DISTINCTROW equivalent in SQL Server - sql-server

Could you please tell me if there is any equivalent of Access' DISTINCTROW for SQL Server?

From reading the documentation on distinctrow, it looks like you want an exists clause or (for generally better performance) a join:
SELECT * FROM A
WHERE EXISTS(SELECT * FROM B WHERE B.a_id = A.a_id)
SELECT DISTINCT A.* FROM A
INNER JOIN B ON A.a_id = B.a_id
-- corrected, should be inner join here

SQL server doesn't support DISTINCTROW instead you should use the keyword DISTINCT
But the answer that Chris Shain has given gives you what you want.

You can just use the keyword DISTINCT instead.

Related

The multi-part identifier "Registry.categoryID" could not be bound [duplicate]

I've seen similar errors on SO, but I don't find a solution for my problem.
I have a SQL query like:
SELECT DISTINCT
a.maxa ,
b.mahuyen ,
a.tenxa ,
b.tenhuyen ,
ISNULL(dkcd.tong, 0) AS tongdkcd
FROM phuongxa a ,
quanhuyen b
LEFT OUTER JOIN ( SELECT maxa ,
COUNT(*) AS tong
FROM khaosat
WHERE CONVERT(DATETIME, ngaylap, 103) BETWEEN 'Sep 1 2011'
AND
'Sep 5 2011'
GROUP BY maxa
) AS dkcd ON dkcd.maxa = a.maxa
WHERE a.maxa <> '99'
AND LEFT(a.maxa, 2) = b.mahuyen
ORDER BY maxa;
When I execute this query, the error result is:
The multi-part identifier "a.maxa" could not be bound. Why?
P/s: if i divide the query into 2 individual query, it run ok.
SELECT DISTINCT
a.maxa ,
b.mahuyen ,
a.tenxa ,
b.tenhuyen
FROM phuongxa a ,
quanhuyen b
WHERE a.maxa <> '99'
AND LEFT(a.maxa, 2) = b.mahuyen
ORDER BY maxa;
and
SELECT maxa ,
COUNT(*) AS tong
FROM khaosat
WHERE CONVERT(DATETIME, ngaylap, 103) BETWEEN 'Sep 1 2011'
AND 'Sep 5 2011'
GROUP BY maxa;
You are mixing implicit joins with explicit joins. That is allowed, but you need to be aware of how to do that properly.
The thing is, explicit joins (the ones that are implemented using the JOIN keyword) take precedence over implicit ones (the 'comma' joins, where the join condition is specified in the WHERE clause).
Here's an outline of your query:
SELECT
…
FROM a, b LEFT JOIN dkcd ON …
WHERE …
You are probably expecting it to behave like this:
SELECT
…
FROM (a, b) LEFT JOIN dkcd ON …
WHERE …
that is, the combination of tables a and b is joined with the table dkcd. In fact, what's happening is
SELECT
…
FROM a, (b LEFT JOIN dkcd ON …)
WHERE …
that is, as you may already have understood, dkcd is joined specifically against b and only b, then the result of the join is combined with a and filtered further with the WHERE clause. In this case, any reference to a in the ON clause is invalid, a is unknown at that point. That is why you are getting the error message.
If I were you, I would probably try to rewrite this query, and one possible solution might be:
SELECT DISTINCT
a.maxa,
b.mahuyen,
a.tenxa,
b.tenhuyen,
ISNULL(dkcd.tong, 0) AS tongdkcd
FROM phuongxa a
INNER JOIN quanhuyen b ON LEFT(a.maxa, 2) = b.mahuyen
LEFT OUTER JOIN (
SELECT
maxa,
COUNT(*) AS tong
FROM khaosat
WHERE CONVERT(datetime, ngaylap, 103) BETWEEN 'Sep 1 2011' AND 'Sep 5 2011'
GROUP BY maxa
) AS dkcd ON dkcd.maxa = a.maxa
WHERE a.maxa <> '99'
ORDER BY a.maxa
Here the tables a and b are joined first, then the result is joined to dkcd. Basically, this is the same query as yours, only using a different syntax for one of the joins, which makes a great difference: the reference a.maxa in the dkcd's join condition is now absolutely valid.
As #Aaron Bertrand has correctly noted, you should probably qualify maxa with a specific alias, probably a, in the ORDER BY clause.
Sometimes this error occurs when you use your schema (dbo) in your query in a wrong way.
for example if you write:
select dbo.prd.name
from dbo.product prd
you will get the error.
In this situations change it to:
select prd.name
from dbo.product prd
if you have given alies name change that to actual name
for example
SELECT
A.name,A.date
FROM [LoginInfo].[dbo].[TableA] as A
join
[LoginInfo].[dbo].[TableA] as B
on [LoginInfo].[dbo].[TableA].name=[LoginInfo].[dbo].[TableB].name;
change that to
SELECT
A.name,A.date
FROM [LoginInfo].[dbo].[TableA] as A
join
[LoginInfo].[dbo].[TableA] as B
on A.name=B.name;
I was struggling with the same error message in SQL SERVER, since I had multiple joins, changing the order of the joins solved it for me.
In my case the issue turned out to be the alias name I had given to the table. "oa" seems to be not acceptable for SQL Server.
What worked for me was to change my WHERE clause into a SELECT subquery
FROM:
DELETE FROM CommentTag WHERE [dbo].CommentTag.NoteId = [dbo].FetchedTagTransferData.IssueId
TO:
DELETE FROM CommentTag WHERE [dbo].CommentTag.NoteId = (SELECT NoteId FROM FetchedTagTransferData)
I was having the same error from JDBC. Checked everything and my query was fine. Turned out, in where clause I have an argument:
where s.some_column = ?
And the value of the argument I was passing in was null. This also gives the same error which is misleading because when you search the internet you end up that something is wrong with the query structure but it's not in my case. Just thought someone may face the same issue
I'm new to SQL, but came across this issue in a course I was taking and found that assigning the query to the project specifically helped to eliminate the multi-part error. For example the project I created was CTU SQL Project so I made sure I started my script with USE [CTU SQL Project] as my first line like below.
USE [CTU SQL Project]
SELECT Advisors.First_Name, Advisors.Last_Name...and so on.
If this error happens in an UPDATE, double-check the JOIN on the table with the column/field that is causing the error.
In my case this was due to the lack of the JOIN itself, which generated the same error due to an unknown field (as Andriy pointed out).
Instead you can try joining tables like,
select
....
from
dkcd
right join
a
, b
This should work
SELECT DISTINCT
phuongxa.maxa ,
quanhuyen.mahuyen ,
phuongxa.tenxa ,
quanhuyen.tenhuyen ,
ISNULL(dkcd.tong, 0) AS tongdkcd
FROM phuongxa ,
quanhuyen
LEFT OUTER JOIN ( SELECT khaosat.maxa ,
COUNT(*) AS tong
FROM khaosat
WHERE CONVERT(DATETIME, ngaylap, 103) BETWEEN 'Sep 1 2011'
AND
'Sep 5 2011'
GROUP BY khaosat.maxa
) AS dkcd ON dkcd.maxa = maxa
WHERE phuongxa.maxa <> '99'
AND LEFT(phuongxa.maxa, 2) = quanhuyen.mahuyen
ORDER BY maxa;
My error was to use a field that did not exist in table.
table1.field1 => is not exist
table2.field1 => is correct
Correct your Table Name.
my error occurred because of using WITH
WITH RCTE AS (
SELECT...
)
SELECT RCTE.Name, ...
FROM
RCTE INNER JOIN Customer
ON RCTE.CustomerID = Customer.ID
when used in join with other tables ...
Did you forget to join some tables? If not then you probably need to use some aliases.
I was also struggling with this error and ended up with the same strategy as the answer. I am including my answer just to confirm that this is a strategy that should work.
Here is an example where I do first one inner join between two tables I know got data and then two left outer joins on tables that might have corresponding rows that can be empty. You mix inner joins and outer joins to get results with data accross tables instead of doing the default comma separated syntax between tables and miss out rows in your desired join.
use somedatabase
go
select o.operationid, o.operatingdate, p.pasid, p.name as patientname, o.operationalunitid, f.name as operasjonsprogram, o.theaterid as stueid, t.name as stuenavn, o.status as operasjonsstatus from operation o
inner join patient p on o.operationid = p.operationid
left outer join freshorganizationalunit f on f.freshorganizationalunitid = o.operationalunitid
left outer join theater t on t.theaterid = o.theaterid
where (p.Name like '%Male[0-9]%' or p.Name like '%KFemale [0-9]%')
First: Do the inner joins between tables you expect to have data matching.
Second part: Continue with outer joins to try to retrieve data in other tables,
but this will not filter out your result set if table outer joining to has not got corresponding data or match on the condition you set up in the on predicate / condition.
This error can also be caused by simply missing a comma , between the column names in the SELECT statement.
eg:
SELECT MyCol1, MyCol2 MyCol3 FROM SomeTable;
For me the issue was that I was stupidly calling a DB function without empty brackets select [apo].[GenerateNationalIdFrance] instead of select [apo].[GenerateNationalIdFrance]() ... took me few minutes to realize that but worth mentioning for juniors out there :-)
For me I was using wrong alias spellings , it worked after correct spelings

how to improve performance when rewrite join SQL?

Suppose I have 2 table need to join. There are 2 way to write the sql:
select * from taba a join tabb b on a.id =b.id where ...
select * from taba a, tabb b where a.id = b.id and ...
which one has better performance or this is only syntax issue with different SQL standard regardless of performance?
Has been already answered here
stackoverflow.com/questions/1129923/is-a-join-faster-than-a-where
The query optimizer usually use more a join than a where clause (so in theory is better the join) but the last word is said by the db engine you're using
The best advice is to try

Use of the IN condition

I can easily create a stored procedure in SQL Server with parameters that I use with =, LIKE and most operators. But when it comes to using IN, I don't really understand what to do, and I can't find a good site to teach me.
Example
CREATE PROCEDURE TEST
#Ids --- What type should go here?
AS BEGIN
SELECT * FROM TableA WHERE ID IN ( #Ids )
END
Is this possible and if so how ?
With SQL Server 2008 and above, you can use Table Valued Parameters.
You declare a table type and can use that as a parameter (read only) for stored procedures that can be used in IN clauses.
For the different options, I suggest reading the relevant article for your version of the excellent Arrays and Lists in SQL Server, by Erland Sommarskog.
I've done this in the past using a Split function that I add to my schema functions as described here
Then you can do the following:
CREATE PROCEDURE TEST
#Ids --- What type should go here?
AS BEGIN
SELECT * FROM TableA WHERE ID IN ( dbo.Split(#Ids, ',') )
END
Just remember that the IN function always expects a table of values as a result. SQL Server is smart enough to convert strings to this table format, so long as they are specifically written in the procedure.
Another option in your specific example though, could be to use a join. This will have a performance improvement, but often does not really meet a real-world example you need. The join version would be:
SELECT *
FROM TableA AS ta
INNER JOIN dbo.Split(#Ids, ',') AS ids
ON ta.Id = ids.items
If your asking what I think your asking, I do this every day..
WITH myData(FileNames)
AS
(
SELECT '0608751970%'
UNION ALL SELECT '1000098846%'
UNION ALL SELECT '1000101277%'
UNION ALL SELECT '1000108488%'
)
SELECT DISTINCT f.*
FROM tblFiles (nolock) f
INNER JOIN myData md
ON b.FileNames LIKE md.FileNames
Or if your doing this based on another table:
WITH myData(FileNames)
AS
(
SELECT RTRIM(FileNames) + '%'
FROM tblOldFiles
WHERE Active=1
)
SELECT DISTINCT f.*
FROM tblFiles (nolock) f
INNER JOIN myData md
ON b.FileNames LIKE md.FileNames

LIKE comparison with IN operator

I want to perform a LIKE style comparison with an IN operator e.g.:
select *
from tbl1
where tbl1.value_to_check in (select allowed_str from allowedvalues)
Where allowed values could contain the following allowed_str:
ab% (allow any string starting ab)
%ab (allow any string ending ab)
The above query obviously does not work in this way however is there a way of achieving this using SQL Server 2008?
No, IN does not support this. Try:
SELECT t1.*
FROM dbo.tbl1 AS t1
INNER JOIN dbo.allowedvalues AS a
ON t1.value_to_check LIKE '%' + a.allowed_str + '%';
As Damien points out, if you are storing ab% and %ab separately, rather than just ab on its own, then you just need:
SELECT t1.*
FROM dbo.tbl1 AS t1
INNER JOIN dbo.allowedvalues AS a
ON t1.value_to_check LIKE a.allowed_str;
I believe the exists clause matches better with the original query.
select *
from tbl1
where exists (select 1
from allowedvalues
where tbl1.value_to_check like '%' + a.allowed_str + '%';)
Note: this will cause a table scan on tbl1 (and might be very slow)

Using the results of a query in OPENQUERY

I have a SQL Server 2005 database that is linked to an Oracle database. What I want to do is run a query to pull some ID numbers out of it, then find out which ones are in Oracle.
So I want to take the results of this query:
SELECT pidm
FROM sql_server_table
And do something like this to query the Oracle database (assuming that the results of the previous query are stored in #pidms):
OPENQUERY(oracledb,
'
SELECT pidm
FROM table
WHERE pidm IN (' +
#pidms + ')')
GO
But I'm having trouble thinking of a good way to do this. I suppose that I could do an inner join of queries similar to these two. Unfortunately, there are a lot of records to pull within a limited timeframe so I don't think that will be a very performant option to choose.
Any suggestions? I'd ideally like to do this with as little Dynamic SQL as possible.
Ahhhh, pidms. Brings back bad memories! :)
You could do the join, but you would do it like this:
select sql.pidm,sql.field2 from sqltable as sql
inner join
(select pidm,field2 from oracledb..schema.table) as orcl
on
sql.pidm = orcl.pidm
I'm not sure if you could write a PL/SQL procedure that would take a table variable from sql...but maybe.....no, I doubt it.
Store openquery results in a temp table, then do an inner join between the SQL table and the temp table.
I don't think you can do a join since OPENQUERY requires a pure string (as you wrote above).
BG: Actually JOIN IN SQLServer to Oracle by OpenQuery works, avoiding #tmp table and allowing JOIN to SQL without Param* - ex.
[SQL SP] LEFT JOIN OPENQUERY(ORADB,
'SELECT COUNT(distinct O.ORD_NUM) LCNT,
O.ORD_MAIN_NUM
FROM CUSTOMER.CUST_FILE C
JOIN CUSTOMER.ORDER_NEW O
ON C.ID = O.ORD_ID
WHERE C.CUS_ID NOT IN (''2'',''3'')
GROUP BY O.ORD_MAIN_MACNUM') LC
ON T.ID = LC.ORD_MAIN_ID*
Cheers, Bill Gibbs

Resources