On Tue, Mar 3, 2009 at 11:46 AM, Ross <real...@gmail.com> wrote:
>
> As I thought about it more, I came up with something similar to what
> both of you suggested. Django's query language has made my SQL
> knowledge suffer a bit--your explanation about why the limit won't
> work makes sense, Jeff.
>
> Thanks a lot for the suggestions.
>
> On Mar 3, 10:39 am, Jeff FW <jeff...@gmail.com> wrote:
> > This time I didn't, apparently, read what you wanted exactly. Here I
> > was, talking about sums, when all you want are the minimum and
> > maximum. Apparently, I shouldn't answer mailing lists in the
> > morning.
> >
> > Anyway, *most* of what I said holds true, as min and max work pretty
> > much the same way as sum--in Python, and in SQL.
> >
> > On Mar 3, 11:36 am, Jeff FW <jeff...@gmail.com> wrote:
> >
> > > Responded too quickly :-)
> >
> > > If you're already getting a list of the top 100 products (and
> > > displaying them, I assume, in a loop,) then totalling up the prices in
> > > Python really won't hurt at all. I'd only suggest going with my
> > > *previous* suggestion if you *weren't* already fetching the top 100
> > > products.
> >
> > > -Jeff
> >
> > > On Mar 3, 11:34 am, Jeff FW <jeff...@gmail.com> wrote:
> >
> > > > The behavior is there because you can't limit an aggregate function
> in
> > > > (AFAIK) SQL in that way. It just doesn't make sense--what would this
> > > > actually mean?
> >
> > > > select sum(price) from product limit 100;
> >
> > > > Really, you'd be limiting the number of *rows* of sum returned,
> which,
> > > > unless you're using GROUP, is going to be 1. To do what you want,
> > > > you'd need a subselect, which (correct me if I'm wrong,) you'd have
> to
> > > > manually write in SQL. Like so:
> >
> > > > select sum(price) from (select price from product order by price desc
> > > > limit 100) as q;
> >
> > > > (That's postgres; your DB may vary.)
> >
> > > > I could be completely off-base here, as I haven't delved very far
> into
> > > > the aggregate code, but from what I can tell, this is the case.
> >
> > > > -Jeff
> >
> > > > On Mar 3, 11:08 am, Alex Gaynor <alex.gay...@gmail.com> wrote:
> >
> > > > > On Tue, Mar 3, 2009 at 8:55 AM, Ross <real...@gmail.com> wrote:
> >
> > > > > > I have started using aggregation, but it seems to ignore any
> slicing I
> > > > > > do on a QuerySet before calling aggregate. This is what I'm
> doing:
> >
> > > > > >
> Product.objects.order_by("-price")[:100].values("price").aggregate(Max
> > > > > > ("price"), Min("price"))
> >
> > > > > > I want the maximum and minimum price for the products with the
> top 100
> > > > > > largest prices. The aggregate function, however, returns the
> maximum
> > > > > > and minimum price for all Products--it ignores the [:100] slice.
> >
> > > > > > Is this an improper use of aggregation, or am I doing something
> wrong
> > > > > > with my query?
> >
> > > > > Before an aggregation is preformed all limits are removed, so you
> are seeing
> > > > > expected behavior. I can't remember why this behavior exists
> though :/
> >
> > > > > Alex
> >
> > > > > --
> > > > > "I disapprove of what you say, but I will defend to the death your
> right to
> > > > > say it." --Voltaire
> > > > > "The people's good is the highest law."--Cicero
> >
>
You could actually probably do it in django's query language by doing:
Product.objects.filter(price__in=Product.objects.order_by('-price').values('price')[:100]).aggregate(Max('price'),
Min('price')).
Alex
--
"I disapprove of what you say, but I will defend to the death your right to
say it." --Voltaire
"The people's good is the highest law."--Cicero
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"Django users" group.
To post to this group, send email to django-users@googlegroups.com
To unsubscribe from this group, send email to
django-users+unsubscr...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/django-users?hl=en
-~----------~----~----~----~------~----~------~--~---
