Go Premium for a chance to win a PS4. Enter to Win

x
  • Status: Solved
  • Priority: Medium
  • Security: Public
  • Views: 129
  • Last Modified:

How can I combine SQL select by columns?

I would like to combine results from a number of tables/queries to produce the table shown in ResultsD.
Please note I am limited to SQL server 2000.
The tables are not related and the fields in the rows in the output ResultsD are therefore not related to each other.

ResultsA
1
2
3
4

ResultsB
Apple
Pear
Lemon

ResultsC
1.0
3.0
5.0
6.0
7.0

ResultsD
1                Apple        1.0
2                Pear          3.0            
3                Lemon      5.0
4                Null           6.0
Null           Null           7.0
0
dgloveruk
Asked:
dgloveruk
  • 5
  • 4
  • 3
  • +1
2 Solutions
 
Vitor MontalvãoMSSQL Senior EngineerCommented:
In SQL Server 2000 you can't. Being a relational database you need to find something that can relate those 3 tables.
In newer versions of SQL Server it's possible to attribute row numbers to the records and use it for relating data. In this case you can't do what you want.
0
 
Jim HornMicrosoft SQL Server Developer, Architect, and AuthorCommented:
Can you explain why you want this?  Perhaps there is an easier workaround.

>Please note I am limited to SQL server 2000.
Okay, but be advised that you're running six versions old (2005 > 2008 > R2 > 2012 > 2014), which means Microsoft support is over, experienced experts are unlikely to maintain skillsets exclusively to 2000, and in the event of any server/database crash the odds of not being able to restore back to 2000 diminish greatly.  I'd tell your employer to spend some money to protect their data and upgrade.
0
 
Scott PletcherSenior DBACommented:
You must use temp tables in SQL 2000; might be best anyway, given the joins you need to do.

--Setup: we must add an identity to each result to join on---------------------------------
SELECT IDENTITY(int, 1, 1) AS ident, *
INTO #ResultsA
FROM ResultsA a
ORDER BY a.col1

SELECT IDENTITY(int, 1, 1) AS ident, *
INTO #ResultsB
FROM ResultsB b
ORDER BY b.col1

SELECT IDENTITY(int, 1, 1) AS ident, *
INTO #ResultsC
FROM ResultsC c
ORDER BY c.col1

--Ok, now we're ready for the final join itself-------------------------------------------------------
SELECT a.col1, b.col1, c.col1
FROM #ResultsA a
FULL OUTER JOIN #ResultsB b ON
    b.ident = a.ident
FULL OUTER JOIN #ResultsC c ON
    c.ident IN (a.ident, b.ident)
0
Prepare for your VMware VCP6-DCV exam.

Josh Coen and Jason Langer have prepared the latest edition of VCP study guide. Both authors have been working in the IT field for more than a decade, and both hold VMware certifications. This 163-page guide covers all 10 of the exam blueprint sections.

 
dgloverukAuthor Commented:
Thank you Scott, to answer Jim, my reason for asking is as follows.  I have a view of results.... For arguments sake lets say customer orders.  There are several things from this list of results that i need distinct lists of e.g salespersons, item categories and payment types to build filters in the user interface with options only relating to the orders being retrieved.  I cannot requery the view for distinct values of each column of interest since the underlying view is rapidly evolving and the view is fairly process intensive anyway to be doing a series of distinct selects on each column.  What i was going to try was querying the view for each of the columns of interest and their distinct values and return them as one table for building filter options from in the user interface.  I thought the approach above would allow me to return a complete set of filters options for several fields at once without having to action a series of queries from the client to generate each filter option.
0
 
Vitor MontalvãoMSSQL Senior EngineerCommented:
What i was going to try was querying the view for each of the columns of interest and their distinct values and return them as one table for building filter options from in the user interface.
Ok, but can't you add to your query a field that can be related to all these records so it can used for a JOIN?
0
 
dgloverukAuthor Commented:
I think what I'm going to do is put the original Query results into a temporary table and then return the distinct selections from this temporary table to get the filter options as well as the temporary table itself.
Originally I didn't go in this direction because I did not want to break up the process of capturing the query results and then getting then querying for filter options since they could have changed and to get the distinct options based on the original query would have been further overhead for each filter.  But what I didn't know which I've read is that you can return multiple sets of data from a single execution of a stored procedure using NextResult and this basically solves my problem of getting the query and a list of filters based on that query, without resubmitting the original query.
Does this sound like a reasonable approach?
0
 
Vitor MontalvãoMSSQL Senior EngineerCommented:
The NextResult is a .NET method, right?
We are giving you only native SQL Server solutions. Can't talk about the NextResult since isn't my expertise area.
0
 
dgloverukAuthor Commented:
That's right.
I'm assuming that internally to SQL server a stored procedure can return multiple sets of results for handling?
If so, I wasn't aware of this else I might have guessed earlier that there would be a method with .NET to cycle through the different sets of results.
0
 
Vitor MontalvãoMSSQL Senior EngineerCommented:
The only way a stored procedure has to return values is using OUTPUT parameters. You can read more about it here.
0
 
dgloverukAuthor Commented:
I've requested that this question be closed as follows:

Accepted answer: 500 points for ScottPletcher's comment #a40468061
Assisted answer: 0 points for dgloveruk's comment #a40468836

for the following reason:

Thank you for your help guys!
0
 
Scott PletcherSenior DBACommented:
>> I'm assuming that internally to SQL server a stored procedure can return multiple sets of results for handling? <<

That is correct.  By default SQL will return the first result set to the caller, but there are methods to get any additional result set(s), if any.

You can also return a single integer return code, using
RETURN <int_value>
By default, SQL will return 0.

Finally, as noted by Vitor, you can return one or more scalar (single-value) parameters by specifying them as "OUTPUT" parameters.
0
 
Scott PletcherSenior DBACommented:
I've done the same thing before.  You have distinct lists but you want to see them in a compact format.  Joining by line# works perfectly in those cases.
0
 
dgloverukAuthor Commented:
Thanks guys!
0

Featured Post

Free Tool: IP Lookup

Get more info about an IP address or domain name, such as organization, abuse contacts and geolocation.

One of a set of tools we are providing to everyone as a way of saying thank you for being a part of the community.

  • 5
  • 4
  • 3
  • +1
Tackle projects and never again get stuck behind a technical roadblock.
Join Now