Mongoid Slug generates a URL slug or permalink based on one or more fields in a Mongoid model. It sits idly on top of stringex, supporting non-Latin characters.
Add to your Gemfile:
gem 'mongoid-slug'
class Book
include Mongoid::Document
include Mongoid::Slug
field :title
slug :title
end
# GET /books/a-thousand-plateaus
book = Book.find params[:book_id]
Mongoid Slug will attempt to determine whether you want to find using the slugs
field or the _id
field by inspecting the supplied parameters.
- Mongoid Slug will perform a find based on
slugs
only if all arguments passed tofind
are of the typeString
. - If your document uses
BSON::ObjectId
identifiers, and all arguments look like validBSON::ObjectId
, then Mongoid Slug will perform a find based on_id
. - If your document uses any other type of identifiers, and all arguments passed to
find
are of the same type, then Mongoid Slug will perform a find based on_id
. - If your document uses
String
identifiers and you want to be able find by slugs or ids, to get the correct behaviour, you should add aslug_id_strategy
option to your_id
field definition. This option should return something that responds tocall
(a callable) and takes one string argument, e.g. a lambda. This callable must return true if the string looks like one of your ids.
Book.fields['_id'].type
=> String
book = Book.find 'a-thousand-plateaus' # Finds by slugs
=> ...
class Post
include Mongoid::Document
include Mongoid::Slug
field :_id, type: String, slug_id_strategy: lambda { |id| id.start_with?('...') }
field :name
slug :name, history: true
end
Post.fields['_id'].type
=> String
post = Post.find 'a-thousand-plateaus' # Finds by slugs
=> ...
post = Post.find '50b1386a0482939864000001' # Finds by bson ids
=> ...
Examine slug.rb for all available options.
To set slugs for existing records run following rake task:
rake mongoid_slug:set
You can pass model names as an option for which you want to set slugs:
rake mongoid_slug:set[Model1,Model2]
Empty slugs are possible and generate a nil
value for the _slugs
field. In the Post
example above, a blank post name
will cause the document record not to contain a _slugs
field in the database. The default _slugs
index is sparse
, allowing that. If you wish to change this behavior add a custom validates_presence_of :_slugs
validator to the document or change the database index to sparse: false
.
By default Mongoid Slug generates slugs with stringex. If this is not desired you can define your own slug generator.
There are two ways to define slug generator.
Configure a block in config/initializers/mongoid_slug.rb
as follows:
Mongoid::Slug.configure do |c|
# create a block that takes the current object as an argument and return the slug
c.slug = proc { |cur_obj|
cur_object.slug_builder.to_url
}
end
class Caption
include Mongoid::Document
include Mongoid::Slug
# create a block that takes the current object as an argument and returns the slug
slug do |cur_object|
cur_object.slug_builder.to_url
end
end
The to_url
method comes from stringex.
You can define a slug builder globally and/or override it per model.
To scope a slug by a reference association, pass :scope
:
class Company
include Mongoid::Document
references_many :employees
end
class Employee
include Mongoid::Document
include Mongoid::Slug
field :name
referenced_in :company
slug :name, scope: :company
end
In this example, if you create an employee without associating it with any company, the scope will fall back to the root employees collection.
Currently, if you have an irregular association name, you must specify the :inverse_of
option on the other side of the assocation.
Embedded objects are automatically scoped by their parent.
Note that the unique index on the Employee
collection in this example is derived from the scope
value and is { _slugs: 1, company_id: 1}
. Therefore :company
must be referenced_in
above the definition of slug
or it will not be able to resolve the association and mistakenly create a { _slugs: 1, company: 1}
index. An alternative is to scope to the field itself as follows:
class Employee
include Mongoid::Document
include Mongoid::Slug
field :name
field :company_id
slug :name, scope: :company_id
end
MongoDB has a default limit around 1KB to the size of the index keys and will raise error 17280, key too large to index
when trying to create a record that causes an index key to exceed that limit. By default slugs are of the form text[-number]
and the text portion is limited in size to Mongoid::Slug::MONGO_INDEX_KEY_LIMIT_BYTES - 32
bytes. You can change this limit with max_length
or set it to nil
if you're running MongoDB with failIndexKeyTooLong set to false
.
class Company
include Mongoid::Document
include Mongoid::Slug
field :name
slug :name, max_length: 24
end
By default when using STI, the scope will be around the super-class.
class Book
include Mongoid::Document
include Mongoid::Slug
field :title
slug :title, history: true
embeds_many :subjects
has_many :authors
end
class ComicBook < Book
end
book = Book.create(title: 'Anti Oedipus')
comic_book = ComicBook.create(title: 'Anti Oedipus')
comic_book.slugs.should_not eql(book.slugs)
If you want the scope to be around the subclass, then set the option by_model_type: true
.
class Book
include Mongoid::Document
include Mongoid::Slug
field :title
slug :title, history: true, by_model_type: true
embeds_many :subjects
has_many :authors
end
class ComicBook < Book
end
book = Book.create(title: 'Anti Oedipus')
comic_book = ComicBook.create(title: 'Anti Oedipus')
comic_book.slugs.should eql(book.slugs)
Enable slug history tracking by setting history: true
.
class Page
include Mongoid::Document
include Mongoid::Slug
field :title
slug :title, history: true
end
The document will then be returned for any of the saved slugs:
page = Page.new title: "Home"
page.save
page.update_attributes title: "Welcome"
Page.find("welcome") == Page.find("home") # => true
Pass words you do not want to be slugged using the reserve
option:
class Friend
include Mongoid::Document
field :name
slug :name, reserve: ['admin', 'root']
end
friend = Friend.create name: 'admin'
Friend.find('admin') # => nil
friend.slug # => 'admin-1'
When reserved words are not specified, the words 'new' and 'edit' are considered reserved by default. Specifying an array of custom reserved words will overwrite these defaults.
The slugs can be localized:
class PageSlugLocalize
include Mongoid::Document
include Mongoid::Slug
field :title, localize: true
slug :title, localize: true
end
This feature is built upon Mongoid localized fields, so fallbacks and localization works as documented in the Mongoid manual.
By default find will search for the document by the id field if the provided id looks like a BSON::ObjectId
, and it will otherwise find by the _slugs field. However, custom strategies can ovveride the default behavior, like e.g:
module Mongoid::Slug::UuidIdStrategy
def self.call id
id =~ /\A([0-9a-fA-F]){8}-(([0-9a-fA-F]){4}-){3}([0-9a-fA-F]){12}\z/
end
end
Use a custom strategy by adding the slug_id_strategy
annotation to the _id
field:
class Entity
include Mongoid::Document
include Mongoid::Slug
field :_id, type: String, slug_id_strategy: UuidIdStrategy
field :user_edited_variation
slug :user_edited_variation, history: true
end
Lets say you want to have a auto-suggest function on your GUI that could provide a preview of what the url or slug could be before the form to create the record was submitted.
You can use the UniqueSlug class in your server side code to do this, e.g.
title = params[:title]
unique = Mongoid::Slug::UniqueSlug.new(Book.new).find_unique(title)
...
# return some representation of unique
Mongoid-slug is work of many of contributors. You're encouraged to submit pull requests, propose features, ask questions and discuss issues. See CONTRIBUTING for details.
Copyright (c) 2010-2017 Hakan Ensari & Contributors, see LICENSE for details.