Datatables is a nifty jquery plugin that adds the ability to paginate, sort, and search your html tables. When dealing with large tables (more than a couple hundred rows) however, we run into performance issues. These can be fixed by using server-side pagination, but this breaks some datatables functionality.
ajax-datatables-rails is a wrapper around datatable's ajax methods that allow synchronization with server-side pagination in a rails app. It was inspired by this railscast. I needed to implement a similar solution in a couple projects I was working on so I extracted it out into a gem.
Add these lines to your application's Gemfile:
gem 'jquery-datatables-rails'
gem 'ajax-datatables-rails'
And then execute:
$ bundle
The following examples assume that we are setting up ajax-datatables-rails for an index of users from a User model
Run the following command:
$ rails generate ajaxdatatable User
This will generate a file named users_datatable.rb in app/datatables. Open the file and customize in the functions as directed by the comments
def initialize(view)
@model_name = User
@columns = # insert array of column names here
@searchable_columns = #insert array of columns that will be searched
super(view)
end-
For
@columns, assign an array of the database columns that correspond to the columns in our view table. For example[users.f_name, users.l_name, users.bio]. This array is used for sorting by various columns -
For
@searchable_columns, assign an array of the database columns that you want searchable by datatables. For example[users.f_name, users.l_name]
This gives us:
def initialize(view)
@model_name = User
@columns = [users.f_name, users.l_name, users.bio]
@searchable_columns = [users.f_name, users.l_name]
super(view)
enddef data
users.map do |user|
[
# comma separated list of the values for each cell of a table row
]
end
endThis method builds a 2d array that is used by datatables to construct the html table. Insert the values you want on each column.
def data
users.map do |user|
[
user.f_name,
user.l_name,
user.bio
]
end
enddef get_raw_records
# insert query here
endThis is where your query goes. For instance, if you want to show all users:
def get_raw_records
User
endDo not put User.all as this will convert to an array, and give you an error
because the offset and limit methods are not defined for an array.
Set up the controller to respond to JSON
def index
respond_to do |format|
format.html
format.json { render json: UsersDatatable.new(view_context) }
end
end- Set up an html
<table>with a<thead>and<tbody> - Add in your table headers if desired
- Don't add any rows to the body of the table, datatables does this automatically
- Add a data attribute to the
<table>tag with the url of the JSON feed
The resulting view may look like this:
<table id="user-table", data-source="<%= users_path(format: :json) %>">
<thead>
<tr>
<th>First Name</th>
<th>Last Name</th>
<th>Brief Bio</th>
</tr>
</thead>
<tbody>
</tbody>
</table>Finally, the javascript to tie this all together. In the appropriate js.coffee file:
$ ->
$('#users-table').dataTable
bProcessing: true
bServerSide: true
sAjaxSource: $('#users-table').data('source')- Fork it
- Create your feature branch (
git checkout -b my-new-feature) - Commit your changes (
git commit -am 'Added some feature') - Push to the branch (
git push origin my-new-feature) - Create new Pull Request