[Dbix-class] search_literal

[Matt S Trout]

On Fri, Sep 21, 2007 at 09:20:16AM +1000, brett gardner wrote:
> Matt S Trout wrote:
> >On Thu, Sep 20, 2007 at 08:43:20AM +0100, Ash Berlin wrote:
> >  
> >>brett gardner wrote:
> >>    
> >>>I think I have found a bug with search_literal. If you get a resultset 
> >>>that came from a search_literal, and try to apply another search_literal 
> >>>to it, the bind values are in the wrong order eg.
> >>>
> >>>my $rs = $schema->resultset('Foo')->search_literal('foo = ?', 1);
> >>>$rs = $rs->search_literal('bar = ?',2);
> >>>
> >>>Spits out the sql
> >>>
> >>>SELECT foo.field
> >>>FROM foo
> >>>WHERE ( ( foo.bar = ? AND foo.foo = ? ) )
> >>>
> >>>With a bind values array of "[1,2]". It should be a bind values array of 
> >>>"[2,1]".
> >>>
> >>>It seems to be adding the second literal string to the front of the 
> >>>where clause, but adding the values to the end of the array.
> >>>
> >>>I'm using 0.08003 but a quick search in the change log doesn't mention 
> >>>anything about search_literal
> >>>
> >>>Cheers,
> >>>Brett Gardner
> >>>
> >>>      
> >>Sounds like a bug to me - can you please whip up a patch that gives us a
> >>failing test case?
> >>    
> >
> >Plus a doc patch saying "this only exists for Class::DBI compatibility, do
> >-not- use in normal DBIx::Class code".
> >
> >  
> Do you mean do not use "search_literal" at all in DBIx::Class code?

Correct.

There's never any reason to and it doesn't chain properly (as you've just
noticed ...)

-- 
      Matt S Trout       Need help with your Catalyst or DBIx::Class project?
   Technical Director                    http://www.shadowcat.co.uk/catalyst/
 Shadowcat Systems Ltd.  Want a managed development or deployment platform?
http://chainsawblues.vox.com/            http://www.shadowcat.co.uk/servers/