[Dbix-class] [ANNOUNCE] 0.08103 RC3 - call for testers! (whops)

[Peter Rabbitson]

Anthony Gladdish wrote:
> Hi,
> 
> Using DBIC v0.08102_07.
> 
> My Result class relationships:
> 
> 1. Course.pm
> 2. Event.pm:
> __PACKAGE__->belongs_to( 'course' => 'Course', 'based_on' );
> 
> My test:
> 
> my $search;
> push( @$search, ('me.code', { 'like', '%i1%' } ) );
> my $rs = $schema->resultset('Event')->search(
>         {
>           -or => $search,          
>         },
>         {
>           distinct => 1,
>           prefetch => 'course',
>         }
> );
> is $rs->count, 1, '$rs ok';
> 
> Fails with:
> 
> DBIx::Class::ResultSet::count(): DBI Exception: DBD::mysql::st execute failed: Duplicate column name 'id' [for Statement "SELECT COUNT( * ) FROM (SELECT me.id, me.code, me.start, me.hours, me.end, me.leader, me.based_on, me.attend_max, me.created_by, me.attend_min, me.RT_ticket, me.status, me.created, me.location, me.price_1, me.price_2, me.price_3, me.price_4, me.price_5, course.id, course.title, course.topics, course.tasks, course.prerequisites, course.overview, course.target, course.quotes, course.code, course.hours, course.sub_title, course.further_tasks, course.short_desc, course.shorter_desc FROM event me  JOIN event_defaults course ON course.id = me.based_on WHERE ( me.code LIKE ? ) GROUP BY me.id, me.code, me.start, me.hours, me.end, me.leader, me.based_on, me.attend_max, me.created_by, me.attend_min, me.RT_ticket, me.status, me.created, me.location, me.price_1, me.price_2, me.price_3, me.price_4, me.price_5, course.id, course.title, course.topics, course.tasks, cou
rse.prerequisites, course.overview, course.target, course.quotes, course.code, course.hours, course.sub_title, course.further_tasks, course.short_desc, course.shorter_desc) count_subq" with ParamValues: 0='%i1%']
> 
> Passes, if I comment out either "distinct" or "prefetch" but this is undesired.
> 
> Am I doing anything wrong?
> Has this been fixed in a newer RC yet that I can test?
> 

This has been fixed by accident, while doing count query optimizations. I just
shipped the 0.08103 - please give it a try and don't hesitate to report any
issues - we can always make another release. Here is the problem with mysql:


This fails with Duplicate column name 'id':  (old incorrect code, would return
the count of all joined rows)

SELECT COUNT( * ) FROM (
  SELECT me.id, me.name, books.id, books.source, books.owner, books.title, books.price
    FROM owners me LEFT JOIN books books ON books.owner = me.id WHERE ( books.id IS NOT NULL )
  GROUP BY me.id, me.name, books.id, books.source, books.owner, books.title, books.price
) count_subq


This passes: (new *correct* code)

SELECT COUNT( * ) FROM (
  SELECT me.id
    FROM owners me LEFT JOIN books books ON books.owner = me.id WHERE ( books.id IS NOT NULL )
  GROUP BY me.id
) count_subq


Cheers