diff seq
112 Interview 2015-06-01 09:00 2015-06-01 11:00 120 1
112 CV updating 2015-06-01 11:00 2015-06-01 11:30 30 2
112 Course 2015-06-02 09:00 2015-06-02 16:00 420 3
112 Interview 2015-06-03 09:00 2015-06-03 10:00 60 4
The final query calculates the cumulative sum of time difference and selects the first record that exceeds 5 hours of activity.
The above query will work for simple interval overlaps, i.e. when just the end date of an activity overlaps the start date of the next activity.
A Geometric Approach
For another issue, I've taken a geometric approach to date
packing. Namely, I convert dates and times to a sql geometry
type and utilize geometry::UnionAggregate to merge the ranges.
I don't believe this will work in sql-server 2005. But your
problem was such an interesting puzzle that I wanted to see
whether the geometrical approach would work. So any future
users running into this problem that have access to a later
version can consider it.
Code Description
In 'numbers':
I build a table representing a sequence
Swap it out with your favorite way to make a numbers table.
For a union operation, you won't ever need more rows than in
your original table, so I just use it as the base to build it.
In 'mergeLines':
I convert the dates to floats and use those floats
to create geometrical points.
I then connect these points via STUnion and STEnvelope.
Finally, I merge all these lines via UnionAggregate. The resulting
'lines' geometry object might contain multiple lines, but if they
overlap, they turn into one line.
In 'redate':
I use the numbers CTE to extract the individual lines inside 'lines'.
I envelope the lines which here ensures that the lines are stored
only as its two endpoints.
I read the endpoint x values and convert them back to their time
representations (This is usually the end goal, but you need more).
I calculate the difference in minutes between activity start and
end dates (I do this first in seconds then divide by 60 for the
sake of a precision issue).
I calculate the cumulative sume of these minutes for each row.
In the outer query:
I align the previous cumulative minutes sum with each current row
I filter for the row where the 5hr goal was met but where the
previous minutes shows that the 5hr goal for the previous row
was not met.
I then calculate where in the current row's range the user has
met the 5 hours, to not only arrive at the date the five hour
goal was met, but the exact time.
The Code
with
numbers as (
select row_number() over (order by (select null)) i
from #activities -- where I put your data
),
mergeLines as (
select activity_client_id,
lines = geometry::UnionAggregate(line)
from #activities
cross apply (select
startP = geometry::Point(convert(float,activity_start_date), 0, 0),
stopP = geometry::Point(convert(float,activity_end_date), 0, 0)
) pointify
cross apply (select line = startP.STUnion(stopP).STEnvelope()) lineify
group by activity_client_id
),
redate as (
select client_id = activity_client_id,
activities_start_date,
activities_end_date,
minutes,
rollingMinutes = sum(minutes) over(
partition by activity_client_id
order by activities_start_date
rows between unbounded preceding and current row
)
from mergeLines ml
join numbers n on n.i between 1 and ml.lines.STNumGeometries()
cross apply (select line = ml.lines.STGeometryN(i).STEnvelope()) l
cross apply (select
activities_start_date = convert(datetime, l.line.STPointN(1).STX),
activities_end_date = convert(datetime, l.line.STPointN(3).STX)
) unprepare
cross apply (select minutes =
round(datediff(s, activities_start_date, activities_end_date) / 60.0,0)
) duration
)
select client_id,
activities_start_date,
activities_end_date,
met_5hr_goal = dateadd(minute, (60 * 5) - prevRoll, activities_start_date)
from (
select *,
prevRoll = lag(rollingMinutes) over (
partition by client_id
order by rollingMinutes
)
from redate
) ranker
where rollingMinutes >= 60 * 5
and prevRoll < 60 * 5;
I have this query through an odbc connection in excel for a refreshable report with data for every 4 weeks. I need to show the dates in each of the 4 weeks even if there is no data for that day because this data is then linked to a Graph. Is there a way to do this?
thanks.
Select b.INV_DT, sum( a.ORD_QTY) as Ordered, sum( a.SHIPPED_QTY) as Shipped
from fct_dly_invoice_detail a, fct_dly_invoice_header b, dim_invoice_customer c
where a.INV_HDR_SK = b.INV_HDR_SK
and b.DIM_INV_CUST_SK = c.DIM_INV_CUST_SK
and a.SRC_SYS_CD = 'ABC'
and a.NDC_NBR is not null
**and b.inv_dt between CURRENT_DATE - 16 and CURRENT_DATE**
and b.store_nbr in (2851, 2963, 3249, 3385, 3447, 3591, 3727, 4065, 4102, 4289, 4376, 4793, 5209, 5266, 5312, 5453, 5569, 5575, 5892, 6534, 6571, 7110, 9057, 9262, 9652, 9742, 10373, 12392, 12739, 13870
)
group by 1
The general purpose solution to this is to create a date dimension table, and then perform an outer join to that date dimension table on the INV_DT column.
There are tons of good resources you can search for on creating a good date dimension table, so I'll just create a quick and dirty (and trivial) example here. I highly recommend some research in that area if you'll be doing a lot of BI/reporting.
If our table we want to report from looks like this:
Table "TABLEZ"
Attribute | Type | Modifier | Default Value
-----------+--------+----------+---------------
AMOUNT | BIGINT | |
INV_DT | DATE | |
Distributed on random: (round-robin)
select * from tablez order by inv_dt
AMOUNT | INV_DT
--------+------------
1 | 2015-04-04
1 | 2015-04-04
1 | 2015-04-06
1 | 2015-04-06
(4 rows)
and our report looks like this:
SELECT inv_dt,
SUM(amount)
FROM tablez
WHERE inv_dt BETWEEN CURRENT_DATE - 5 AND CURRENT_DATE
GROUP BY inv_dt;
INV_DT | SUM
------------+-----
2015-04-04 | 2
2015-04-06 | 2
(2 rows)
We can create a date dimension table that contains a row for every date (or ate last 1024 days in the past and 1024 days in the future using the _v_vector_idx view in this example).
create table date_dim (date_dt date);
insert into date_dim select current_date - idx from _v_vector_idx;
insert into date_dim select current_date + idx +1 from _v_vector_idx;
Then our query would look like this:
SELECT d.date_dt,
SUM(amount)
FROM tablez a
RIGHT OUTER JOIN date_dim d
ON a.inv_dt = d.date_dt
WHERE d.date_dt BETWEEN CURRENT_DATE -5 AND CURRENT_DATE
GROUP BY d.date_dt;
DATE_DT | SUM
------------+-----
2015-04-01 |
2015-04-02 |
2015-04-03 |
2015-04-04 | 2
2015-04-05 |
2015-04-06 | 2
(6 rows)
If you actually needed a zero value instead of a NULL for the days where you had no data, you could use a COALESCE or NVL like this:
SELECT d.date_dt,
COALESCE(SUM(amount),0)
FROM tablez a
RIGHT OUTER JOIN date_dim d
ON a.inv_dt = d.date_dt
WHERE d.date_dt BETWEEN CURRENT_DATE -5 AND CURRENT_DATE
GROUP BY d.date_dt;
DATE_DT | COALESCE
------------+----------
2015-04-01 | 0
2015-04-02 | 0
2015-04-03 | 0
2015-04-04 | 2
2015-04-05 | 0
2015-04-06 | 2
(6 rows)
I agree with #ScottMcG that you need to get the list of dates. However if you are in a situation where you aren't allowed to create a table. You can simplify things. All you need is a table that has at least 28 rows. Using your example, this should work.
select date_list.dt_nm, nvl(results.Ordered,0) as Ordered, nvl(results.Shipped,0) as Shipped
from
(select row_number() over(order by sub.arb_nbr)+ (current_date -28) as dt_nm
from (select rowid as arb_nbr
from fct_dly_invoice_detail b
limit 28) sub ) date_list left outer join
( Select b.INV_DT, sum( a.ORD_QTY) as Ordered, sum( a.SHIPPED_QTY) as Shipped
from fct_dly_invoice_detail a inner join
fct_dly_invoice_header b
on a.INV_HDR_SK = b.INV_HDR_SK
and a.SRC_SYS_CD = 'ABC'
and a.NDC_NBR is not null
**and b.inv_dt between CURRENT_DATE - 16 and CURRENT_DATE**
and b.store_nbr in (2851, 2963, 3249, 3385, 3447, 3591, 3727, 4065, 4102, 4289, 4376, 4793, 5209, 5266, 5312, 5453, 5569, 5575, 5892, 6534, 6571, 7110, 9057, 9262, 9652, 9742, 10373, 12392, 12739, 13870)
inner join
dim_invoice_customer c
on b.DIM_INV_CUST_SK = c.DIM_INV_CUST_SK
group by 1 ) results
on date_list.dt_nm = results.inv_dt
This process has several steps which are reflected in various tables of a database:
Production --> UPDATE to the inventory table using something like
UPDATE STOR SET
STOR.BLOC1 = T.BLOC1,
STOR.BLOC2 = T.BLOC2,
STOR.BLOC3 = T.BLOC3,
STOR.PRODUCTION = T.PROD,
STOR.DELTA = T.DELTA
FROM BLDG B INNER JOIN STOR S
ON S.B_ID = B.B_ID
CROSS APPLY dbo.INVENTORIZE(B.B_ID) AS T;
The above feeds a log table with a TRIGGER like this:
CREATE TRIGGER trgrCYCLE
ON STOR
FOR UPDATE
AS
INSERT INTO dbo.INVT
(TS, BLDG, PROD, ACT, VAL)
SELECT CURRENT_TIMESTAMP, B_ID, PRODUCTION,
CASE WHEN DELTA < 0 THEN 'SELL' ELSE 'BUY' END,
DELTA
FROM inserted WHERE COALESCE(DELTA,0) <> 0
And finally, every update should INSERT a row into a financials table which I added to the TRIGGER above:
INSERT INTO dbo.FINS
(COMPANY, TS, COST2, BAL)
SELECT CORP, CURRENT_TIMESTAMP, COST,
((SELECT TOP 1 BAL FROM FINS WHERE COMPANY = CORP ORDER BY TS DESC)- COST)
FROM inserted WHERE COALESCE(COST,0) <> 0
The problem is with this line:
((SELECT TOP 1 BAL FROM FINS WHERE COMPANY = CORP ORDER BY TS DESC)- COST)
which is meant to calculate the latest balance of an account. But because the CROSS APPLY treats all the INSERTS as a batch, the calculation is done off of the same last record and I get an incorrect balance figure. Example:
COST BALANCE
----------------
1,000 <-- initial balance
-150 850
-220 780 <-- should be 630
What would be the way to solve that? A trigger on the FINS table instead for the balance calculation?
Understanding existing logic in your query
UPDATE statement will fire a trigger only once for a set or batch satisfying join conditions, Inserted statement will have all the records that are being updated. This is because of BATCH processing not because of CROSS APPLY but because of UPDATE.
In this query of yours
SELECT CORP, CURRENT_TIMESTAMP, COST,
((SELECT TOP 1 BAL FROM FINS WHERE COMPANY = CORP ORDER BY TS DESC)- COST)
FROM inserted WHERE COALESCE(COST,0) <> 0
For each CORP from an Outer query, same BAL will be returned.
(SELECT TOP 1 BAL FROM FINS WHERE COMPANY = CORP ORDER BY TS DESC)
That being said, your inner query will be replaced by 1000(value you used in your example) every time CORP = 'XYZ'
SELECT CORP, CURRENT_TIMESTAMP, COST, (1000- COST)
FROM inserted WHERE COALESCE(COST,0) <> 0
Now your inserted statement has all the records that are being inserted. So every record's cost will be subtracted by 1000. Hence you are getting unexpected result.
Suggested solution
As per my understanding, you want to calculate some cumulative frequency kind of thing. Or last running total
Data Preparation for problem statement. Used my dummy data to give you an idea.
--Sort data based on timestamp in desc order
SELECT PK_LoginId AS Bal, FK_RoleId AS Cost, AddedDate AS TS
, ROW_NUMBER() OVER (ORDER BY AddedDate DESC) AS Rno
INTO ##tmp
FROM dbo.M_Login WHERE AddedDate IS NOT NULL
--Check how data looks
SELECT Bal, Cost, Rno, TS FROM ##tmp
--Considering ##tmp as your inserted table,
--I just added Row_Number to apply Top 1 Order by desc logic
+-----+------+-----+-------------------------+
| Bal | Cost | Rno | TS |
+-----+------+-----+-------------------------+
| 172 | 10 | 1 | 2012-12-05 08:16:28.767 |
| 171 | 10 | 2 | 2012-12-04 14:36:36.483 |
| 169 | 12 | 3 | 2012-12-04 14:34:36.173 |
| 168 | 12 | 4 | 2012-12-04 14:33:37.127 |
| 167 | 10 | 5 | 2012-12-04 14:31:21.593 |
| 166 | 15 | 6 | 2012-12-04 14:30:36.360 |
+-----+------+-----+-------------------------+
Alternative logic for subtracting cost from last running balance.
--Start a recursive query to subtract balance based on cost
;WITH cte(Bal, Cost, Rno)
AS
(
SELECT t.Bal, 0, t.Rno FROM ##tmp t WHERE t.Rno = 1
UNION ALL
SELECT c.Bal - t.Cost, t.Cost, t.Rno FROM ##tmp t
INNER JOIN cte c ON t.RNo - 1 = c.Rno
)
SELECT * INTO ##Fin FROM cte;
SELECT * FROM ##Fin
Output
+-----+------+-----+
| Bal | Cost | Rno |
+-----+------+-----+
| 172 | 0 | 1 |
| 162 | 10 | 2 |
| 150 | 12 | 3 |
| 138 | 12 | 4 |
| 128 | 10 | 5 |
| 113 | 15 | 6 |
+-----+------+-----+
You have to tweet your columns little bit to get this functionality into your trigger.
I think you can try a trigger on the Fins.
You can use IDENT_CURRENT('Table')) to take the last primary key from the table and make a select.
I think it's better than "select top 1".
To to take the last balance value, set a variable last_bal = select bal from FINS where primary_key = Ident_Current("FINS")
well
first sql is a game where it work with groups or rather "set" so always you have think about that.
if you work with a simple item is correct, it maybe be better approach
declare #myinsert table(id int identity(1,1), company VArchar(35), ts datetime, cost2 smallmoney, bal smallmoney)
insert into #myinsert(company,ts, cost2, bal)
SELECT CORP, CURRENT_TIMESTAMP, COST,
FROM inserted WHERE COALESCE(COST,0) <> 0
declare #current int
select #current = min(id) from #myinsert
while exists(select * from #myinsert where id = #current)
begin
INSERT INTO dbo.FINS
(COMPANY, TS, COST2, BAL)
SELECT COMPANY, CURRENT_TIMESTAMP, COST,
((SELECT TOP 1 BAL FROM FINS WHERE COMPANY = my.COMPANY ORDER BY TS DESC)- COST)
from #myinsert my where id = #current
select #current = min(id) from #myinsert where id > #current
end
i am not giving you exact query .For a moment forget trigger.Because you are unable to test your query .
I suggest to use Output clause .This will atleast help you to construct proper query and test it.
this query is running ok,(if you can use merge then that is best).
Declare #t table
(
BLOC1,BLOC2,BLOC3 ,PRODUCTION ,DELTA --whatever column is require here
)
UPDATE STOR SET
STOR.BLOC1 = T.BLOC1,
STOR.BLOC2 = T.BLOC2,
STOR.BLOC3 = T.BLOC3,
STOR.PRODUCTION = T.PROD,
STOR.DELTA = T.DELTA
Output inserted.BLOC1 ,inserted.BLOC2, and so on into #t
FROM BLDG B INNER JOIN STOR S
ON S.B_ID = B.B_ID
CROSS APPLY dbo.INVENTORIZE(B.B_ID) AS T;
now you have inserted value in table variable #t
SELECT CORP, CURRENT_TIMESTAMP, COST,
BAL,Row_Number() over(partition by company order by TS desc) RN
FROM #t inner join FINS on COMPANY = CORP
WHERE COALESCE(COST,0) <> 0
Verify this query till here.Think of optimizing or trigger later on.
I think i gave good suggestion.and I guess subtraction is not a problem.I am telling to put everything in output clause and analyze the query and test it.
you can use CTE inside trigger also but how will you test it.
;With CTE as
(
SELECT CORP, CURRENT_TIMESTAMP, COST,BAL
ROW_NUMBER()over(ORDER BY TS DESC )rn
FROM inserted
inner join FINS on COMPANY = CORP
WHERE COALESCE(COST,0) <> 0
)
select * from CTE --check this what you are getting
Something like that, Isn't complete.
CREATE TRIGGER trgrCYCLE
ON STOR
FOR UPDATE
AS
begin
declare #last_bal int
declare #company varchar(50)
declare #ts --type
declare #cost int
declare #bal --type
--etc whatever you need
select #company = company, #ts= ts , #cost = cost , #bal = bal from INSERTED
--others selects and sets
set #last_bal = select bal from dbo.FINS where you_primary_key = IDENT_CURRENT('FINS'))
set #last_bal = #last_bal - #cost
Insert INTO FINS (company, ts, cost2, bal) VALUES (#company, #ts, #cost, #last_bal) where --your conditions
end
If, similar to #Shantanu's method, you could associate a sequence with inserted, the virtual table associated with the trigger you could do this by subtracting all the COSTs that come before the current record.
This could be accomplished by adding a rowversion to STOR, which will be updated automatically with each delete.
Then instead of:
((SELECT TOP 1 BAL FROM FINS WHERE COMPANY = CORP ORDER BY TS DESC)- COST)
from inserted ...
make the rowversion RV, and:
(SELECT SUM(X.B) FROM
(SELECT TOP 1 BAL B
FROM FINS
WHERE COMPANY = CORP
ORDER BY TS DESC
UNION
SELECT -COST B
FROM inserted ii
WHERE ii.RV >= i.RV AND ii.CORP = i.CORP
) AS X)
FROM inserted i WHERE COALESCE(COST,0) <> 0
Should do what you want. You could conceivably do this with a timestamp that was more find-grained than CURRENT_TIMESTAMP which, I believe, goes down only to seconds but that requires you update it in the UPDATE statement. The rowversion may cause problems with your STOR insert statements.