On Tue, 7 Feb 2006, Brandon Black wrote:

run in a big app with a big/complex database schema (in development or
production - just in general).  It's purpose is to get people up and
running quickly against an existing database that wasn't built with
DBIx::Class in mind, or is created/managed by other tools at the time.

I just went "huh?" here.. What is different about a DB that's been designed with DBIx::Class in mind? Or did you just mean "properly normalised" ?

It tries to divine the ideal DBIx::Class::Schema setup based on what
it can see on your live running database server, but it cannot
possibly get it all correct - it will always be just a good
approximation of what you might have done if you defined the Schema
manually when it comes to relationships and such.

[I think I've written the above in some form like 4 times now to
various people, perhaps I should patch that into the Schema::Loader
documentation?]

Great idea! ;)

Unless your app/db is fairly simplistic, you will probably want to
migrate off of Schema::Loader after a while, and you'd do that by
using the DBIx::Class SQLT support to dump the
Schema::Loader-generated Schema out to a manual definition that you
can edit and refine from there.

IMO (and I may be off base), it should be possible to use Loader, which will have a standard set of predictable results (which should be detailed in the docs), and then only write your own classes for the parts you need to change / expand. Some flag should possibly set whether the user-created classes replace or extend the Loader ones .. ?

Jess


_______________________________________________
List: http://lists.rawmode.org/cgi-bin/mailman/listinfo/dbix-class
Wiki: http://dbix-class.shadowcatsystems.co.uk/
IRC: irc.perl.org#dbix-class
SVN: http://dev.catalyst.perl.org/repos/bast/trunk/DBIx-Class/

Reply via email to