Wow that was fast and is exactly what I was looking for :D
Thanks,
Nick.
On Feb 3, 4:04 am, Russell Keith-Magee <freakboy3...@gmail.com> wrote:
> On Tue, Feb 3, 2009 at 8:34 PM, Russell Keith-Magee
>
>
>
> <freakboy3...@gmail.com> wrote:
> > On Tue, Feb 3, 2009 at 11:16 AM, alex.gay...@gmail.com
> > <alex.gay...@gmail.com> wrote:
>
> >> On Feb 2, 9:05 pm, nsitarz <nsit...@gmail.com> wrote:
> >>> Hey,
>
> >>> Back when the ORMaggregatesupport was a patch in trac I used to
> >>> createcustomaggregateobjects that looked like this:
>
> >>> class Date(Aggregate):
> >>> pass
>
> >>> class DateHour(Aggregate):
> >>> def __init__(self, lookup):
> >>> super(DateHour, self).__init__(lookup)
> >>> self.sql_template = '(DATE(${field}) + INTERVAL HOUR(${field})
> >>> HOUR)'
>
> >>> However now thataggregatesupport has made its way into trunk the api
> >>> seems to have changed a bit. After reading theaggregatemodule it
> >>> looks to me like I'd have to reimplement theAggregateclass myself in
> >>> order to have this same kind of functionality. Is there a supported
> >>> way of creatingcustomaggregateclasses for use with the trunk ORM?
>
> >>> I'd appreciate any help that anybody can offer on this.
> ...
> > What you will end up with is a backend-specificaggregateclass that
> > appears to be a genericaggregatefor the purposes of annotate() and
> >aggregate() calls.
>
> Curiosity got the better of me... so here's a worked (and documented)
> example that reimplements the Max()aggregate:
>
> from django.db.models.sql.aggregates importAggregate
> class MyMax(Aggregate):
> sql_function = 'MAX'
> """A base class to make it easy for end users to define their own
> customSQL aggregates.
>
> The subclass should define the following two class properties:
> * sql_function - the name of the SQL function to invoke
>
> Optionally, you can define
> * sql_template - a format string that is used to compose the
> SQL that will be sent to the database. The template will be
> provided with the following substitution variables:
> - ``function``, the sql fuction that will be invoked
> - ``field``, the resolved name of the column to be
> operated on.
> The template will also be provided with any keyword argument
> provided to theaggregatewhen it was defined.
> The default template is '%(function)s(%(field)s)'
> * is_ordinal - a boolean, True if the result of theaggregate
> will always be a count, regardless of the field on which the
> aggregateis applied. False by default.
> * is_computed - a boolean, True if the result of theaggregate
> will always be a float, regardless of the field on which the
> aggregateis applied. False by default.
> """
> def __init__(self, lookup, **extra):
> self.lookup = lookup
> self.extra = extra
>
> def _default_alias(self):
> return '%s__%s' % (self.lookup, self.__class__.__name__.lower())
> default_alias = property(_default_alias)
>
> def add_to_query(self, query, alias, col, source, is_summary):
> super(MyMax, self).__init__(col, source, is_summary, **self.extra)
> query.aggregate_select[alias] = self
>
> Usage:
>
> Book.objects.annotate(MyMax('pubdate'))
>
> Yours,
> Russ Magee %-)
--~--~-~--~~~---~--~~
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
-~--~~~~--~~--~--~---
