Tim

In my continuing quest I tried

select distinct x.type,name,type,collation
from information.tablecolumns TC
left join information.IndexColumns XC on
XC.ColumnName = TC.Name
and
XC.TableName = TC.TableName
left join information.indexes X ON X.Name = XC.IndexName
and
X.TableName = TC.TableName
and x.type = 'Text'
WHERE TC.tablename = 'MandN'

EDBMan tells me there are 25 rows. The query plan tells me there are 25 rows. I count 34 rows. Either I or ElevateDB can't count

Roy Lambert

Tim

I think its sql. It gets worse.

select distinct x.type,name,type,collation
from information.tablecolumns TC
left join information.IndexColumns XC on
XC.ColumnName = TC.Name
and
XC.TableName = TC.TableName
left join information.indexes X ON X.Name = XC.IndexName
and
X.TableName = TC.TableName
and x.type = 'Text'
WHERE TC.tablename = 'MandN'
order by name,x.type,type,collation

Returns 29 rows. I added the order clause because I thought that would be the best way to see if there were any duplicates. There are but they differ depending if the order clause is there or not.

The database I'm using is the one on the DVD


Roy Lambert

Roy,

<< EDBMan tells me there are 25 rows. The query plan tells me there are 25
rows. I count 34 rows. Either I or ElevateDB can't count >>

The correct answer is 25 rows, and this is another issue that has already
been fixed, apparently, because it works just fine using the 1.03 candidate
code.  As indicated before, the DISTINCT, grouping, and having was rewritten
to be faster and more low-level.

--
Tim Young
Elevate Software
www.elevatesoft.com

Roy,

<< I think its sql. It gets worse. >>

Same situation - works fine here with the 1.03 (not released) code.

--
Tim Young
Elevate Software
www.elevatesoft.com