dbreap

Namespaced seed data - saved in YML. Much easier to see what's going on with your seed data than a bunch of ruby code that no-one maintains.

Share essential database records between developers, staging, and demo environments.

# Gemfile
gem "dbreap"

bundle install
rails generate dbreap:install

This creates db/fixtures/ and installs the seed loader into db/seeds.rb (appends if the file already exists).

Working on a new feature - Does having some example data help other developers on dev work with and test your new code in the UI?

Create your data in the UI or console - and commit the seed YML alongside the feature code.

# Reap specific tables
rake db:reap FIXTURES=products,categories

# Reap everything
rake db:reap FIXTURES=ALL

Fixtures are written to db/fixtures/<env>/. Commit these files.

# Clear existing data, then seed from fixtures
rake db:empty
rake db:seed

db:seed uses Rails' standard db/seeds.rb, configured by the install generator.

bundle install
bundle exec lefthook install  # installs pre-commit hooks
bundle exec rspec

• Fixture files are ordered by id — reap → seed → reap produces identical output
• JSON columns are preserved as structured data
• String columns containing numeric values (e.g. "123") are not cast to integers
• schema_migrations and ar_internal_metadata are always skipped
• db:empty uses DELETE (not TRUNCATE) — schema and sequences are preserved, no DDL lock
• Run db:empty before db:seed to avoid duplicate records on repeated seeds
• Fixture files are namespaced by environment (db/fixtures/<env>/) — development, staging, and demo data can coexist