Module: ActionView::Helpers::PrototypeHelper

Included in:
ActionView::Helpers, JavaScriptHelper
Defined in:
lib/action_view/helpers/prototype_helper.rb

Overview

Prototype is a JavaScript library that provides DOM manipulation, Ajax functionality, and more traditional object-oriented facilities for JavaScript. This module provides a set of helpers to make it more convenient to call functions from Prototype using Rails, including functionality to call remote Rails methods (that is, making a background request to a Rails action) using Ajax. This means that you can call actions in your controllers without reloading the page, but still update certain parts of it using injections into the DOM. A common use case is having a form that adds a new element to a list without reloading the page or updating a shopping cart total when a new item is added.

Usage

To be able to use these helpers, you must first include the Prototype JavaScript framework in your pages.

javascript_include_tag 'prototype'

(See the documentation for ActionView::Helpers::JavaScriptHelper for more information on including this and other JavaScript files in your Rails templates.)

Now you’re ready to call a remote action either through a link…

link_to_remote "Add to cart",
  :url => { :action => "add", :id => product.id },
  :update => { :success => "cart", :failure => "error" }

…through a form…

<% form_remote_tag :url => '/shipping' do -%>
  <div><%= submit_tag 'Recalculate Shipping' %></div>
<% end -%>

…periodically…

periodically_call_remote(:url => 'update', :frequency => '5', :update => 'ticker')

…or through an observer (i.e., a form or field that is observed and calls a remote action when changed).

<%= observe_field(:searchbox,
     :url => { :action => :live_search }),
     :frequency => 0.5,
     :update => :hits,
     :with => 'query'
     %>

As you can see, there are numerous ways to use Prototype’s Ajax functions (and actually more than are listed here); check out the documentation for each method to find out more about its usage and options.

Common Options

See link_to_remote for documentation of options common to all Ajax helpers; any of the options specified by link_to_remote can be used by the other helpers.

Designing your Rails actions for Ajax

When building your action handlers (that is, the Rails actions that receive your background requests), it’s important to remember a few things. First, whatever your action would normally return to the browser, it will return to the Ajax call. As such, you typically don’t want to render with a layout. This call will cause the layout to be transmitted back to your page, and, if you have a full HTML/CSS, will likely mess a lot of things up. You can turn the layout off on particular actions by doing the following:

class SiteController < ActionController::Base
  layout "standard", :except => [:ajax_method, :more_ajax, :another_ajax]
end

Optionally, you could do this in the method you wish to lack a layout:

render :layout => false

You can tell the type of request from within your action using the request.xhr? (XmlHttpRequest, the method that Ajax uses to make background requests) method.

def name
  # Is this an XmlHttpRequest request?
  if (request.xhr?)
    render :text => @name.to_s
  else
    # No?  Then render an action.
    render :action => 'view_attribute', :attr => @name
  end
end

The else clause can be left off and the current action will render with full layout and template. An extension to this solution was posted to Ryan Heneise’s blog at ArtOfMission.

layout proc{ |c| c.request.xhr? ? false : "application" }

Dropping this in your ApplicationController turns the layout off for every request that is an “xhr” request.

If you are just returning a little data or don’t want to build a template for your output, you may opt to simply render text output, like this:

render :text => 'Return this from my method!'

Since whatever the method returns is injected into the DOM, this will simply inject some text (or HTML, if you tell it to). This is usually how small updates, such updating a cart total or a file count, are handled.

Updating multiple elements

See JavaScriptGenerator for information on updating multiple elements on the page in an Ajax response.

Defined Under Namespace

Classes: JavaScriptGenerator

Constant Summary collapse

CALLBACKS =
Set.new([ :create, :uninitialized, :loading, :loaded,
:interactive, :complete, :failure, :success ] +
(100..599).to_a)
AJAX_OPTIONS =
Set.new([ :before, :after, :condition, :url,
:asynchronous, :method, :insertion, :position,
:form, :with, :update, :script, :type ]).merge(CALLBACKS)

Instance Method Summary collapse

Instance Method Details

#button_to_remote(name, options = {}, html_options = {}) ⇒ Object

Creates a button with an onclick event which calls a remote action via XMLHttpRequest The options for specifying the target with :url and defining callbacks is the same as link_to_remote.



263
264
265
# File 'lib/action_view/helpers/prototype_helper.rb', line 263

def button_to_remote(name, options = {}, html_options = {})
  button_to_function(name, remote_function(options), html_options)
end

#evaluate_remote_responseObject

Returns ‘eval(request.responseText)’ which is the JavaScript function that form_remote_tag can call in :complete to evaluate a multiple update return document using update_element_function calls.



433
434
435
# File 'lib/action_view/helpers/prototype_helper.rb', line 433

def evaluate_remote_response
  "eval(request.responseText)"
end

#form_remote_tag(options = {}, &block) ⇒ Object

Returns a form tag that will submit using XMLHttpRequest in the background instead of the regular reloading POST arrangement. Even though it’s using JavaScript to serialize the form elements, the form submission will work just like a regular submission as viewed by the receiving side (all elements available in params). The options for specifying the target with :url and defining callbacks is the same as link_to_remote.

A “fall-through” target for browsers that doesn’t do JavaScript can be specified with the :action/:method options on :html.

Example:

# Generates:
#      <form action="/some/place" method="post" onsubmit="new Ajax.Request('',
#      {asynchronous:true, evalScripts:true, parameters:Form.serialize(this)}); return false;">
form_remote_tag :html => { :action =>
  url_for(:controller => "some", :action => "place") }

The Hash passed to the :html key is equivalent to the options (2nd) argument in the FormTagHelper.form_tag method.

By default the fall-through action is the same as the one specified in the :url (and the default method is :post).

form_remote_tag also takes a block, like form_tag:

# Generates:
#     <form action="/" method="post" onsubmit="new Ajax.Request('/',
#     {asynchronous:true, evalScripts:true, parameters:Form.serialize(this)});
#     return false;"> <div><input name="commit" type="submit" value="Save" /></div>
#     </form>
<% form_remote_tag :url => '/posts' do -%>
  <div><%= submit_tag 'Save' %></div>
<% end -%>


331
332
333
334
335
336
337
338
339
340
# File 'lib/action_view/helpers/prototype_helper.rb', line 331

def form_remote_tag(options = {}, &block)
  options[:form] = true

  options[:html] ||= {}
  options[:html][:onsubmit] =
    (options[:html][:onsubmit] ? options[:html][:onsubmit] + "; " : "") +
    "#{remote_function(options)}; return false;"

  form_tag(options[:html].delete(:action) || url_for(options[:url]), options[:html], &block)
end

Returns a link to a remote action defined by options[:url] (using the url_for format) that’s called in the background using XMLHttpRequest. The result of that request can then be inserted into a DOM object whose id can be specified with options[:update]. Usually, the result would be a partial prepared by the controller with render :partial.

Examples:

# Generates: <a href="#" onclick="new Ajax.Updater('posts', '/blog/destroy/3', {asynchronous:true, evalScripts:true});
#            return false;">Delete this post</a>
link_to_remote "Delete this post", :update => "posts",
  :url => { :action => "destroy", :id => post.id }

# Generates: <a href="#" onclick="new Ajax.Updater('emails', '/mail/list_emails', {asynchronous:true, evalScripts:true});
#            return false;"><img alt="Refresh" src="/images/refresh.png?" /></a>
link_to_remote(image_tag("refresh"), :update => "emails",
  :url => { :action => "list_emails" })

You can override the generated HTML options by specifying a hash in options[:html].

link_to_remote "Delete this post", :update => "posts",
  :url  => post_url(@post), :method => :delete,
  :html => { :class  => "destructive" }

You can also specify a hash for options[:update] to allow for easy redirection of output to an other DOM element if a server-side error occurs:

Example:

# Generates: <a href="#" onclick="new Ajax.Updater({success:'posts',failure:'error'}, '/blog/destroy/5',
#            {asynchronous:true, evalScripts:true}); return false;">Delete this post</a>
link_to_remote "Delete this post",
  :url => { :action => "destroy", :id => post.id },
  :update => { :success => "posts", :failure => "error" }

Optionally, you can use the options[:position] parameter to influence how the target DOM element is updated. It must be one of :before, :top, :bottom, or :after.

The method used is by default POST. You can also specify GET or you can simulate PUT or DELETE over POST. All specified with options[:method]

Example:

# Generates: <a href="#" onclick="new Ajax.Request('/person/4', {asynchronous:true, evalScripts:true, method:'delete'});
#            return false;">Destroy</a>
link_to_remote "Destroy", :url => person_url(:id => person), :method => :delete

By default, these remote requests are processed asynchronous during which various JavaScript callbacks can be triggered (for progress indicators and the likes). All callbacks get access to the request object, which holds the underlying XMLHttpRequest.

To access the server response, use request.responseText, to find out the HTTP status, use request.status.

Example:

# Generates: <a href="#" onclick="new Ajax.Request('/words/undo?n=33', {asynchronous:true, evalScripts:true,
#            onComplete:function(request){undoRequestCompleted(request)}}); return false;">hello</a>
word = 'hello'
link_to_remote word,
  :url => { :action => "undo", :n => word_counter },
  :complete => "undoRequestCompleted(request)"

The callbacks that may be specified are (in order):

:loading

Called when the remote document is being loaded with data by the browser.

:loaded

Called when the browser has finished loading the remote document.

:interactive

Called when the user can interact with the remote document, even though it has not finished loading.

:success

Called when the XMLHttpRequest is completed, and the HTTP status code is in the 2XX range.

:failure

Called when the XMLHttpRequest is completed, and the HTTP status code is not in the 2XX range.

:complete

Called when the XMLHttpRequest is complete (fires after success/failure if they are present).

You can further refine :success and :failure by adding additional callbacks for specific status codes.

Example:

# Generates: <a href="#" onclick="new Ajax.Request('/testing/action', {asynchronous:true, evalScripts:true,
#            on404:function(request){alert('Not found...? Wrong URL...?')},
#            onFailure:function(request){alert('HTTP Error ' + request.status + '!')}}); return false;">hello</a>
link_to_remote word,
  :url => { :action => "action" },
  404 => "alert('Not found...? Wrong URL...?')",
  :failure => "alert('HTTP Error ' + request.status + '!')"

A status code callback overrides the success/failure handlers if present.

If you for some reason or another need synchronous processing (that’ll block the browser while the request is happening), you can specify options[:type] = :synchronous.

You can customize further browser side call logic by passing in JavaScript code snippets via some optional parameters. In their order of use these are:

:confirm

Adds confirmation dialog.

:condition

Perform remote request conditionally by this expression. Use this to describe browser-side conditions when request should not be initiated.

:before

Called before request is initiated.

:after

Called immediately after request was initiated and before :loading.

:submit

Specifies the DOM element ID that’s used as the parent of the form elements. By default this is the current form, but it could just as well be the ID of a table row or any other DOM element.

:with

A JavaScript expression specifying the parameters for the XMLHttpRequest. Any expressions should return a valid URL query string.

Example:

:with => "'name=' + $('name').value"

You can generate a link that uses AJAX in the general case, while degrading gracefully to plain link behavior in the absence of JavaScript by setting html_options[:href] to an alternate URL. Note the extra curly braces around the options hash separate it as the second parameter from html_options, the third.

Example:

link_to_remote "Delete this post",
  { :update => "posts", :url => { :action => "destroy", :id => post.id } },
  :href => url_for(:action => "destroy", :id => post.id)


255
256
257
# File 'lib/action_view/helpers/prototype_helper.rb', line 255

def link_to_remote(name, options = {}, html_options = nil)
  link_to_function(name, remote_function(options), html_options || options.delete(:html))
end

#observe_field(field_id, options = {}) ⇒ Object

Observes the field with the DOM ID specified by field_id and calls a callback when its contents have changed. The default callback is an Ajax call. By default the value of the observed field is sent as a parameter with the Ajax call.

Example:

# Generates: new Form.Element.Observer('suggest', 0.25, function(element, value) {new Ajax.Updater('suggest',
#         '/testing/find_suggestion', {asynchronous:true, evalScripts:true, parameters:'q=' + value})})
<%= observe_field :suggest, :url => { :action => :find_suggestion },
     :frequency => 0.25,
     :update => :suggest,
     :with => 'q'
     %>

Required options are either of:

:url

url_for-style options for the action to call when the field has changed.

:function

Instead of making a remote call to a URL, you can specify javascript code to be called instead. Note that the value of this option is used as the body of the javascript function, a function definition with parameters named element and value will be generated for you for example:

observe_field("glass", :frequency => 1, :function => "alert('Element changed')")

will generate:

new Form.Element.Observer('glass', 1, function(element, value) {alert('Element changed')})

The element parameter is the DOM element being observed, and the value is its value at the time the observer is triggered.

Additional options are:

:frequency

The frequency (in seconds) at which changes to this field will be detected. Not setting this option at all or to a value equal to or less than zero will use event based observation instead of time based observation.

:update

Specifies the DOM ID of the element whose innerHTML should be updated with the XMLHttpRequest response text.

:with

A JavaScript expression specifying the parameters for the XMLHttpRequest. The default is to send the key and value of the observed field. Any custom expressions should return a valid URL query string. The value of the field is stored in the JavaScript variable value.

Examples

:with => "'my_custom_key=' + value"
:with => "'person[name]=' + prompt('New name')"
:with => "Form.Element.serialize('other-field')"

Finally

:with => 'name'

is shorthand for

:with => "'name=' + value"

This essentially just changes the key of the parameter.

Additionally, you may specify any of the options documented in the Common options section at the top of this document.

Example:

# Sends params: {:title => 'Title of the book'} when the book_title input
# field is changed.
observe_field 'book_title',
  :url => 'http://example.com/books/edit/1',
  :with => 'title'


547
548
549
550
551
552
553
# File 'lib/action_view/helpers/prototype_helper.rb', line 547

def observe_field(field_id, options = {})
  if options[:frequency] && options[:frequency] > 0
    build_observer('Form.Element.Observer', field_id, options)
  else
    build_observer('Form.Element.EventObserver', field_id, options)
  end
end

#observe_form(form_id, options = {}) ⇒ Object

Observes the form with the DOM ID specified by form_id and calls a callback when its contents have changed. The default callback is an Ajax call. By default all fields of the observed field are sent as parameters with the Ajax call.

The options for observe_form are the same as the options for observe_field. The JavaScript variable value available to the :with option is set to the serialized form by default.



563
564
565
566
567
568
569
# File 'lib/action_view/helpers/prototype_helper.rb', line 563

def observe_form(form_id, options = {})
  if options[:frequency]
    build_observer('Form.Observer', form_id, options)
  else
    build_observer('Form.EventObserver', form_id, options)
  end
end

#periodically_call_remote(options = {}) ⇒ Object

Periodically calls the specified url (options[:url]) every options[:frequency] seconds (default is 10). Usually used to update a specified div (options[:update]) with the results of the remote call. The options for specifying the target with :url and defining callbacks is the same as link_to_remote. Examples:

# Call get_averages and put its results in 'avg' every 10 seconds
# Generates:
#      new PeriodicalExecuter(function() {new Ajax.Updater('avg', '/grades/get_averages',
#      {asynchronous:true, evalScripts:true})}, 10)
periodically_call_remote(:url => { :action => 'get_averages' }, :update => 'avg')

# Call invoice every 10 seconds with the id of the customer
# If it succeeds, update the invoice DIV; if it fails, update the error DIV
# Generates:
#      new PeriodicalExecuter(function() {new Ajax.Updater({success:'invoice',failure:'error'},
#      '/testing/invoice/16', {asynchronous:true, evalScripts:true})}, 10)
periodically_call_remote(:url => { :action => 'invoice', :id => customer.id },
   :update => { :success => "invoice", :failure => "error" }

# Call update every 20 seconds and update the new_block DIV
# Generates:
# new PeriodicalExecuter(function() {new Ajax.Updater('news_block', 'update', {asynchronous:true, evalScripts:true})}, 20)
periodically_call_remote(:url => 'update', :frequency => '20', :update => 'news_block')


292
293
294
295
296
# File 'lib/action_view/helpers/prototype_helper.rb', line 292

def periodically_call_remote(options = {})
   frequency = options[:frequency] || 10 # every ten seconds by default
   code = "new PeriodicalExecuter(function() {#{remote_function(options)}}, #{frequency})"
   javascript_tag(code)
end

#remote_form_for(record_or_name_or_array, *args, &proc) ⇒ Object Also known as: form_remote_for

Creates a form that will submit using XMLHttpRequest in the background instead of the regular reloading POST arrangement and a scope around a specific resource that is used as a base for questioning about values for the fields.

Resource

Example:

<% remote_form_for(@post) do |f| %>
  ...
<% end %>

This will expand to be the same as:

<% remote_form_for :post, @post, :url => post_path(@post), :html => { :method => :put, :class => "edit_post", :id => "edit_post_45" } do |f| %>
  ...
<% end %>

Nested Resource

Example:

<% remote_form_for([@post, @comment]) do |f| %>
  ...
<% end %>

This will expand to be the same as:

<% remote_form_for :comment, @comment, :url => post_comment_path(@post, @comment), :html => { :method => :put, :class => "edit_comment", :id => "edit_comment_45" } do |f| %>
  ...
<% end %>

If you don’t need to attach a form to a resource, then check out form_remote_tag.

See FormHelper#form_for for additional semantics.



376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
# File 'lib/action_view/helpers/prototype_helper.rb', line 376

def remote_form_for(record_or_name_or_array, *args, &proc)
  options = args.extract_options!

  case record_or_name_or_array
  when String, Symbol
    object_name = record_or_name_or_array
  when Array
    object = record_or_name_or_array.last
    object_name = ActionController::RecordIdentifier.singular_class_name(object)
    apply_form_for_options!(record_or_name_or_array, options)
    args.unshift object
  else
    object      = record_or_name_or_array
    object_name = ActionController::RecordIdentifier.singular_class_name(record_or_name_or_array)
    apply_form_for_options!(object, options)
    args.unshift object
  end

  concat(form_remote_tag(options))
  fields_for(object_name, *(args << options), &proc)
  concat('</form>'.html_safe)
end

#remote_function(options) ⇒ Object

Returns the JavaScript needed for a remote function. Takes the same arguments as link_to_remote.

Example:

# Generates: <select id="options" onchange="new Ajax.Updater('options',
# '/testing/update_options', {asynchronous:true, evalScripts:true})">
<select id="options" onchange="<%= remote_function(:update => "options",
    :url => { :action => :update_options }) %>">
  <option value="0">Hello</option>
  <option value="1">World</option>
</select>


448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
# File 'lib/action_view/helpers/prototype_helper.rb', line 448

def remote_function(options)
  javascript_options = options_for_ajax(options)

  update = ''
  if options[:update] && options[:update].is_a?(Hash)
    update  = []
    update << "success:'#{options[:update][:success]}'" if options[:update][:success]
    update << "failure:'#{options[:update][:failure]}'" if options[:update][:failure]
    update  = '{' + update.join(',') + '}'
  elsif options[:update]
    update << "'#{options[:update]}'"
  end

  function = update.empty? ?
    "new Ajax.Request(" :
    "new Ajax.Updater(#{update}, "

  url_options = options[:url]
  url_options = url_options.merge(:escape => false) if url_options.is_a?(Hash)
  function << "'#{escape_javascript(url_for(url_options))}'"
  function << ", #{javascript_options})"

  function = "#{options[:before]}; #{function}" if options[:before]
  function = "#{function}; #{options[:after]}"  if options[:after]
  function = "if (#{options[:condition]}) { #{function}; }" if options[:condition]
  function = "if (confirm('#{escape_javascript(options[:confirm])}')) { #{function}; }" if options[:confirm]

  return function
end

#submit_to_remote(name, value, options = {}) ⇒ Object

Returns a button input tag with the element name of name and a value (i.e., display text) of value that will submit form using XMLHttpRequest in the background instead of a regular POST request that reloads the page.

# Create a button that submits to the create action
#
# Generates: <input name="create_btn" onclick="new Ajax.Request('/testing/create',
#     {asynchronous:true, evalScripts:true, parameters:Form.serialize(this.form)});
#     return false;" type="button" value="Create" />
<%= submit_to_remote 'create_btn', 'Create', :url => { :action => 'create' } %>

# Submit to the remote action update and update the DIV succeed or fail based
# on the success or failure of the request
#
# Generates: <input name="update_btn" onclick="new Ajax.Updater({success:'succeed',failure:'fail'},
#      '/testing/update', {asynchronous:true, evalScripts:true, parameters:Form.serialize(this.form)});
#      return false;" type="button" value="Update" />
<%= submit_to_remote 'update_btn', 'Update', :url => { :action => 'update' },
   :update => { :success => "succeed", :failure => "fail" }

options argument is the same as in form_remote_tag.



421
422
423
424
425
426
427
428
# File 'lib/action_view/helpers/prototype_helper.rb', line 421

def submit_to_remote(name, value, options = {})
  options[:with] ||= 'Form.serialize(this.form)'

  html_options = options.delete(:html) || {}
  html_options[:name] = name

  button_to_remote(value, options, html_options)
end

#update_page(&block) ⇒ Object

Yields a JavaScriptGenerator and returns the generated JavaScript code. Use this to update multiple elements on a page in an Ajax response. See JavaScriptGenerator for more information.

Example:

update_page do |page|
  page.hide 'spinner'
end


1028
1029
1030
# File 'lib/action_view/helpers/prototype_helper.rb', line 1028

def update_page(&block)
  JavaScriptGenerator.new(@template, &block).to_s.html_safe
end

#update_page_tag(html_options = {}, &block) ⇒ Object

Works like update_page but wraps the generated JavaScript in a <script> tag. Use this to include generated JavaScript in an ERb template. See JavaScriptGenerator for more information.

html_options may be a hash of <script> attributes to be passed to ActionView::Helpers::JavaScriptHelper#javascript_tag.



1038
1039
1040
# File 'lib/action_view/helpers/prototype_helper.rb', line 1038

def update_page_tag(html_options = {}, &block)
  javascript_tag update_page(&block), html_options
end