Rake db:schema:load insists on querying my database for data I haven't loaded

Rails 2.3.2. Ruby 1.8.7. MySQL 5.x

I have used “rake db:schema:dump” to generate a migration structure to
go from development to production. I want to use “rake
db:schema:load” to load it into my new database. Sounds like a plan.

When I run “rake db:schema:load” on the new database (development or
production), I get the following error:

rake aborted! Mysql::Error: Table 'mojo_development.statuses' doesn't exist: SELECT * FROM `statuses` WHERE (`statuses`.`name` = 'New') LIMIT 1

(production gives the same, but database name is mojo_production).

The Select is not in the schema. If I trace it back, I find the
Select is generated from the following method in my Mojo model:

def self.new_state
for_name(‘New’).first
end

where for_name is a :named_scope generating the select statement
above.

So my questions are:

  1. WHY does db:schema:load run queries against my database when the
    data (and tables) aren’t there?

and

  1. HOW do I fix this so it won’t happen again?

Posting the results of running the rake command with --trace always
helps, but I’d say the likely culprit here is something in your
environment / initializers. Does that function get called from
environment.rb or an initializer?

–Matt J.

Eric Smith wrote:

Rails 2.3.2. Ruby 1.8.7. MySQL 5.x

I have used “rake db:schema:dump” to generate a migration structure to
go from development to production. I want to use “rake
db:schema:load” to load it into my new database. Sounds like a plan.

When I run “rake db:schema:load” on the new database (development or
production), I get the following error:

Your rake task is picking up your development environment from
config/environment.rb.

Here’s what you can do to start off with a clean production schema
provided you are the only one working on your db and you have
foolishly/wisely modified your db schema without going through a proper
migration strategy…

  1. perform a rake db:schema:dump
  2. remove all prior migration files
  3. perform a rake db:migrate
  4. open the migration file just created, and cut and paste the contents
    from the schema.rb file that you created in step 1. (during the paste,
    just copy the relevant parts, avoid the database create part…)
  5. Now you can use that migration file on production
  6. Ask that girl in finance out on a date like you promised yourself 2
    weeks ago.

hth

ilan

On Aug 11, 5:31 am, Polydectes [email protected] wrote:

  1. WHY does db:schema:load run queries against my database when the
    data (and tables) aren’t there?

and

  1. HOW do I fix this so it won’t happen again?

rake db:schema:load requires the environment task (which loads your
app config and so on)
in production config.cache_classes is true, which means that your
applications models are loaded
your Mojo model is loaded, and reaches the line

named_scope :some_scope, :conditions => {… self.new_state }

which runs your new_state ,method, which queries the database

One possible solution: alter your environment so that cache classes
doesn’t get set to true in cases like this. I outlined a few ways of
doing this at
http://www.spacevatican.org/2008/12/28/when-cache_classes-gets-you-down

Fred

This forum is not affiliated to the Ruby language, Ruby on Rails framework, nor any Ruby applications discussed here.

| Privacy Policy | Terms of Service | Remote Ruby Jobs