Iain wrote:
> This is what we do, however we forgot to turn off cascading on the
> extra relationship e.g.
>
> {cascade_delete => 0, cascade_copy => 0, join_type => 'left'}
>
> and it caused us all kinds of fun.
hehe. ;-)
Is it necessary to put the cascade_* attrs on both sides, or is it
enough
On Thu, 2009-02-19 at 18:13 +0100, Bernhard Graf wrote:
> Nigel Metheringham wrote:
>
> > Not easily for a single query, however a relationship can have a
> > join_type attribute - relationship attributes are documented in
> > DBIx::Class::Relationship::Base under add_relationship
>
> I know this
Nigel Metheringham wrote:
> Not easily for a single query, however a relationship can have a
> join_type attribute - relationship attributes are documented in
> DBIx::Class::Relationship::Base under add_relationship
I know this, but I actually wanted to use both join types.
> You can always have
michael reddick wrote:
> something like this i think is what you're after...
>
> $schema->resultset('User')->search(
> { items.id => { '!=' => undef },
> {
> join => 'items'
> group_by => 'me.id',
> }
> );
I think that would work, although using the "right" join type seems more
natura
On 19 Feb 2009, at 16:35, Bernhard Graf wrote:
Is it possible to specify the join type for the result set?
Not easily for a single query, however a relationship can have a
join_type attribute - relationship attributes are documented in
DBIx::Class::Relationship::Base under add_relationshi
On Thu, Feb 19, 2009 at 10:35 AM, Bernhard Graf wrote:
> Is it possible to specify the join type for the result set?
>
> E.g. I have a user who might have several items, but I want only those
> users, that actually have items.
>
> E.g. this gives me a left join:
>
> $schema->resultset('User')->se
Is it possible to specify the join type for the result set?
E.g. I have a user who might have several items, but I want only those
users, that actually have items.
E.g. this gives me a left join:
$schema->resultset('User')->search(
{},
{
join => 'items'
group_by => 'me.id',
}
)
