2012-01-08 12:06:15 -05:00
# ARel [![Build Status](https://secure.travis-ci.org/rails/arel.png)](http://travis-ci.org/rails/arel) [![Dependency Status](https://gemnasium.com/rails/arel.png)](https://gemnasium.com/rails/arel)
2010-09-24 19:45:02 -04:00
* http://github.com/rails/arel
## DESCRIPTION
2008-02-23 19:49:49 -05:00
2011-04-11 12:47:47 -04:00
Arel is a SQL AST manager for Ruby. It
2011-11-10 15:02:40 -05:00
1. Simplifies the generation of complex SQL queries
2011-04-11 12:47:47 -04:00
2. Adapts to various RDBMS systems
It is intended to be a framework framework; that is, you can build your own ORM
with it, focusing on innovative object and collection modeling as opposed to
database compatibility and query generation.
2008-02-23 19:49:49 -05:00
2010-09-24 19:45:02 -04:00
## Status
2008-05-19 16:49:43 -04:00
2013-04-14 14:39:07 -04:00
For the moment, Arel uses Active Record's connection adapters to connect to the various engines, connection pooling, perform quoting, and do type conversion.
2008-05-19 16:49:43 -04:00
2010-09-24 19:45:02 -04:00
## A Gentle Introduction
2008-02-23 19:49:49 -05:00
Generating a query with ARel is simple. For example, in order to produce
SELECT * FROM users
2009-08-23 16:24:59 -04:00
2008-02-23 19:49:49 -05:00
you construct a table relation and convert it to sql:
2010-10-19 13:05:51 -04:00
users = Arel::Table.new(:users)
2011-04-27 06:49:06 -04:00
query = users.project(Arel.sql('*'))
query.to_sql
2009-08-23 16:24:59 -04:00
2010-09-24 19:45:02 -04:00
### More Sophisticated Queries
2008-05-19 16:49:43 -04:00
Here is a whirlwind tour through the most common relational operators. These will probably cover 80% of all interaction with the database.
2008-02-23 19:49:49 -05:00
2008-05-19 16:49:43 -04:00
First is the 'restriction' operator, `where` :
2008-02-23 19:49:49 -05:00
2008-05-19 16:49:43 -04:00
users.where(users[:name].eq('amy'))
# => SELECT * FROM users WHERE users.name = 'amy'
What would, in SQL, be part of the `SELECT` clause is called in Arel a `projection` :
users.project(users[:id]) # => SELECT users.id FROM users
2009-08-23 16:24:59 -04:00
2008-05-19 16:49:43 -04:00
Joins resemble SQL strongly:
users.join(photos).on(users[:id].eq(photos[:user_id]))
# => SELECT * FROM users INNER JOIN photos ON users.id = photos.user_id
What are called `LIMIT` and `OFFSET` in SQL are called `take` and `skip` in Arel:
users.take(5) # => SELECT * FROM users LIMIT 5
users.skip(4) # => SELECT * FROM users OFFSET 4
2009-08-23 16:24:59 -04:00
2008-05-19 16:49:43 -04:00
`GROUP BY` is called `group` :
2012-12-13 19:16:36 -05:00
users.project(users[:name]).group(users[:name]) # => SELECT users.name FROM users GROUP BY users.name
2008-05-19 16:49:43 -04:00
2010-03-10 21:14:49 -05:00
The best property of the Relational Algebra is its "composability", or closure under all operations. For example, to restrict AND project, just "chain" the method invocations:
2008-05-19 16:49:43 -04:00
users \
.where(users[:name].eq('amy')) \
.project(users[:id]) \
# => SELECT users.id FROM users WHERE users.name = 'amy'
All operators are chainable in this way, and they are chainable any number of times, in any order.
2008-02-23 19:49:49 -05:00
2008-05-19 16:49:43 -04:00
users.where(users[:name].eq('bob')).where(users[:age].lt(25))
2009-08-23 16:24:59 -04:00
2008-05-19 16:49:43 -04:00
Of course, many of the operators take multiple arguments, so the last example can be written more tersely:
users.where(users[:name].eq('bob'), users[:age].lt(25))
2008-02-23 19:49:49 -05:00
2010-10-19 13:05:51 -04:00
The `OR` operator works like this:
2008-05-19 16:49:43 -04:00
users.where(users[:name].eq('bob').or(users[:age].lt(25)))
2009-08-23 16:24:59 -04:00
2010-10-19 13:05:51 -04:00
The `AND` operator behaves similarly.
2008-05-20 13:11:07 -04:00
2010-09-24 19:45:02 -04:00
### The Crazy Features
2008-02-23 19:49:49 -05:00
2008-05-19 16:49:43 -04:00
The examples above are fairly simple and other libraries match or come close to matching the expressiveness of Arel (e.g., `Sequel` in Ruby).
2008-02-23 19:49:49 -05:00
2011-01-29 01:40:39 -05:00
#### Inline math operations
2012-03-12 14:05:12 -04:00
Suppose we have a table `products` with prices in different currencies. And we have a table `currency_rates` , of constantly changing currency rates. In Arel:
2011-01-29 01:40:39 -05:00
products = Arel::Table.new(:products)
products.columns # => [products[:id], products[:name], products[:price], products[:currency_id]]
currency_rates = Arel::Table.new(:currency_rates)
currency_rates.columns # => [currency_rates[:from_id], currency_rates[:to_id], currency_rates[:date], currency_rates[:rate]]
Now, to order products by price in user preferred currency simply call:
products.
join(:currency_rates).on(products[:currency_id].eq(currency_rates[:from_id])).
where(currency_rates[:to_id].eq(user_preferred_currency), currency_rates[:date].eq(Date.today)).
order(products[:price] * currency_rates[:rate])
2010-09-24 19:45:02 -04:00
#### Complex Joins
2008-02-23 19:49:49 -05:00
2008-05-19 16:49:43 -04:00
Where Arel really shines in its ability to handle complex joins and aggregations. As a first example, let's consider an "adjacency list", a tree represented in a table. Suppose we have a table `comments` , representing a threaded discussion:
2010-10-19 13:05:51 -04:00
comments = Arel::Table.new(:comments)
2009-08-23 16:24:59 -04:00
2008-05-19 16:49:43 -04:00
And this table has the following attributes:
2008-02-24 20:26:32 -05:00
2010-10-19 13:05:51 -04:00
comments.columns # => [comments[:id], comments[:body], comments[:parent_id]]
2008-02-23 19:49:49 -05:00
2008-05-19 16:49:43 -04:00
The `parent_id` column is a foreign key from the `comments` table to itself. Now, joining a table to itself requires aliasing in SQL. In fact, you may alias in Arel as well:
2008-02-23 19:49:49 -05:00
2008-05-19 16:49:43 -04:00
replies = comments.alias
comments_with_replies = \
comments.join(replies).on(replies[:parent_id].eq(comments[:id]))
# => SELECT * FROM comments INNER JOIN comments AS comments_2 WHERE comments_2.parent_id = comments.id
2009-08-23 16:24:59 -04:00
2008-05-19 16:49:43 -04:00
This will return the first comment's reply's body.