I'm not fond of using fixtures for seed data, so I, like many people (including some Rails Core guys) use migrations (very sparingly) to add some pieces of data that will be used in development.
There are other ways to solve the "seed" problem, but using migrations has been the easiest to use, for me, when I want some data in my database for development only. I'm not saying this is the right solution, but it has been the easiest one for me, thus far. I'm working on something to fix this and extract seed data from migrations and still keep that simplicity that I like when including seed data in migrations. I'm fine with DHH's decision on rake db:reset. It makes sense for the viewpoint that he has presented. I just don't agree with creating a new namespace for the reason you provided. Regardless, I've written a plugin to cover these issues for my own usage. The link is in the above discussion on rails core forum. On Dec 10, 10:42 am, "Matt Aimonetti" <[EMAIL PROTECTED]> wrote: > It's more of a political problem than anything guys. DHH doesn't want to > hear about rake db:rest and the migration should not be used a seeder. So in > production mode you should always use the schema.rb file. > > I created another namespace in the hope to be able to still have the old > behavior, but understand that this behavior should not be used on a regular > basis. > > -Matt > > -- > --------http://railsontherun.com > > On 12/10/07, Patrick Crowley <[EMAIL PROTECTED]> wrote: > > > > > > I personally do not like the name rake migrations:reset, it's a bit > > > long and I think anything having to do with altering the database, > > > should be in the db namespace. > > > +1 > > > -- Patrick --~--~---------~--~----~------------~-------~--~----~ SD Ruby mailing list [email protected] http://groups.google.com/group/sdruby -~----------~----~----~----~------~----~------~--~---
