Well, here we are 7 months later, and I am finally back around to working on this Sql Sever conversion again. I have done a small test with your suggestion, and it seems like it is going to work out pretty well.

Thanks!



>>>The alternative way is to pass local cursor to the SQL Server, e.g. something like this
>>>text to lcInsert noshow
>>>declare @LocalTable (field1 varchar(10), field2 int)
>>>endtext
>>>
>>>select local cursor
>>>scan
>>>
>>> using textmerge create insert statements for the local cursor
>>>endscan
>>
>>Does that @LocalTable remain in the Sql Server database afterwards?
>>
>>And, what if you have multiple users doing this process at the same time? They will all be trying to use the same @LocalTable and the records would get all jumbled up.
>>
>For such tiny local cursors used as join filters you can locally pipe their PKs into a

server.pk IN (client.pk1, client.pk2... client.pk_max)

value_set clause to be used on the server in the where-clause.
>
>

Select Users.email ;
>	From Jobs  ;
>		Join Users On Jobs.quoted_by = Users.emp_num ;
>	Where qty_recd <> before_qty_recd ;
>              and Jobs.job_num in (&CSV_ListOfValuesProducedLocallyIntoString)
>	Order By 1 ;
>	Group By 1 ;