0.03
Repository is archived
No commit activity in last 3 years
No release in over 3 years
Sequences need some love. pg_sequencer teaches Rails what sequences are, and will dump them to schema.rb, and also lets you create/drop sequences in migrations.
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
 Dependencies

Development

Runtime

>= 3.0.0
 Project Readme

NOTICE: Project is no longer maintained¶ ↑

As of November 7th, 2018, Code42 has chosen to no longer actively host this project. The project was archive at this point. All issues have been closed and pull requests rejected. While archived, the source is still available, but this project may be permanantly removed from GitHub in the near future.

pg_sequencer¶ ↑

pg_sequencer adds methods to your migrations to allow you to create, drop and change sequence objects in PostgreSQL. It also dumps sequences to schema.rb.

This is especially useful if you are connecting to a legacy database where the primary key field is declared as an INTEGER and a sequence is queried for the value of the next record.

The design of pg_sequencer is heavily influenced on Matthew Higgins’ Foreigner gem:

Installation¶ ↑

Add this to your Gemfile:

gem 'pg_sequencer'

API¶ ↑

pg_sequencer adds the following methods to migrations:

  • create_sequence(sequence_name, options)

  • change_sequence(sequence_name, options)

  • drop_sequence(sequence_name)

The methods closely mimic the syntax of the PostgreSQL SQL for CREATE SEQUENCE, DROP SEQUENCE and ALTER SEQUENCE. See the REFERENCES section below for more information.

Options¶ ↑

For create_sequence and change_sequence, all options are the same, except create_sequence will look for :start or :start_with, and change_sequence will look for :restart or :restart_with.

  • :increment/:increment_by (integer) - The value to increment the sequence by.

  • :min (integer/false) - The minimum value of the sequence. If specified as false (e.g. :min => false), “NO MINVALUE” is sent to Postgres.

  • :max (integer/false) - The maximum value of the sequence. May be specified as “:max => false” to generate “NO MAXVALUE”

  • :start/:start_with (integer) - The starting value of the sequence (create_sequence only)

  • :restart/:restart_with (integer) The value to restart the sequence with (change_sequence only)

  • :cache (integer) - The number of values the sequence should cache.

  • :cycle (boolean) - Whether the sequence should cycle. Generated at “CYCLE” or “NO CYCLE”

Examples¶ ↑

Creating a sequence¶ ↑

Create a sequence called “seq_user”, incrementing by 1, min of 1, max of 2000000, starts at 1, caches 10 values, and disallows cycles:

create_sequence("seq_user", {
  :increment => 1,
  :min => 1,
  :max => 2000000,
  :start => 1,
  :cache => 10,
  :cycle => false
})

This is equivalent of the following query:

CREATE SEQUENCE seq_user INCREMENT BY 1 MIN 1 MAX 2000000 START 1 CACHE 10 NO CYCLE

Reset a sequence’s value:¶ ↑

change_sequence "seq_accounts", :restart_with => 50

This is equivalent to:

ALTER SEQUENCE seq_accounts RESTART WITH 50

Removing a sequence:¶ ↑

drop_sequence "seq_products"

Caveats / Bugs¶ ↑

  • Tested with postgres 9.0.4, should work down to 8.1.

  • Listing all the sequences in a database creates n+1 queries (1 to get the names and n to describe each sequence). Is there a way to fully describe all sequences in a database in one query?

  • The “SET SCHEMA” fragment of the ALTER command is not implemented.

  • Oracle/other databases not supported

  • Other unknown bugs :)

References¶ ↑