Click here to monitor SSC
SQLServerCentral is supported by Red Gate Software Ltd.
 
Log in  ::  Register  ::  Not logged in
 
 
 
        
Home       Members    Calendar    Who's On


Add to briefcase

Incremental pattern matching Expand / Collapse
Author
Message
Posted Friday, July 30, 2010 6:34 AM
Valued Member

Valued MemberValued MemberValued MemberValued MemberValued MemberValued MemberValued MemberValued Member

Group: General Forum Members
Last Login: Tuesday, April 08, 2014 6:47 AM
Points: 69, Visits: 557
Hi all,

I've got a column, called user_code char(3) that is used to store unique codes. The patterns are as follows

Character 1 is A-Z
Character 2 is 0-9,A-Z
Character 3 is 1-9,0,A-Z

So, the data starts at

A01
A02
...
AZZ
B01
B02
..
BZZ
..
ZZZ

Now, what I need to do is write a T-SQL statement that will tell me what the next available (ie unused) user_code can be. I wish I could just do a MAX(user_code) to find the most recently used one and then work out the next value in the sequence, but it's not that simple, as the user_codes assigned aren't in any meaningful order..

If you could help me with this, that'd be awesome. There are no constraints here really, if it means creating another lookup table with all the options in, then that's allowed, as long as the solution works!
Post #961290
Posted Friday, July 30, 2010 12:44 PM
SSC Veteran

SSC VeteranSSC VeteranSSC VeteranSSC VeteranSSC VeteranSSC VeteranSSC VeteranSSC Veteran

Group: General Forum Members
Last Login: Friday, December 31, 2010 9:46 AM
Points: 274, Visits: 473
I suggest creating another table containing every possible 3-char value, clustered on that code (fillfactor 100). Then join to that table and find the first one that isn't in use.

OR

If you (almost) never delete rows where you're adding the 3-char code, delete each 3-char code from the lookup table as you insert it into the main/detail table. Then you would always just select the MIN(code3) from the lookup table.


Scott Pletcher, SQL Server MVP 2008-2010
Post #961583
Posted Friday, July 30, 2010 1:34 PM


Ten Centuries

Ten CenturiesTen CenturiesTen CenturiesTen CenturiesTen CenturiesTen CenturiesTen CenturiesTen Centuries

Group: General Forum Members
Last Login: Wednesday, December 04, 2013 1:49 PM
Points: 1,104, Visits: 1,174
Although the absolute fastest performance is going to be what Scott said, it is possible to do it inline. (And this was worth doing anyway so you could use the same logic to quickly generate the 30K codes if that's what you choose to do.)

--fake table / data
CREATE TABLE #yourTable
(user_code char(3))
INSERT INTO #yourTable (user_code)
select 'A01' union select 'A02' union select 'B01'

--query with inline tally
;WITH E1(N) AS (SELECT 1 UNION ALL SELECT 1 UNION ALL SELECT 1 UNION ALL
SELECT 1 UNION ALL SELECT 1 UNION ALL SELECT 1),--6
E2(N) AS (SELECT 1 FROM E1 a, E1 b),--36
cteTally(N) AS (SELECT ROW_NUMBER() OVER (ORDER BY N) FROM E2),
cteAlphaNum AS
(select N,
Value = case when N <= 26 then char(c1.N + 64)
else cast(N - 27 as varchar) end
from cteTally c1),
cteFinalList AS
(select c1.Value + c2.Value + c3.Value as user_code
from cteAlphaNum c1
cross join cteAlphaNum c2
cross join cteAlphaNum c3
where c1.N <= 26
and (c2.N <= 26 or c3.N <> 27))
select top 1 fl.user_code
from cteFinalList fl
left join #yourTable t
on t.user_code = fl.user_code
where t.user_code is null
order by fl.user_code asc



└> bt


Forum Etiquette: How to post data/code on a forum to get the best help
Post #961606
« Prev Topic | Next Topic »

Add to briefcase

Permissions Expand / Collapse