FreeTDS Development Group

["ZIGLIO, Frediano, VF-IT" <Frediano.Ziglio AT vodafone.com>]

...
> > 
> > If we end up adding in rowcounts generated by triggered 
> > updates/inserts/deletes - so what ? we would be returning a valid 
> > total of the number of records affected by the original statement.
> 
> For my purposes, I think this would work.  For a general 
> concept of "correctness" - from my perspective as a user - 
> SQLRowCount() should give the same answer for the same 
> command, regardless of whether it was prepared and executed, 
> or just execdirected.  In the case of triggered activity, does
> SQLRowCount() include rows affected by the triggers when you 
> use SQLExecDirect()?  I'll try to find a moment to test this 
> under Windows later today.
> 
> Big thanks to everyone who is looking into this for me!
> 

Hi Matthew!
The problem is that results from triggers are exactly the same from your
update. Using SQLPrepare is like creating a store procedure and calling
this store procedure instead of executing directly. The wire change very
deeply. If you issue (with SQLPrepare) a "delete ... select ..." you get
(with FreeTDS or MS ODBC drivers) only a single recordset... the same
query with SQLExecDirect (but without parameters) return two recordset.
every delete/insert/update before any rows are just ignored. See
moreandcount test (not last commit); this test reflect exactly the
behaviour of MS and FreeTDS ODBC driver.

freddy77