Re: long way to WCHAR...

On Tuesday 04 November 2008 15:28:46 Frediano Ziglio wrote:
> 2008/10/31 Frediano Ziglio <[email protected]>
>
> >> Hi Freddy,
> >>
> >> I have sent you a gzip of the odbc trace privately so as not to annoy
> >> any list
> >> users with attachments.
> >>
> >>
> >> To start with I re-read my post and noticed I used confusing
> >> terminology.
> >>
> >> "The data gets mixed up between the queries when populating my tables."
> >>
> >> Should of read.
> >>
> >> "The data gets mixed up between the queries when displaying my data in
> >> my program."
> >>
> >>
> >> I use threads to run the queries that display and format the data in my
> >> program. This is so long running queries do not block the interface.
> >>
> >> So as far as I can tell, I have one connection being used by multiple
> >> threads
> >> at the same time running queries.
> >>
> >> Royce
> >>
> >> Thanks. I think the problem is not only with threads, SQLGetData use
> >> just
> >
> > current_results but with cursors this can be the results of another
> > cursor, we should use stmt->cursor->res_info. SQLFetch/SQLFetchScroll
> > lock the statement so current_results is safe here...
> > I'll write a test... and obviously a fix :)
>
> I wrote a (backported) patch in CVS HEAD, see
> http://freetds.cvs.sourceforge.net/viewvc/freetds/freetds/src/odbc/odbc.c?r
>1=1.503&r2=1.504&sortby=date Test on cursor3
>
> bye
>   freddy77

Freddy you're a star, my program is no longer mixing up the query results. (I 
am using CVS HEAD)

Thanks,
Royce