[Dbix-class] possible bug in DBIx::Class::Row

[Matt S Trout]

On Wed, May 13, 2009 at 02:18:15PM +0400, Oleg Pronin wrote:
> in DBIx::Class::Row::insert
> 
> if(!$self->{_rel_in_storage}) {
>     #heavy code
> }
> 
> #insert code
> 
> 
> Example:
> 
> $obj1 = $rs->new({...});
> $obj2 = $obj1->new_related('abc', {..});
> 
> $obj1->insert; #ok
> 
> #this cause 'heavy code' to execute because $self->{_rel_in_storage}
> is undefined although $obj1 has been inserted in previous line
> $obj2->insert;
> 
> 
> I think _rel_in_storage should be updated when related object is
> inserted or should be evaluated at insert

I'd call this a performance pessimisation rather than a bug, but I do see
what you mean. If you modify the code to set _rel_in_storage there, do any
of the tests fail?

-- 
        Matt S Trout         Catalyst and DBIx::Class consultancy with a clue
     Technical Director      and a commit bit: http://shadowcat.co.uk/catalyst/
 Shadowcat Systems Limited
  mst (@) shadowcat.co.uk        http://shadowcat.co.uk/blog/matt-s-trout/