I tried to find the bug in the sources and found
out that TIB_Grid gets a bit confused when one uses
the scroll wheel of the mouse.

With rowselect=true, the record pointer remains on
the current record and the current record remains
highlighted. Up to now, this is OK (I guess).

But the when you are scrolling (using the wheel),
another row gets highlighted, so there are to highlighted rows.
You can even increase confusion pressing the up or down key -
then the IB_Grid highlights the selected row - moves the record
pointer to it - but forgets to erase the old record pointer.
Then you have one highlighted row, but two marked by a record pointer.

I've put three hours of my time at work trying to find the error (I guess
my boss won't like it), but I am stuck.
I would be able to proceed if someone could tell me which event is generated
when the scroll wheel of the mouse is used.


TIA,

Christian

> -----Original Message-----
> From: Jason Wharton [mailto:jwharton@...]
> Sent: Tuesday, February 06, 2001 7:17 PM
> To: IBObjects@yahoogroups.com
> Subject: Re: [IBO] IB_Grid, Rowselect and TrackGridRow
>
>
> It sounds like a bug. Get up to the latest source and see if
> you can fix it
> in IBG_Custom.IMP and send me the changes. I'm sure it is
> probably something
> simple. I'll add this to my list of items to check myself.
>
> Regards,
> Jason Wharton
> CPS - Mesa AZ
> http://www.ibobjects.com
>
>
> ----- Original Message -----
> From: "Christian Gütter" <cguetter@...>
> To: <ibobjects@yahoogroups.com>
> Sent: Tuesday, February 06, 2001 5:22 AM
> Subject: [IBO] IB_Grid, Rowselect and TrackGridRow
>
>
> > Hi,
> >
> > I am using a TIB_Grid with Rowselect = True
> > and TrackGridRow = False.
> >
> > When I use the wheel of my mouse to scroll
> > through the grid, the record pointer does not
> > move (the line which is pointed at remains
> > highlighted) and the row I scroll to is also
> > highlighted. So there are to highlighted rows,
> > which is very confusing.
> > It is even more confusing because TrackGridRow
> > is false. This property seems to ignore the state
> > 'false'.
> >
> > Could this be a bug or am I too dumb?
> >