I have 3 tables. They are inner joined. When I choose to select from all three tables I get a cartesian product.
I've used distinct and I've tried cross apply with top 1. Top 1 brings back the right amount of records but it repeats the fields used in that select top N.
Basic question. Can you select from 3 different tables and avoid a cartesian result? I can have all three tables joined and with distinct I can get records from two of the tables without a cartesian. It is when I choose to select from a third is where the cartesian appears.
If this is possible, what other tsql commands/constructs should I be experimenting with?
http://imageshack.us/f/255/50353790.png/
SELECT CRT.[TransactionID]
,CRT.[creditrewardsID]
,CRT.[OwnerID]
, CRT.[TransactionDate]
,CRT.[ItemID]
,CRT.[VALUE]
,CRM.First
,CRM.MI
,CRM.Last
,CTI.fn
,CTI.ln
FROM [ownership].[dbo].[creditrewardsTransactions] CRT
Join [ownership].[dbo].[creditrewardsMembers] CRM
on CRT.creditrewardsid = CRM.[creditrewardsID]
Join [Exchange].[dbo].[CreditTourInfo] CTI
on CRM.CRMemberNum = CTI.PRIMECRPNum
--where CRT.creditrewardsID = 11111
To answer your "basic" question, yes you can easily join 3 tables and not receive a Cartesian product. However, you need to ensure you have proper One-to-One or One-to-Many joins.
The issue above likely has to do with the data you have in the creditrewardsMembers and CreditTourInfo tables. Joining the CRMemberNum and PRIMECRPNum fields is likely a Many-to-Many join, which will give you a Cartesian product.
You can run these SQL statements to confirm. If they both return records, then you have a Many-to-Many join issue.
SELECT CRM.CRMemberNum, COUNT(*)
FROM [ownership].[dbo].[creditrewardsMembers] CRM
GROUP BY CRM.CRMemberNum
HAVING COUNT(*) > 1
SELECT CTI.PRIMECRPNum, COUNT(*)
FROM [Exchange].[dbo].[CreditTourInfo] CTI
GROUP BY CTI.PRIMECRPNum
HAVING COUNT(*) > 1
If the fields are in fact the correct ones to join, then you may need to just pick the first record from one of the tables in your query like this.
SELECT CRT.[TransactionID]
,CRT.[creditrewardsID]
,CRT.[OwnerID]
,CRT.[TransactionDate]
,CRT.[ItemID]
,CRT.[VALUE]
,CRM.First
,CRM.MI
,CRM.Last
,CTI.fn
,CTI.ln
FROM [ownership].[dbo].[creditrewardsTransactions] CRT
JOIN [ownership].[dbo].[creditrewardsMembers] CRM
ON CRT.creditrewardsid = CRM.[creditrewardsID]
JOIN (
SELECT PRIMECRPNum, MIN(fn) as [fn], MIN(ln) as [ln]
FROM [Exchange].[dbo].[CreditTourInfo]
GROUP BY PRIMECRPNum
) CTI
ON CRM.CRMemberNum = CTI.PRIMECRPNum
Ultimately, it's going to depend on the limitations of your specific data model and the tradeoffs you may need to make. (ie. getting the first record only from [CreditTourInfo])
Related
I have a distinct list of part numbers from one table. It is basically a table that contains a record of all the company's part numbers. I want to add columns that will pull data from different tables but only pertaining to the part number on that row of the distinct part list.
For example: if I have part A, B, C from the unique part list I want to add columns for Purchase quantity, repair quantity, loan quantity, etc... from three totally unique tables.
So it's almost like I need 3 subqueries that will sum of that data from the different tables for each part.
Can anybody steer me in the direction of how to do this? Please and thank you so much!
One method is correlated subqueries. Something like this:
select p.*,
(select count(*)
from purchases pu
where pu.part_id = p.part_id
) as num_purchases,
(select count(*)
from repairs r
where r.part_id = p.part_id
) as num_repairs,
(select count(*)
from loans l
where l.part_id = p.part_id
) as num_loans
from parts p;
Another option is joins with aggregation before the join. Or lateral joins (which are quite similar to correlated subqueries).
I am trying to run a query to get all the customers from my database. These are my tables in a diagram :
when running the query by joining the table Companies_Customers and the Customers table based on the customerId in both tables(doesn't show in the join table in the pic), I get duplicate rows, which is not the desired outcome.
This is normal from a database standpoint since a Customer can be related to different companies (Companies can share single customer).
My question is how do I get rid of the duplication via SQL.
There can be 2 approaches to your problem.
Either only select data from Customers table:
SELECT * FROM Customers
Or select from both tables joined together, but without CompanyName and with GROUP BY CompanyCustomerId - although I highly suggest the first approach.
I have three tables: Clinic, Stock and StockLog.
I need to get all rows where Stock.stock < 5. I need to also show if an order has been placed and what amount it is; which is found in the table Stocklog.
The issue is that a user can set his stock level in the Stock table without placing an order which would go to Stocklog.
I need a query that can : return the rows in the Stock table and get the related order amounts in the Stocklog table. If no order has been placed in StockLog, then set amount to order amount to zero.
I have tried :
SELECT
Clinic.Name,
Stock.NameOfMedication, Stock.Stock,
StockLog.OrderAmount
FROM
Clinic
JOIN
Stock ON Stock.ClinicID = Clinic.ClinicID
JOIN
StockLog ON StockLog.StockID = Stock.StockID
WHERE
Stock.Stock <= 5
The issue with my query is that I lose rows which are not found in StockLog.
Any help on how to write this.
Thank you.
I am thinking the query should look like this:
SELECT c.Name, s.NameOfMedication, s.Stock,
COALESCE(sl.OrderAmount, 0) as OrderAmount
FROM Stock s LEFT JOIN
Clinic c
ON s.ClinicID = c.ClinicID LEFT JOIN
StockLog sl
ON sl.StockID = s.StockID
WHERE s.Stock <= 5 ;
You want to keep all rows in Stock (subject to the WHERE condition). So think: "make Stock the first table in the FROM and use LEFT JOIN for all the other tables."
If you want to keep all the rows that result from joining Clinic and Stock, then use a LEFT OUTER JOIN with StockLog. I don't know which SQL you're using (SQL Server, MySQL, PostgreSQL, Oracle), so I can't give you a precise example, but searching for "left outer join" in the relevant documentation should work.
See this Stack Overflow post for an explanation of the various kinds of joins.
I have two tables, Invoices and members, connected by PK/FK relationship through the field InvoiceNum. I have created the following sql and it works fine, and pulls 44 records as expected.
SELECT
INVOICES.InvoiceNum,
INVOICES.GroupNum,
INVOICES.DivisionNum,
INVOICES.DateBillFrom,
INVOICES.DateBillTo
FROM INVOICES
INNER JOIN MEMBERS ON INVOICES.InvoiceNum = MEMBERS.InvoiceNum
WHERE MEMBERS.MemberNum = '20032526000'
Now, I want to replace INVOICES.GroupNum and INVOICES.DivisionNum in the above query with GroupName and DivisionName. These values are present in the Groups and Divisions tables which also have the corresponding Group_num and Division_num fields. I have created the following sql. The problem is that it now pulls 528 records instead of 44!
SELECT
INVOICES.InvoiceNum,
INVOICES.DateBillFrom,
INVOICES.DateBillTo,
DIVISIONS.DIVISION_NAME,
GROUPS.GROUP_NAME
FROM INVOICES
INNER JOIN MEMBERS ON INVOICES.InvoiceNum = MEMBERS.InvoiceNum
INNER JOIN GROUPS ON INVOICES.GroupNum = GROUPS.Group_Num
INNER JOIN DIVISIONS ON INVOICES.DivisionNum = DIVISIONS.Division_Num
WHERE MEMBERS.MemberNum = '20032526000'
Any help is greatly appreciated.
You have at least one relation between your tables which is missing in your query. It gives you extra records. Find all common fields. Say, are divisions related to groups?
The statement is fine, as far as the SQL syntax goes.
But the question you have to ask yourself (and answer it):
How many rows in Groups do you get for that given GroupNum?
Ditto for Divisions - how many rows exist for that DivisionNum?
It would appear that those numbers aren't unique - multiple rows exist for each number - therefore you get multiple rows returned
I have a database with two tables
One with games
and one with participants
A game is able to have more participants and these are in a different table.
Is there a way to combine these two into one query?
Thanks
You can combine them using the JOIN operator.
Something like
SELECT *
FROM games g
INNER JOIN participants p ON p.gameid = g.gameid
Explanation on JOIN operators
INNER JOIN - Match rows between the two tables specified in the INNER
JOIN statement based on one or more
columns having matching data.
Preferably the join is based on
referential integrity enforcing the
relationship between the tables to
ensure data integrity.
o Just to add a little commentary to the basic definitions
above, in general the INNER JOIN
option is considered to be the most
common join needed in applications
and/or queries. Although that is the
case in some environments, it is
really dependent on the database
design, referential integrity and data
needed for the application. As such,
please take the time to understand the
data being requested then select the
proper join option.
o Although most join logic is based on matching values between
the two columns specified, it is
possible to also include logic using
greater than, less than, not equals,
etc.
LEFT OUTER JOIN - Based on the two tables specified in the join
clause, all data is returned from the
left table. On the right table, the
matching data is returned in addition
to NULL values where a record exists
in the left table, but not in the
right table.
o Another item to keep in mind is that the LEFT and RIGHT OUTER
JOIN logic is opposite of one another.
So you can change either the order of
the tables in the specific join
statement or change the JOIN from left
to right or vice versa and get the
same results.
RIGHT OUTER JOIN - Based on the two tables specified in the join
clause, all data is returned from the
right table. On the left table, the
matching data is returned in addition
to NULL values where a record exists
in the right table but not in the left
table.
Self -Join - In this circumstance, the same table is
specified twice with two different
aliases in order to match the data
within the same table.
CROSS JOIN - Based on the two tables specified in the join clause, a
Cartesian product is created if a
WHERE clause does filter the rows.
The size of the Cartesian product is
based on multiplying the number of
rows from the left table by the number
of rows in the right table. Please
heed caution when using a CROSS JOIN.
FULL JOIN - Based on the two tables specified in the join clause,
all data is returned from both tables
regardless of matching data.
example
table Game has columns (gameName, gameID)
table Participant has columns (participantID, participantName, gameID)
the GameID column is the "link" between the 2 tables. you need a common column you can join between 2 tables.
SELECT gameName, participantName
FROM Game g
JOIN Participat p ON g.gameID = p.gameID
This will return a data set of all games and the participants for those games.
The list of games will be redundant unless you structure it some other way due to multiple participants to that game.
sample data
WOW Bob
WOW Jake
StarCraft2 Neal
Warcraft3 James
Warcraft3 Rich
Diablo Chris