Hi Luf,

Thanks, patches applied, though I have commented out the mylogs inside
the inner loop of CC_mapping as they *really* churn out a lot of
additional detail. I left them there though so if needed they can just
be uncommented.

Interestingly, this doesn't seem to fix the VB6 rowcount issue which I
still see as -1. Not sure why though - ADO uses SQLRowcount and now gets
the correct result, and SQLGetDiagField with SQL_HANDLE_STMT and
SQL_DIAG_CURSOR_ROW_COUNT which also appears to return the correct
value.

Needs some further thought I think....

Regards, Dave

> -----Original Message-----
> From: pgsql-odbc-owner@postgresql.org
> [mailtogsql-odbc-owner@postgresql.org] On Behalf Of Ludek Finstrle
> Sent: 01 December 2005 20:21
> To: pgsql-odbc@postgresql.org
> Subject: [ODBC] Some changes
>
> Hello list,
>
> when I was seeking patch for server side prepare I found two
> improvements and add some mylog output.
>
> I send here 3 patch + one contains all of them to review.
>
> 1) psqlodbc-skip_cmd_fetch.diff - this skip QR_fetch_tuples when
> backend returns PGRES_COMMAND_OK - there is no data to fetch
> 2) psqlodbc-rowcount.diff - this add support for returning rowcount
> from select statements (with UseDeclareFetch disabled - as
> non-libpq versions)
> 3) psqlodbc-mylog.diff - add some mylog output for better tracing
>
> 4) psqlodbc-all-skip_cmd_fetch-rowcount-mylog.diff
> this one contains all (three) above in one becouse they
> are created
> againist CVS version and they coincidents a little:
>
> $ psqlodbc>patch -p1 < ..\psqlodbc-skip_command_fetch.diff
> $ patching file `connection.c'
>
> $ psqlodbc>patch -p1 < ..\psqlodbc-rowcount.diff
> $ patching file `connection.c'
> $ Hunk #1 succeeded at 1850 (offset 5 lines).
>
> $ psqlodbc>patch -p1 < ..\psqlodbc-mylog.diff
> $ patching file `connection.c'
> $ Hunk #1 succeeded at 1800 (offset 5 lines).
> $ Hunk #3 succeeded at 1887 (offset 13 lines).
> $ Hunk #5 succeeded at 1911 (offset 13 lines).
> $ Hunk #7 succeeded at 2007 (offset 13 lines).
>
> Please feel free to send feedback here
>
> Luf
>

---------------------------(end of broadcast)---------------------------
TIP 9: In versions below 8.0, the planner will ignore your desire to
choose an index scan if your joining column's datatypes do not
match