[Dbix-class] Lazy Columns

Peter Rabbitson rabbit+dbic at rabbit.us
Tue May 11 19:15:37 GMT 2010


Bill Moseley wrote:
> On Tue, May 11, 2010 at 9:40 AM, Peter Rabbitson <rabbit+dbic at rabbit.us
> <mailto:rabbit%2Bdbic at rabbit.us>> wrote:
> 
> 
>     This means you didn't even read the original thread, which clearly
>     mentioned
>     http://search.cpan.org/~frew/DBIx-Class-0.08121/lib/DBIx/Class/Row.pm#has_column_loaded
> 
> 
> It was four days ago, I forgot.  Sorry.  If I would have remembered I
> guess I would have know you were implying earlier today "Ordered should
> be checking has_column_loaded()".  In that case I'll try later today to
> create a failing test.
> 
>  
> 
>     > Although now the database takes more of a beating... ;)
> 
>     http://dev.catalystframework.org/svnweb/bast/blame/DBIx-Class/0.08/trunk/lib/DBIx/Class/Ordered.pm
>     Read the comment at lines 791-794, I need a volunteer to make me a
>     list of databases
>     which can do that without stepping on its own toes. Then we can make
>     exceptions based
>     on underlying storage.
> 
> 
> Geeze, reduced to the lowest common denominator.   What about a flag to
> indicate that it's ok to do it the smart way?
> 

To which I said:

> I need a volunteer to make me a list of databases which can do that without
> stepping on its own toes. Then we can make exceptions based on underlying
> storage.

Remember DBIC is correct first and fast only afterwards. This is far from the
only "lowest common denominator" reduction in the code, but we currently still
have trouble to test multiple databases. Work is being done on that front
but it is sadly slow.



More information about the DBIx-Class mailing list