FreeTDS Development Group

[Sebastien FLAESCH <sf AT 4js.com>]

Hi all,

From my own experience, writing different ODBC interfaces to DB2, SQL Server
(Native Client and older NDAC client), I am always doing:

    r = SQLExecute(st->stmtHandle);
    if (!SQL_SUCCEEDED(r) && r != SQL_NO_DATA) {
        /* handle error */
    }

My 2 cents.
Seb

James K. Lowden wrote:
> Raju Mohanapu wrote:
> > SQLExecute (preparedUpdateStatement) returns SQL_SUCCESS even when no 
> > rows are affected. [...] I believe this 
> > is the behavior indicated in ODBC docs as well.
>
> http://msdn.microsoft.com/en-us/library/ms713584(VS.85).aspx
>
>         "If SQLExecute executes a searched update or delete statement that 
> does
> not affect any rows at the data source, the call to SQLExecute returns
> SQL_NO_DATA."
>
> This was discussed extensively toward the end of February 2008.  See
> specifically Peter Deacon's post Wed, 27 Feb 2008 17:29:30 PST Sb:
> "SQL_NO_DATA or SQL_SUCCESS".  Observed driver behavior differs from the
> vague documented specified behavior, if that's possible!  
>
> > For performance reasons, I wouldn't like to execute 
> > another command on the database to find whether any rows are affected by
> >
> > the just executed update statement. It would be ideal for me if the 
> > SQLExecute returned SQL_NO_DATA if no rows are affected. 
>
> http://msdn.microsoft.com/en-us/library/ms711835(VS.85).aspx
>
>         "SQLRowCount returns the number of rows affected by an UPDATE, 
> INSERT, or
> DELETE statement...."
>
> SQLRowCount does not interrogate the server.  It returns information the
> driver already has.  The server included the rowcount in the DONE packet
> it sent to indicate the command was successfully completed.  
>
> SQLRowCount is also the only portable way to learn the rowcount, zero or
> nonzero.  :-)
>
> HTH.  
>
> --jkl
>
>
> _______________________________________________
> FreeTDS mailing list
> FreeTDS AT lists.ibiblio.org
> http://lists.ibiblio.org/mailman/listinfo/freetds