Transact SQL :: Adding Case When Statement With Group By Query Doesn't Aggregate Records
Aug 28, 2015
I have a a Group By query which is working fine aggregating records by city. Now I have a requirement to focus on one city and then group the other cities to 'Other'. Here is the query which works:
Select [City]= CASE WHEN [City] = 'St. Louis' THEN 'St. Louis' ELSE 'Other Missouri City' END, SUM([Cars]) AS 'Total Cars'
From [Output-MarketAnalysis]
Where [City] IN ('St. Louis','Kansas City','Columbia', 'Jefferson City','Joplin') AND [Status] = 'Active'
Group by [City]
Here is the result:
St. Louis 1000
Kansas City 800
Columbia 700
Jefferson City 650
Joplin 300
When I add this Case When statement to roll up the city information it changes the name of the city to 'Other Missouri City' however it does not aggregate all Cities with the value 'Other Missouri City':
Select [City]= CASE WHEN [City] = 'St. Louis' THEN 'St. Louis' ELSE 'Other Missouri City' END, SUM([Cars]) AS 'Total Cars'
From [Output-MarketAnalysis]
Where [City] IN ('St. Louis','Kansas City','Columbia', 'Jefferson City','Joplin') AND [Status] = 'Active'
Group by [City]
Here is the result:
St. Louis 1000
Other Missouri City 800
Other Missouri City 700
Other Missouri City 650
Other Missouri City 300
If I just use a simple select statement, I find that I have 8286 records within a specified date range.
If I use the select statement to pull records that were created from 5pm and later and then add it to another select statement with records created before 5pm, I get a different count: 7521 + 756 = 8277
Is there something I am doing incorrectly in the following sql?
DECLARE @startdate date = '03-06-2015' DECLARE @enddate date = '10-31-2015' DECLARE @afterTime time = '17:00' SELECT General_Count = (SELECT COUNT(*) as General FROM Unidata.CrumsTicket ct
I've always been mistified why you can't use a column alias in the group byclause (i.e. you have to re-iterate the entire expression in the group byclause after having already done it once in the select statement). I'mmostly a SQL hobbiest, so it's possible that I am not doing this in the mostefficient manner. Anyone care to comment on this with relation to thefollowing example (is there a way to acheive this without re-stating theentire CASE statement again in the Group By clause?):Select 'Age' =CaseWHEN(SubmittedOn >= DATEADD(dd, - 30, GETDATE())) AND (SubmittedOn <=DATEADD(dd, - 0, GETDATE()))THEN '0-30 Days Old'WHEN(SubmittedOn >= DATEADD(dd, - 60, GETDATE())) AND (SubmittedOn <=DATEADD(dd, - 31, GETDATE())) Then '31-60 Days Old'WHEN(SubmittedOn >= DATEADD(dd, - 90, GETDATE())) AND (SubmittedOn <=DATEADD(dd, - 61, GETDATE())) Then '61-90 Days Old'WHEN(SubmittedOn <= DATEADD(dd, - 91, GETDATE())) THEN '91+ Days Old'ELSE cast(SubmittedOn as varchar(22))end,max(SubmittedOn), COUNT(SCRID) AS NbrSCRsFrom SCRViewWHERE(StatusSort < 90) ANDCustomerID = 8 andUserID = 133group byCaseWHEN(SubmittedOn >= DATEADD(dd, - 30, GETDATE())) AND (SubmittedOn <=DATEADD(dd, - 0, GETDATE()))THEN '0-30 Days Old'WHEN(SubmittedOn >= DATEADD(dd, - 60, GETDATE())) AND (SubmittedOn <=DATEADD(dd, - 31, GETDATE())) Then '31-60 Days Old'WHEN(SubmittedOn >= DATEADD(dd, - 90, GETDATE())) AND (SubmittedOn <=DATEADD(dd, - 61, GETDATE())) Then '61-90 Days Old'WHEN(SubmittedOn <= DATEADD(dd, - 91, GETDATE())) THEN '91+ Days Old'ELSE cast(SubmittedOn as varchar(22))endOrder by max(submittedon) descThanks,Chad
I was asked to add an additional column to an existing query. I'm using Microsoft Query with a MS SQL 2000 server, and don't have much knowledge of SQL in general. Here's the existing query:
SELECT A.COMPANYCODE, A.INVOICENUMBER, A.LINENUMBER, A.SONUMBER, A.CUSTOMERCODE, A.SHIPPERNUMBER, A.INVOICEDATE, A.ITEMCODE, A.QUANTITYINVOICED, A.UNITPRICE AS 'InvPrice', A.QUANTITYINVOICED * A.UNITPRICE AS 'ExtInvPrice', INVENTORY.UNITPRICE AS 'StdPrice', INVENTORY.STANDARDCOST, A.QUANTITYINVOICED * INVENTORY.STANDARDCOST AS 'ExtCost', (A.QUANTITYINVOICED * A.UNITPRICE) - (A.QUANTITYINVOICED * INVENTORY.STANDARDCOST) AS 'GM$', (INVENTORY.UNITPRICE - A.UNITPRICE) * -1 AS 'PriceListDiff' FROM ABW.DBO.SALESANALYSISHISTORY A, ABW.DBO.INVENTORY INVENTORY WHERE INVENTORY.COMPANYCODE = A.COMPANYCODE AND INVENTORY.ITEMCODE = A.ITEMCODE AND ((A.COMPANYCODE = 'csp') AND (A.QUANTITYINVOICED <>$ 0) AND (A.INVOICEDATE BETWEEN '03/1/08' AND '03/31/08')) ORDER BY A.INVOICEDATE, A.ITEMCODE
They want a column added to the current query where if A.Unitprice is greater than or equal to Inventory.UnitPrice then populate the column with A.QuantityInvoiced*A.UnitPrice. I posted on another forum, and the advice I got was to add this:
SELECT NewColumn = CASE WHEN A.Unitprice >= Inventory.UnitPrice THEN A.QuantityInvoiced * A.Unitprice ELSE 'null' END, FROM ABW.DBO.SALESANALYSISHISTORY A, ABW.DBO.INVENTORY INVENTORY
I'm not sure how to integrate it to the current query, so I've tried running it by itself, and I get the error: Didn't expect 'A' after the SELECT column list.
Any help would be greatly appreciated to modify the current query to display the new column.
Below is my SQl which just counts the number of appointments and grouped by clinic. This is great but what I'd like to add is the percentage within each clinic.
For example Clinic BRESRAD1 has a total of 61 appointments, of which 75.41% are Normal Appointments and 24.59% are Diagnostic, Ideally I would like the percentage in the next column.
BRESRAD1 Normal Appointment 46 BRESRAD1 Diagnostic Appointment 15 BRESRAD2 Normal Appointment 17 BRESRAD2 Diagnostic Appointment 12 BRESRAD3 Normal Appointment 34 BRESRAD3 Diagnostic Appointment 43
My SQL is as follows:
SELECT ClinicCode, CASE WHEN [ApptTypeDesc] LIKE '%Diag%' THEN 'Diagnostic Appointment' ELSE 'Normal Appointment' END AS [Diagnostic Appt], COUNT(OPAppointmentID) AS CountOfOPAppointmentID FROM dbo.OP_APPOINTMENT WHERE (AttendStatusNatCode IN ('5', '6')) AND (ApptFinYr = '2014/15') GROUP BY ClinicCode, CASE WHEN [ApptTypeDesc] LIKE '%Diag%' THEN 'Diagnostic Appointment' ELSE 'Normal Appointment' END ORDER BY ClinicCode
table A: | ID | FRUIT | VEGETABLE | GOOD | -------------------------------------------- | 1 | orange | cabbage | no | | 1 | apple | lettuce | yes | | 1 | kiwi | broccoli | no | | 1 | pear | kale | yes |
table B: | ID | FRUIT | VEGETABLE | ------------------------------- | 1 | apple | lettuce | | 2 | pear | kale |
If the fruit and vegetable in table A is found in table B, then set the GOOD column = yes, else no.
This is what I have so far.
update tableA set GOOD = (case when tableA.id = C.id then 'yes' else 'no' end ) from (select tableA.id as id from tableA A left join tableB B on B.fruit = A.fruit and B.vegetable = A.vegetable) C
I am using SQL 2005. I have some data from an old application that did not follow the rules for normalization. The table is for Invoices, and the table allows for 13 purchase items per record. So in each row of my table I have a non-unique integer field itemID, itemID1, itemID2 ... itemID12. For each itemID I also have "lbs_total" and "line_total" (which is price * lbs_total) - so itemID, lbs_total, line_total ... itemID1, lbs_total1, line_total1 ... etc. It's a mess, I know.Each row has a unique Customer Number ("cno") and an Invoice Date ("inv_date"). My proc needs to allow for params for the item number, and a start date and end date for BETWEEN on the inv_date.I also need to get the aggregate for the lbs_total and the line_total.
Well adding it to a group by or function skews the result set. How to write this query so it displays as I need it to? This is what I have thus far, and it works as it should UNTIL I add in the line of
cast(cte.[C] As float)/cast(sum(cte.[C]) over() as float)*100 As [Rate1],
Presents the error of: Msg 8120, Level 16, State 1, Line 35 Column 'cte.[C]' is invalid in the select list because it is not contained in either an aggregate function or the GROUP BY clause.
This is my full on query -- with 3 CTE's involved to get me the actual result set I am after.
;with cte as ( select [state], case when exists (select 1 from table2 R where R.centername = d.centername) then 1 else 0 end as [L], case when exists (select 1 from table3 C where C.centername = d.centername) then 1 else 0 end as [C] FROM maintable d ),
Tried a few ways and I can ge this to work at the end in then WHEN part. Just struggling to put this together to be accepted as a CASE WHEN statement, probably missing the obvious.
Case when Postcode like '%[abcdefghijklmnopqrstuvwxyz%]' then 'Lowercase Postcode' else 'Postcode OK' end as [DQPostcode]
collate Latin1_General_CS_AS
Simple terms looking for all instances of Lowercase characters in the Postcode field
I want to add an INNER JOIN based on a condition. If @Sports is not '' then I want to add the inner join statement...The following doesnt work:declare @Sports nvarchar(50)set @sports='1' SELECT DISTINCT tblUserData.UserName,tblUserData.UserCode,Description,IsDonator,IsVIP,BirthDate,IsMale,ShowAgeOnly,tblCountries.CountryPicture,case @LanguageColumnName when 'nl' then tblSexuality.nl when 'en' then tblSexuality.en when 'de' then tblSexuality.deendas sexualityname,tblCountries.CountryName,ThumbNailPicture,UserRanking,LastActivityDate,NearestBigCity,DistanceToNearestBigCity,ROW_NUMBER() OVER (ORDER BY IsVip DESC,UserRanking DESC,LastActivityDate DESC) as RowNum FROM aspnet_UsersINNER JOIN tblUserData ON aspnet_Users.UserId = tblUserData.UserIDINNER JOIN tblCountries ON tblUserData.Country=tblCountries.CountryID INNER JOIN tblSexuality ON tblUserData.Sexuality=tblSexuality.SexualityID ,case @Sportswhen <>'' INNER JOIN tblUsersAndSports ON tblUserData.UserCode=tblUsersAndSports.UserCodeend
SELECT COUNT(DISTINCT CASE WHEN visit_type = 0 THEN visitor_id END) AS [New Visitors], COUNT(DISTINCT CASE WHEN visit_type = 0 THEN visitor_id END) AS [Returning Visitors] FROM content_hits_tbl WHERE (hit_date BETWEEN DATEADD(mm, - 1, GETDATE()) AND GETDATE())
=======================
How do I add up both COUNT/CASE columns? Would it be: SUM([New Visitors] + [Returning Visitors]) AS Total
I tried this and it doesn't work. I get invalid column names error for both.
I have even tried: SUM([COUNT(DISTINCT CASE WHEN visit_type = 0 THEN visitor_id END)] + [COUNT(DISTINCT CASE WHEN visit_type = 0 THEN visitor_id END)]) AS Total
You would think that there would be some gui functionality in VS08 that would do this...
I tend to learn from example and am used to powershell. If for instance in powershell I wanted to get-something and store it in a variable I could, then use it again in the same code. In this example of a table order items where there are order_num, quantity and item_prices how could I declare ordertotal as a variable then instead of repeating it again at "having sum", instead use the variable in its place?
Any example of such a use of a variable that still lets me select the order_num, ordertotal and group them etc? I hope to simply replace in the "having section" the agg function with "ordertotal" which bombs out.
select order_num, sum(quantity*item_price) as ordertotal from orderitems group by order_num having sum(quantity*item_price) >=50 order by ordertotal;
Hi guys, I am having a little diffulty displaying two columns next to each other in a case/group by statement as code shown below. I was wondering if i could have the results displayed such that the gst_amount and total_amount are in two separate columns (as they currently are) however the results of the rows are in the same row not in separate rows (as they are currently). I dont think i can do this in the current case statement that i have (i.e: the two case statements). Any feedback would be appreciated
SELECT distinct PERIOD.STARTDATE, temp_111.EVENTTYPEID, case when temp_111.[name] like '%GST%' then sum(temp_111.CONTRIBUTIONVALUE) end as GST_AMOUNT, case when temp_111.[name] not like '%GST%' then sum(temp_111.CONTRIBUTIONVALUE) end as Total_Amount FROM temp_111 INNER JOIN PERIOD ON temp_111.PERIODSEQ = PERIOD.PERIODSEQ WHERE (NOT temp_111.PRODUCTID = 'IIIE' OR temp_111.PRODUCTID IS NULL) AND temp_111.PERIODSEQ in ('111') group by PERIOD.STARTDATE, temp_111.EVENTTYPEID, temp_111.[name]
material ========= material_id project_type project_id qty 1 AB Corporate 1 3 2 Other Project 2 7
i have taken AB Corporate for AB_Corporate_project ,Other Project for Other_project
sample query i write :--
select m.material_id ,m.project_type,m.project_id,m.qty,ab.ab_crp_id, ab.custname ,op.other_proj_id,op.other_custname,op. po case if m.project_type = 'AB Corporate' then select * from AB_Corporate_project where ab.ab_crp_id = m.project_id else if m.project_type = 'Other Project' then select * from Other_project where op.other_proj_id=m.project_id end from material m,AB_Corporate_project ab,Other_project op
but this query not work,also it gives errors
i want sql query to show data as follows
material_id project_type project_id custname other_custname qty 1 AB Corporate 1 abc -- 3 2 Other Project 2 -- dsd 7
so plz help me how can i write sql query for to show the output plz send a sql query
I've been beating my head against this for a bit and haven't been able to figure this out I want to pull 1 set of values between a date time range if @Report = '1' but if @Report = 2 or 3 I want it to drop 3 of the parameters so it will pull all items. by dropping the where parameters for @BeginRangeDate and @LookOutDate and L.COLineStatus. My report works great if I go in and run Report 1 with the date parameters in the where statement and also works great for Reports 2, 3 if I REM out those 3 items in the Where statement. Here is what I currently have
When I created a CASE statement (This is at work, Pat:)) it is about 30-40 lines long. I gave it a name and set the name = to the case statement:
ie,
SELECT fieldname1 = CASE WHEN condition THEN 'blah blah' WHEN condition THEN 'blah blah' WHEN condition THEN 'blah blah' ELSE thisandthat END , fieldname2 , fieldname3 FROM tablename1 GROUP BY CASE WHEN condition THEN 'blah blah' WHEN condition THEN 'blah blah' WHEN condition THEN 'blah blah' ELSE thisandthat END, , fieldname2, fieldname3
etc.
The long CASE statement in my GROUP BY is awkward to me. Is this the only way to do it? I tried using the fieldname1 but it comes back as an invalid field name and asks for the "expression".
I have a form with a dropdown or combo box, the user can select <All>or pick a user name. If they pick a user name my where clause worksfine, buts what's the best way to write "Select All" if they choosethe <All>This is what I have so far, but I don't think I should be using theLIKE operator.WHERE tblCase.qarep LIKE CASE @myqarep WHEN '<All>' THEN '%' ELSE@myqarep ENDand tblOffice.officecode LIKE CASE @myoffice WHEN -1 THEN '%' ELSE@myoffice ENDthanks for your help!!
In the below query, I can get the individual/single group by columns as well as multiple but I cannot control the order in which I would like to group by the data.
So lets say I want to group the data by OS->browser->browser_version(just one example) then I cannot achieve that as the order of OS column comes later in the query.
I know one option would be to write a dynamic SQL but i dont want to do that because of performance reasons. Any other way this can be achieved?
select case when @include_browser = 1 then browser_name end as browser_name, case when @include_browser_version = 1 then browser_version end as browser_version,
Is it possible to combine a CASE statement with two different columns to retrieve data into one result column? In one column it has multiple JobCode, but this needs to be divided. The only way I can see to do this is take the data from another column to get the results.Ex. JobCode - has one code for several job descriptions (there are about 30), but all within this code cannot have the same level of access. So I need to divide them out, and put them in one column for AccessLevel.
JobTitle - has one code for one job, (but there are over 100).I want to pull from both columns to get the results I need to assign appropriate access level in one column.
Case JobCode (they all have same job code, but everyone in this job code should not have same access) When '45' Then '1' (Principal, Asst. Prin, or any Administrator, Counselors) When '25' Then '2' (this could be teacher, etc. ) Case JobTitle (this is how access should be) When '12345' Then '1' (this is Administration only)
I am trying to run the below but I get an error of 'Incorrect syntax ')'' --- I have tried every angle I can think of around the parens to fix this but nothing I do is working.
UPDATE abcdefg SET [Date] = GETDate(), [readytogo] = ( CASE WHEN [customername] NOT IN (Select [customername] from [server].[database].[dbo].[view]) THEN 'Yes' ELSE 'Needs Verification'
Output should look like this..look for only Type=Accept until deny is reached. After Deny,if there is a Accept ignore it.
ID PersonID Date Type Qty 1 1 01/01/2011 Accept 5 (show only one Accept row=1 becoz Type is Accept and date is same,Qtyis same) 3 2 02/01/2010 Accept 10 (show Accept row=3,ignore deny row)
When i am running below snippet execution plan is showing constant scan instead of referring subquery table.
I want to know how this query working. and why in execution plan there is no scan /seek which will basically indicate that particular table is getting referred.
select count(*) from A where exists (select count(1) from B where A.a=B.a)
execution plan has to show scan or seek for subquery. Surprisingly, output is coming as expected.
I have used the below update query. However, its updating only the first value. Like its updating AB with volume when c.Type = ABC, similarly for CD. Its not updating based on the 2nd or the next case condition.
Update XYZ Set AB = a.Amt * (CASE WHEN c.Type = 'ABC' THEN (c.volume) WHEN c.TYPE = 'DEF' THEN (c.volume) WHEN c.Type = 'GHI' THEN (c.volume) Else 0 END), CD = CASE WHEN c.Type = 'MARGIN' THEN '4105.31' WHEN c.Type = 'ABC' THEN '123.1' WHEN c.Type = 'DEF' THEN '234.2' WHEN c.Type = 'GHI' THEN '567.1' END from table1 a join table2 b on a.Cust = b.Customer join table3 c on b.account = c.account and b.channel =c.channel
Why its not working properly? But if i use Select statement instead of update query its working properly.
I am trying to add the letters 'MS' in front of value while using a case statement. If Dispo = 2 I want it to pull back 'Inactive', else I want it to pull back the Value with MS in front (eg. "MS14"). The data in the value column are numbers. Would I use a CONCAT? If so where does that need to go?
Case when dispo = 2 then 'Inactive' else cast(Value as varchar(11)) end ,
I have 20 rows in dbo.test so i will get 20 as my output, now i need to write a case statement here such that when count(*) = 0 then it should display text filed 'NO Data' else it should display the count.
I need to perform an update where there are multiple scenarios which determine the value that is entered. Below is a sort've psuedo code of how it needs to be.
Update MyTable SET MyColumn = CASE WHEN MyCol1 = 'Value1' Then NewValue Else WHEN MyCol1 <> 'Value1' And MyCol2 = 'Active' Then 'Value1'
In the scenario where MyCol1 <> Value1 and MyCol2 <> 'Active' then no update would occur and the original value would remain intact.
