How does works "group by" correct

[SoftIl]

Great, thats a very good site.
Many thanks for this tip. :-)

Best regards,
Carsten

----- Original Message ----- 
From: "Halldor Gislason" <halldorg at vortex.is>
To: "Valentina Developers" <valentina at lists.macserve.net>
Sent: Monday, August 16, 2004 4:57 PM
Subject: Re: How does works "group by" correct


> Hi Carsten,
> Here is a good refernce with examples:
> 
> http://www.1keydata.com/sql/sqlgroupby.html
> 
> best regards,
> Halldor
> 
> On Aug 16, 2004, at 10:38 AM, SoftIl wrote:
> 
> > Hallo Halldor,
> > thanks for your answer.
> > Now I understand where my mistake was.
> > I didn't know that I can use goup by only with an aggregate function.
> > Now it is clear, why it could not work.
> >
> > Thanks
> > Carsten
> >
> >
> > ----- Original Message -----
> > From: "Halldor Gislason" <halldorg at vortex.is>
> > To: "Valentina Developers" <valentina at lists.macserve.net>
> > Sent: Thursday, August 12, 2004 11:49 PM
> > Subject: Re: How does works "group by" correct
> >
> >
> >> "select *" returns all the fields so a group by does not really have a
> >> purpose there.
> >>
> >> the purpose of the group by is to group the results of some aggregate
> >> function by some fields so a group by without such a function has
> >> little point
> >>
> >> you may instead be looking for the UNIQUE statement that will not
> >> return records with duplicate values of of the field values you 
> >> specify
> >> as unique - I do not know how well Valentina supports UNIQUE
> >>
> >> to group by more than one field simply do:
> >>
> >> select a,b, count(*) from c group by a,b
> >>
> >> or
> >>
> >> select a,b, sum(c) from d group by a,b
> >>
> >> h
> >>
> >>
> >>
> >> On Aug 12, 2004, at 9:34 PM, SoftIl wrote:
> >>
> >>> Hi all,
> >>> I've got some problems with "group by":
> >>>
> >>> 1)
> >>> If I try this:
> >>> select * from myTable group by date
> >>> the "group by" statement hasn't an effect.
> >>>
> >>> If I write this
> >>> select *,count(any column) from myTable group by date
> >>> it works correct.
> >>>
> >>> Why is it so? In no documentation I could find this, that I have to 
> >>> use
> >>> "count()" if I'd like to use "group by".
> >>>
> >>> 2)
> >>> How can I use "group by" for more than one column?
> >>> This statement works not correct. I get all records return:
> >>> select *,count(any column) from myTable group by date,ohrnummer
> >>>
> >>> Thanks for your help.
> >>> Carsten
> >>>
> >>>
> >>> _______________________________________________
> >>> Valentina mailing list
> >>> Valentina at lists.macserve.net
> >>> http://lists.macserve.net/mailman/listinfo/valentina
> >>>
> >>
> >> _______________________________________________
> >> Valentina mailing list
> >> Valentina at lists.macserve.net
> >> http://lists.macserve.net/mailman/listinfo/valentina
> >>
> >
> > _______________________________________________
> > Valentina mailing list
> > Valentina at lists.macserve.net
> > http://lists.macserve.net/mailman/listinfo/valentina
> >
> 
> _______________________________________________
> Valentina mailing list
> Valentina at lists.macserve.net
> http://lists.macserve.net/mailman/listinfo/valentina
>