create table Test (a varchar(11));
create table Test2 (a varchar(11));
insert into Test values('a'),('b'),('c');
This one works just fine
delete from Test output deleted.a into Test2(a)
However this one doesn't work
delete t from Test t output deleted.a into Test2(a)
Incorrect syntax near 'OUTPUT'.
I want to be able to use a table alias there since the actual query is something like this:
DELETE t
FROM Test t
LEFT JOIN blah blah
WHERE blah blah
OUTPUT DELETED.a INTO Test2(a)
The syntax you want is DELETE <alias> OUTPUT ... FROM ...:
DELETE T
OUTPUT deleted.a INTO dbo.Test2(a)
FROM dbo.Test T;
So, for your LEFT JOIN example, it would be:
DELETE t
OUTPUT DELETED.a
INTO Test2(a)
FROM dbo.Test t
LEFT JOIN dbo.blah b on t.a = b.b --A lEFT JOIN on a simple DELETE seems odd mind
WHERE t.c = 2;
Related
I am trying to copy data from one table into another table if these rows are not already in the table. In addition I want to add the value from a column in a third table based on the value of the column in the original table being copied across.
I have inherited two tables one of which is to be made redundant (because the database is going to be made defunct) and hence the information needs to be moved to another table which is not identical in design (however the datatypes are).
INSERT INTO [AMData].[dbo].[AttendanceRecord] A ([TypeID], [Date], [EmpID], [EhrID], [SickReason], [AbsID])
SELECT
T.[ID], F.[EHrDate], F.[EHrEmpID], F.[EHrID], F.[EHRUserComment], F.[AbsID]
FROM
[Focus].[dbo].[AllAbsence] F
INNER JOIN
[AMData].[dbo].[AttendanceTypes] T ON F.[AbsID] = T.[FocusID]
WHERE
NOT EXISTS (SELECT *
FROM [AMData].[dbo].[AttendanceRecord] A, [Focus].[dbo].[AllAbsence] F
WHERE A.[EhrID] = F.[EHrID]
AND F.[EHrDate] BETWEEN '2018/01/01' AND '2018/12/01'
AND F.[AbsID] <> 0
AND F.[AbsID] <> 2);
I am getting a syntax error when I try to run the code. I am not used to running insert code so there is probably obviously wrong.
A guess, but I suspect the problem is you are aliasing the table AttendanceRecord. You can't alias a table in the INSERT clause. The error is actually telling you this:
Ln: 1 Col: 47 - Incorrect syntax near 'A'
Remove the A and this will work (well, the below SQL doesn't have a syntax error, so I use the word "work" loosely):
INSERT INTO [AMData].[dbo].[AttendanceRecord] ([TypeID],
[Date],
[EmpID],
[EhrID],
[SickReason],
[AbsID])
SELECT T.[ID],
F.[EHrDate],
F.[EHrEmpID],
F.[EHrID],
F.[EHRUserComment],
F.[AbsID]
FROM [Focus].[dbo].[AllAbsence] AS F
INNER JOIN [AMData].[dbo].[AttendanceTypes] AS T ON F.[AbsID] = T.[FocusID]
WHERE NOT EXISTS (SELECT *
FROM [AMData].[dbo].[AttendanceRecord] AS A,
[Focus].[dbo].[AllAbsence] AS F
WHERE A.[EhrID] = F.[EHrID]
AND F.[EHrDate] BETWEEN '2018/01/01' AND '2018/12/01'
AND F.[AbsID] <> 0
AND F.[AbsID] <> 2);
I have two tables test1 and test2 both with single column containing some values.
I have applied the inner and outer joins but have confusion with the output.
Create table test1
( id int)
insert into test1 values (1)
insert into test1 values (1)
insert into test1 values (1)
Create table test2
( id int)
insert into test2 values (1)
insert into test2 values (1)
insert into test2 values (NULL)
select a.id from test1 a inner join test2 b on a.id = b.id
I was expecting,
1
1
Null
as output for inner join, left join and right join.
But the original output was,
1
1
1
1
1
1
Could you please help me in understanding this on all the joins.
Each one of the three 1s in test1 joined with each one of the two 1s in test2, and this yielded the 3x2=6 rows you got in the result-set. There is nothing different between the first,second and third 1 in test1 and nothing different between the first and second 1 in test2.
Also, keep in mind that all of the following conditions:
NULL = 1
NULL <> 1
NULL = NULL
NULL <> NULL
are false. All conditions which have a NULL at one side will evaluate to false. This is because a NULL represents an unknown value.
What you expected is quite wrong, as you can see. It seems you expected the first row of test1 to be joined with the first row of test2 and so on. There is no such "magic" in sql - the entire logic of the join is placed within the ON clause, which joined the 1s as expained before.
I'm having trouble figuring this out.
According to Jeff Atwood A Visual Explanation of SQL Joins Left outer join produces a complete set of records from Table A, with the matching records (where available) in Table B. If there is no match, the right side will contain null.
The left table (TableA) doesn't have duplicates. The right tableB has 1 or 2 entries for each client number. The PrimaryTP designates one as primary with 1 and the other has 0.
I shouldn't have to include the line And B.PrimaryTP = 1 because TableA doesn't have duplicates. Yet if I leave it out I get duplicate client numbers. Why?
Can you help me understand how this works. It's being very confusing to me. The logic of And B.PrimaryTP = 1 escapes me. Yet it seems to work. Still, I'm scared to trust it if I don't understand it. Can you help me understand it. Or do I have a logic error hidden in the query?
SELECT A.ClientNum --returns a list with no duplicate client numbers
FROM (...<TableA>
) as A
Left Outer Join
<TableB> as B
on A.ClientNum = B.ClientNum
--eliminate mismatch of (ClientNum <> FolderNum)
Where A.ClientNum Not In
(
Select ClientNum From <TableB>
Where ClientNum Is Not Null
And ClientNum <> IsNull(FolderNum, '')
)
--eliminate case where B.PrimaryTP <> 1
And B.PrimaryTP = 1
The difference between an INNER JOIN and a LEFT JOIN is just that the LEFT JOIN still returns the rows in Table A when there are no corresponding rows in Table B.
But it's still a JOIN, which means that if there is more than one corresponding row in Table B, it will join the row from Table A to each one of them.
So if you want to make sure that you get no more than one result for each row in Table A, you have to make sure that no more than one row from Table B is found - hence the And B.PrimaryTP = 1.
If you have one client number in A and two matches in Table B, then you will get duplicates.
Suppose you have the following data,
Table-A(client Num) Table-B(client Num)
1 2
2 2
The left Join Results
Table-A(client Num) Table-B(client Num)
1 (null)
2 2
2 2
This is the cause of duplicates. So you need to take distinct values form Table B or perform Distinct on the result set.
I shouldn't have to include the line And B.PrimaryTP = 1 because TableA doesn't have duplicates. Yet if I leave it out I get duplicate client numbers. Why?
Because both rows in the right table match a row in the left table. There is no way for SQL Server to output a triangular result; it must show the columns from both tables for every joined row. And this is true for INNER JOIN as well.
DECLARE #a TABLE(a INT);
DECLARE #b TABLE(b INT);
INSERT #a VALUES(1),(2);
INSERT #b VALUES(1),(1);
SELECT a.a, b.b FROM #a AS a
LEFT OUTER JOIN #b AS b ON a.a = b.b;
SELECT a.a, b.b FROM #a AS a
INNER JOIN #b AS b ON a.a = b.b;
Results:
a b
-- ----
1 1
1 1
2 NULL
a b
-- --
1 1
1 1
On the link that you gave the joins are explained very good. So the problem is that you have several records from table A (no matter that there are no duplicates) is that to 1 record in A there are 2 records in B (in some cases). To avoid this you can use either DISTINCT clause, either GROUP BY clause.
The LEFT OUTER JOIN will give you all the records from A with all the matching records from B. The difference with an INNER JOIN is that if there are no matching records in B, an INNER join will omit the record from A entirely, while the LEFT join will then still include a row with the results from A.
In your case, however, you may also want to check out the DISTINCT keyword.
I've come across a scenario where I need to return a complex set of calculated values at a crossover point from "legacy" to current.
To cut a long story short I have something like this ...
with someofit as
(
select id, col1, col2, col3 from table1
)
select someofit.*,
case when id < #lastLegacyId then
(select ... from table2 where something = id) as 'bla'
,(select ... from table2 where something = id) as 'foo'
,(select ... from table2 where something = id) as 'bar'
else
(select ... from table3 where something = id) as 'bla'
,(select ... from table3 where something = id) as 'foo'
,(select ... from table3 where something = id) as 'bar'
end
from someofit
No here lies the problem ...
I don't want to be constantly doing that case check for each sub selection but at the same time when that condition applies I need all of the selections within the relevant case block.
Is there a smarter way to do this?
if I was in a proper OO language I would use something like this ...
var common = GetCommonSuff()
foreach (object item in common)
{
if(item.id <= lastLegacyId)
{
AppendLegacyValuesTo(item);
}
else
{
AppendCurrentValuesTo(item);
}
}
I did initially try doing 2 complete selections with a union all but this doesn't work very well due to efficiency / number of rows to be evaluated.
The sub selections are looking for total row counts where some condition is met other than the id match on either table 2 or 3 but those tables may have millions of rows in them.
The cte is used for 2 reasons ...
firstly it pulls only the rows from table 1 i am interested in so straight away im only doing a fraction of the sub selections in each case.
secondly its returning the common stuff in a single lookup on table 1
Any ideas?
EDIT 1 :
Some context to the situation ...
I have a table called "imports" (table 1 above) this represents an import job where we take data from a file (csv or similar) and pull the records in to the db.
I then have a table called "steps" this represents the processing / cleaning rules we go through and each record contains a sproc name and a bunch of other stuff about the rule.
There is then a join table that represents the rule for a particular import "ImportSteps" (table 2 above - for current data), this contains a "rowsaffected" column and the import id
so for the current jobs my sql is quite simple ...
select 123 456
from imports
join importsteps
for the older legacy stuff however I have to look through table 3 ... table 3 is the holding table, it contains every record ever imported, each row has an import id and each row contains key values.
on the new data rowsaffected on table 2 for import id x where step id is y will return my value.
on the legacy data i have to count the rows in holding where col z = something
i need data on about 20 imports and this data is bound to a "datagrid" on my mvc web app (if that makes any difference)
the cte i use determines through some parameters the "current 20 im interested in" those params represent start and end record (ordered by import id).
My biggest issue is that holding table ... it's massive .. individual jobs have been known to contain 500k + records on their own and this table holds years of imported rows so i need my lookups on that table to be as fast as possible and as few as possible.
EDIT 2:
The actual solution (suedo code only) ...
-- declare and populate the subset to reduce reads on the big holding table
declare table #holding ( ... )
insert into #holding
select .. from holding
select
... common stuff from inner select in "from" below
... bunch of ...
case when id < #legacy then (select getNewValue(id, stepid))
else (select x from #holding where id = ID and ... ) end as 'bla'
from
(
select ROW_NUMBER() over (order by importid desc) as 'RowNum'
, ...
) as I
-- this bit handles the paging
where RowNum >= #StartIndex
and RowNum < #EndIndex
i'm still confident i can clean it up more but my original query that looked something like bills solution was about 45 seconds in execution time, this is about 7
I take it the subqueries must return a single scalar value, correct? This point is important because it is what ensures the LEFT JOINs will not multiply the result.
;with someofit as
(
select id, col1, col2, col3 from table1
)
select someofit.*,
bla = coalesce(t2.col1, t3.col1),
foo = coalesce(t2.col2, t3.col2),
bar = coalesce(t2.bar, t3.bar)
from someofit
left join table2 t2 on t2.something=someofit.id and somefit.id < #lastLegacyId
left join table3 t3 on t3.something=someofit.id and somefit.id >= #lastLegacyId
Beware that I have used id >= #lastLegacyId as the complement of the condition, by assuming that id is not nullable. If it is, you need an IsNull there, i.e. somefit.id >= isnull(#lastLegacyId,somefit.id).
Your edit to the question doesn't change the fact that this is an almost literal translation of the O-O syntax.
foreach (object item in common) --> "from someofit"
{
if(item.id <= lastLegacyId) --> the precondition to the t2 join
{
AppendLegacyValuesTo(item); --> putting t2.x as first argument of coalesce
}
else --> sql would normally join to both tables
--> hence we need an explicit complement
--> condition as an "else" clause
{
AppendCurrentValuesTo(item); --> putting t3.x as 2nd argument
--> tbh, the order doesn't matter since t2/t3
--> are mutually exclusive
}
}
function AppendCurrentValuesTo --> the correlation between t2/t3 to someofit.id
Now, if you have actually tried this and it doesn't solve your problem, I'd like to know where it broke.
Assuming you know that there are no conflicting ID's between the two tables, you can do something like this (DB2 syntax, because that's what I know, but it should be similar):
with combined_tables as (
select ... as id, ... as bla, ...as bar, ... as foo from table 2
union all
select ... as id, ... as bla, ...as bar, ... as foo from table 3
)
select someofit.*, combined_ids.bla, combined_ids.foo, combined_ids.bar
from someofit
join combined_tables on someofit.id = combined_tables.id
If you had cases like overlapping ids, you could handle that within the combined_tables() section
I have a problem which my limited SQL knowledge is keeping me from understanding.
First the problem:
I have a database which I need to run a report on, it contains configurations of a users entitlements. The report needs to show a distinct list of these configurations and a count against each one.
So a line in my DB looks like this:
USER_ID SALE_ITEM_ID SALE_ITEM_NAME PRODUCT_NAME CURRENT_LINK_NUM PRICE_SHEET_ID
37715 547 CultFREE CultPlus 0 561
the above line is one row of a users configuration, for every user ID there can be 1-5 of these lines. So the definition of a configuration is multiple rows of data sharing a common User ID with variable attributes..
I need to get a distinct list of these configurations across the whole table, leaving me just one configuration set for every instance where > 1 has that configuration and a count of instances of that configuration.
Hope this is clear?
Any ideas?!?!
I have tried various group by's and unions, also the grouping sets function to no avail.
Will be very greatful if anyone can give me some pointers!
Ouch that hurt ...
Ok so problem:
a row represents a configurable line
users may be linked to more than 1 row of configuration
configuration rows when grouped together form a configuration set
we want to figure out all of the distinct configuration sets
we want to know what users are using them.
Solution (its a bit messy but the idea is there, copy and paste in to SQL management studio) ...
-- ok so i imported the data to a table named SampleData ...
-- 1. import the data
-- 2. add a new column
-- 3. select all the values of the config in to the new column (Configuration_id)
--UPDATE [dbo].[SampleData]
--SET [Configuration_ID] = SALE_ITEM_ID + SALE_ITEM_NAME + [PRODUCT_NAME] + [CURRENT_LINK_NUM] + [PRICE_SHEET_ID] + [Configuration_ID]
-- 4. i then selected just the distinct values of those and found 6 distinct Configuration_id's
--SELECT DISTINCT [Configuration_ID] FROM [dbo].[SampleData]
-- 5. to make them a bit easier to read and work with i gave them int values instead
-- for me it was easy to do this manually but you might wanna do some trickery here to autonumber them or something
-- basic idea is to run the step 4 statement but select into a new table then add a new primary key column and set identity spec on it
-- that will generate u a bunch of incremental numbers for your config id's so u can then do something like ...
--UPDATE [dbo].[SampleData] sd
--SET Configuration_ID = (SELECT ID FROM TempConfigTable WHERE Config_ID = sd.Configuration_ID)
-- at this point you have all your existing rows with a unique ident for the values combined in each row.
-- so for example in my dataset i have several rows where only the user_id has changed but all look like this ...
--SALE_ITEM_ID SALE_ITEM_NAME PRODUCT_NAME CURRENT_LINK_NUM PRICE_SHEET_ID Configuration_ID
--54101 TravelFREE TravelPlus 0 56101 1
-- now you have a config id you can start to work on building sets up ...
-- each user is now matched with 1 or more config id
-- 6. we use a CTE (common table expression) to link the possibles (keeps the join small) ...
--WITH Temp (ConfigID)
--AS
--(
-- SELECT DISTINCT SD.Configuration_Id --SD2.Configuration_Id, SD3.Configuration_Id, SD4.Configuration_Id, SD5.Configuration_Id,
-- FROM [dbo].[SampleData] SD
--)
-- this extracts all the possible combinations using the CTE
-- on the basis of what you told me, max rows per user is 6, in the result set i have i only have 5 distinct configs
-- meaning i gain nothing by doing a 6th join.
-- cross joins basically give you every combination of unique values from the 2 tables but we joined back on the same table
-- so its every possible combination of Temp + Temp (ConfigID + ConfigID) ... per cross join so with 5 joins its every combination of
-- Temp + Temp + Temp + Temp + Temp .. good job temp only has 1 column with 5 values in it
-- 7. uncomment both this and the CTE above ... need to use them together
--SELECT DISTINCT T.ConfigID C1, T2.ConfigID C2, T3.ConfigID C3, T4.ConfigID C4, T5.ConfigID C5
--INTO [SETS]
--FROM Temp T
--CROSS JOIN Temp T2
--CROSS JOIN Temp T3
--CROSS JOIN Temp T4
--CROSS JOIN Temp T5
-- notice the INTO clause ... this dumps me out a new [SETS] table in my db
-- if i go add a primary key to this and set its ident spec i now have unique set id's
-- for each row in the table.
--SELECT *
--FROM [dbo].[SETS]
-- now here's where it gets interesting ... row 1 defines a set as being config id 1 and nothing else
-- row 2 defines set 2 as being config 1 and config 2 and nothing else ... and so on ...
-- the problem here of course is that 1,2,1,1,1 is technically the same set as 1,1,1,2,1 from our point of view
-- ok lets assign a set to each userid ...
-- 8. first we pull the distinct id's out ...
--SELECT DISTINCT USER_ID usr, null SetID
--INTO UserSets
--FROM SampleData
-- now we need to do bit a of operating on these that's a bit much for a single update or select so ...
-- 9. process findings in a loop
DECLARE #currentUser int
DECLARE #set int
-- while theres a userid not linked to a set
WHILE EXISTS(#currentUser = SELECT TOP 1 usr FROM UserSets WHERE SetId IS NULL)
BEGIN
-- figure out a set to link it to
SET #set = (
SELECT TOP 1 ID
FROM [SETS]
-- shouldn't really do this ... basically need to refactor in to a table variable then compare to that
-- that way the table lookup on ur main data is only 1 per User_id
WHERE C1 IN (SELECT DISTINCT Configuration_id FROM SampleData WHERE USER_ID = #currentUser)
AND C2 IN (SELECT DISTINCT Configuration_id FROM SampleData WHERE USER_ID = #currentUser)
AND C3 IN (SELECT DISTINCT Configuration_id FROM SampleData WHERE USER_ID = #currentUser)
AND C4 IN (SELECT DISTINCT Configuration_id FROM SampleData WHERE USER_ID = #currentUser)
AND C5 IN (SELECT DISTINCT Configuration_id FROM SampleData WHERE USER_ID = #currentUser)
)
-- hopefully that worked
IF(#set IS NOT NULL)
BEGIN
-- tell the usersets table
UPDATE UserSets SET SetId = #set WHERE usr = #currentUser
set #set = null
END
ELSE -- something went wrong ... set to 0 to prevent endless loop but any userid linked to set 0 is a problem u need to look at
UPDATE UserSets SET SetId = 0 WHERE usr = #currentUser
-- and round we go again ... until we are done
END
SELECT
USER_ID,
SALE_ITEM_ID, ETC...,
COUNT(*) WhateverYouWantToNameCount
FROM TableNAme
GROUP BY USER_ID