git clone git://github.com/markiz/schema_enumerator.git && \
cd schema_enumerator && \
rake install

require 'schema_enumerator'
enum = SchemaEnumerator.new({:adapter => 'mysql2', :database => 'test'})
# see http://sequel.rubyforge.org/rdoc/files/doc/opening_databases_rdoc.html
# for help on connection to db
people, users = enum.table("people", "users")
p people.fields
p people.indices

puts people.diff(users, :color)
# {:fields=>
#    {:id=>
#      {:allow_null=>false,
#       :default=>nil,
#       :primary_key=>true,
#       :db_type=>"integer"},
#     :name=>
# -    {:allow_null=>true,
# +    {:allow_null=>false,
# ...

# Also advanced diff:
pp people.diff(users, :hash)
# {:missing_fields=>{},
#  :extra_fields=>{},
#  :changed_fields=>
#   {:name=>
#     {:other=>
#       {:default=>nil,
#        :primary_key=>false,
#        :allow_null=>true,
#        :db_type=>"string"},
#      :own=>
#       {:default=>nil,
#        :primary_key=>false,
#        :allow_null=>false,
#        :db_type=>"string"}}},
#  :missing_indices=>{},
#  :extra_indices=>{[:name]=>{:unique=>false, :columns=>[:name]}}}

enum = SchemaEnumerator.new({:adapter => 'mysql2', :database => 'test'})
people, users = enum.table(:people, :users)
# Generates non-destructive migration to make `users`
# as identical to `people` as possible
puts SchemaEnumerator::MigrationGenerator.new(:people, :users).sequel_migration
# alter_table(:users) do
#   add_column :job, "varchar(80)", {:null=>true}
#   add_index [:name], {:unique=>false}
# end