• This repository has been archived on 20/Nov/2023
  • Stars
    star
    135
  • Rank 269,297 (Top 6 %)
  • Language
    CoffeeScript
  • License
    MIT License
  • Created over 13 years ago
  • Updated over 8 years ago

Reviews

There are no reviews yet. Be the first to send feedback to the community and the maintainers!

Repository Details

Arel ported to node js β€” with some changes

Rel

https://github.com/yang/rel

Description

Rel is a SQL AST manager for Node JS. It is a straight port of https://github.com/rails/arel. Although it does have some changes of note. These are:

  1. No reliance on a database connection. This library builds queries only.
  2. It obviously can't use all of the ruby-isms like over-riding the array operator so methods are used instead.

It still holds the same goals as Arel which are:

  1. Simplifies the generation complex of SQL queries.
  2. Adapts to various RDBMS systems

Before you ask, there will also be a port of ActiveRecord coming in the next little bit.

Installation

npm install rel

Unfortunately the package is currently purely Coffeescript, with no accompanying JS compilation in the git repo. To use this with a JS project, you can compile the files in-place with:

npm run build

This is a bandaid for the time being.

Introduction

users = new Rel.Table 'users'
users.project(Rel.star()).toSql()

Will produce

SELECT * FROM users

A more complicated example of command queries is:

users.where(users.column('name').eq('amy'))
# => SELECT * FROM users WHERE users.name = 'amy'

In SQL the selection would contain what you are getting from the database, this is called a projection in Rel.

users.project(users.column('id')) # => SELECT users.id FROM users

Joins resemble SQL:

users.join(photos).on(users.column('id').eq(photos.column('user_id')))
# => SELECT * FROM users INNER JOIN photos ON users.id = photos.user_id

Limit and offset and called take and skip:

users.take(5) # => SELECT * FROM users LIMIT 5
users.skip(4) # => SELECT * FROM users OFFSET 4

GROUP BY is called group:

users.group(users.column('name')) # => SELECT * FROM users GROUP BY name

You can chain all operators, for example:

users.where(users.column('name').eq('amy')).project(users.column('id'))
# => SELECT users.id FROM users WHERE users.name = 'amy'

Another example:

users.where(users.column('name').eq('bob')).where(users.column('age').lt(25))

You can also pass in multiple arguments:

users.where(users.column('name').eq('bob'), users.column('age').lt(25))

The OR operator works like this:

users.where(users.column('name').eq('bob').or(users.column('age').lt(25)))

This is the same as the AS operator.

Contributors

Founded and majority of development to date by Carl Woodward (@carlwoodward).

Current developer and maintainer is Yang Zhang (@yang).

Development

To run tests, run npm test.

Setup notes