[Catalyst] Re: a noob delete/cascade problem

Matt S Trout dbix-class at trout.me.uk
Wed Sep 12 17:47:25 GMT 2007


On Tue, Sep 11, 2007 at 10:20:50PM -0700, Dustin Suchter wrote:
> Well, I found a solution to my problem. I'm not sure it is the most
> elegant, but here goes:
> 
> I figured that just putting the proper cascade logic into my actual
> DB schema would fix things, and it does. I changed my table
> definition to:
>
> I think a better solution (TIMTOWTDI) is to for DBIx to somehow
> automatically detect the constraint and transact a delete that
> removes the child rows first and then the parent rows, in order to
> make sure everything gets deleted. I think that one is a pipe dream
> so I'll happily use my solution above for now.

Couple of notes:

(1) it's DBIx::Class, DBIx:: is the namespace for -all- DBI extensions
(2) it has its own mailing list

As for automatically detecting the constraint, if you use $schema->deploy to
let DBIC generate your CREATE TABLE statements for you it puts in the
constraint cascades automatically so you'll never see a problem.

We intentionally don't override cascade/fk settings on your database to avoid
buggy code causing mass dataloss (I've seen this many times before now with
ORMs that try to be helpful the way you're requesting) - DBIC -will- try the
delete -after- deleting the main record for DBs like SQLite without FKs, but
in the case where your database has constraints DBIx::Class prefers to respect
them.

Either do the cascade clearly in your own code or make sure your DB schema's
correct - at least that way when a cascade wipes out a load of data it's
easy for the maintenance programmer to see why and how it's happened :)

-- 
      Matt S Trout       Need help with your Catalyst or DBIx::Class project?
   Technical Director    Want a managed development or deployment platform?
 Shadowcat Systems Ltd.  Contact mst (at) shadowcatsystems.co.uk for a quote
http://chainsawblues.vox.com/                    http://www.shadowcat.co.uk/ 



More information about the Catalyst mailing list