Flowdock

Recent good notes

RSS feed
May 2, 2009
8 thanks

Create a Hash from two Arrays

Here is my favorite idiom for creating a Hash from an Array of keys and an Array of values:

keys = [:a, :b]
values = [1,2]
h = Hash[*keys.zip(values).flatten]      # => {:b=>2, :a=>1}
May 2, 2009
3 thanks

Test if one array includes the elements of another

You can just use a set difference (aka minus) to see if one array includes all elements of another

not_included = [1,2,3] - (1..9).to_a
not_included      # => []

not_included = [1,2,3,'A'] - (1..9).to_a
not_included      # => ["A"]

Use intersection to test if any of the one are in the other:

shared = [1,2,3,'A'] & (1..9).to_a
shared     # => [1, 2, 3]
April 28, 2009
3 thanks

Tip: Define from_param(...) as Opposite

Often when defining a to_param method, it’s handy to introduce an opposite method for decoding them. For example:

class User < ActiveRecord::Base
  def self.from_param(param)
    find_by_name!(param)
  end

  def to_param
    name
  end
end

While you can just as easily redefine the find() method, this may be confusing since the expectation is that find() works with numerical IDs, or whatever the key column is defined as.

April 28, 2009 - (>= v2.3.2)
5 thanks

A very thorough explanation of use

Ryan Daigle has a great article about 2.3’s new nest forms which does a really good job of explaining how to use this and some of the potential gotchas. Highly recommended:

http://ryandaigle.com/articles/2009/2/1/what-s-new-in-edge-rails-nested-attributes

April 25, 2009
4 thanks

Set :use_route to nil to let Rails pick the best route

Imagine the following case. You have two landing pages, one generic one, and an account specific one. The urls are as follows:

map.landing 'landing', :controller => 'landing', :action => 'index'
map.account_landing 'accounts/:account_id/landing', :controller => 'landing', :action => 'index'

Now imagine you want a path to the landing page, using the most specific route possible. If you have an account_id, use it, if not, skip it.

You could do

url_for(:controller => 'landing', :action => 'index', :account_id => current_account)

If current_account is set you’ll get “/accounts/:account_id/landing” if not, you’ll get “/landing”. However, that just looks ugly.

Enter :use_route => nil.

landing_path(:account_id => nil)                    # => '/landing'
landing_path(:account_id => 1)                      # => '/landing?account_id=1'
landing_path(:account_id => nil, :use_route => nil) # => '/landing'
landing_path(:account_id => 1, :use_route => nil)   # => '/accounts/1/landing'

Setting :use_route to nil, is equivalent to the earlier #url_for example.

April 23, 2009
5 thanks
April 23, 2009
6 thanks

Argument Ordering

Be aware that the order of arguments for this method is the opposite of File.join:

File.expand_path('foo', '/bar')   # => "/bar/foo"
File.join('foo', '/bar')          # => "foo/bar"
April 23, 2009
5 thanks

Handy shorthand for array manipulation

You may write something like this:

>> ['a', 'b', 'c'].collect{|letter| letter.capitalize}
=> ["A", "B", "C"]

But it looks so much nicer this way:

>> ['a', 'b', 'c'].collect(&:capitalize)
=> ["A", "B", "C"]
April 21, 2009
3 thanks

Format not coming out properly?

Date, Time and DateTime may have different formats defined.

If you do:

@user.created_at.to_formatted_s(:long_ordinal)

You will get (or something):

April 16th, 2009 22:03 

But if you do:

@user.created_at.to_date.to_formatted_s(:long_ordinal)

You will get:

April 16th, 2009

So, be sure you know which one you are working with.

April 21, 2009
3 thanks

To throw an exception, use Kernel#raise

Other languages use the term throw for raising exceptions, but Ruby has a specific raise call for that.

April 21, 2009
7 thanks

Do not forget to add indexes

Don’t forget to add indexes to HATM table:

add_index :developers_projects, [:developer_id, :project_id]
April 21, 2009 - (>= v2.3.2)
5 thanks

strip_tags method not functioning in controllers, models, or libs

It comes up with an error about white_list_sanitizer undefined in the class you’re using it in. To get around this, use:

ActionController::Base.helpers.strip_tags('string')

To shorten this, add something like this in an initializer:

class String
  def strip_tags
    ActionController::Base.helpers.strip_tags(self)
  end
end

then call it with:

'string'.strip_tags
April 21, 2009 - (>= v2.3.2)
3 thanks

sanitize method not functioning in controllers, models, or libs

It comes up with an error about white_list_sanitizer undefined in the class you’re using it in. To get around this, use:

ActionController::Base.helpers.sanitize('string')

To shorten this, add something like this in an initializer:

class String
  def sanitize
    ActionController::Base.helpers.sanitize(self)
  end
end

then call it with:

'string'.sanitize
April 16, 2009
12 thanks

Parameters for Hash#inject

When running inject on a Hash, the hash is first converted to an array before being passed through.

The typical Enumerable#inject approach would be to simply capture the value:

array.inject(...) do |c, v|
end

In the case of a Hash, v is actually a key/value pair Array. That is the key is v.first and the value is v.last, however using the pair this way is awkward and can lead to confusion.

Better to simply expand the parameters in the block definition:

hash.inject(...) do |c, (k, v)|
end

Where c is the traditional carry variable and k/v represent key and value respectively.

April 16, 2009
3 thanks

Various use cases

Example

user = User.new
user.name = 'Akhil Bansal'
user.save

user =  User.new(:name => 'Akhil')
user.save

User.new do |u|
  u.name = 'Akhil'
  u.save
end
April 9, 2009
11 thanks

Define handlers in order of most generic to most specific

The later the definition of the rescue handler, the higher the priority:

rescue_from Exception, :with => :error_generic
rescue_from Exception::ComputerOnFire, :with => :panic

Declaring the Exception catch-all handler last would have the side-effect of precluding any other handlers from running.

This is what is meant by being “searched…from bottom to top”.

April 9, 2009
4 thanks

Method has moved to ActionController::Rescue::ClassMethods module

This method has simply moved, still works the same way in 2.3+

New location: ActiveSupport::Rescuable::ClassMethods#rescue_from

April 8, 2009 - (>= v2.3.2)
13 thanks

Setting child_index while using nested attributes mass assignment

When using nested attributes mass assignment sometimes you will want to add new records with javascript. You can do it with pure javascript, but if HTML is long your javascript will be long and messy and it will not be DRY as probably you already have a partial for it.

So to add a partial dynamically you can do something like that (notice string “index_to_replace_with_js”):

link_to_function

def add_object_link(name, form, object, partial, where)
  options = {:parent => true}.merge(options)
  html = render(:partial => partial, :locals => { :form => form}, :object => object)
  link_to_function name, %{
    var new_object_id = new Date().getTime() ;
    var html = jQuery(#{js html}.replace(/index_to_replace_with_js/g, new_object_id)).hide();
    html.appendTo(jQuery("#{where}")).slideDown('slow');
  }
end

js method in one of helpers (from minus mor plugin)

def js(data)
  if data.respond_to? :to_json
    data.to_json
  else
    data.inspect.to_json
  end
end

This method will generate link adding generated partial to html.

The thing that is not mentioned in docs is how to set child_index. You must add it as an argument in hash.

Example of partial

<% form.fields_for :tasks, task, :child_index => 
       (task.new_record? ? "index_to_replace_with_js" : nil) do |tasks_form| %>
  <% tasks_form.text_field :name %>
<% end %>

Using add_object_link

<% form_for :project do |form| %>
  <div id="tasks">
    <%# displaying existing tasks %>
  </div>

  <%= add_object_link("New task, form, Task.new, "task", "#tasks") %>

<% end %>

Thanks to child_index after insertion it will change indexes to current time in miliseconds so added tasks will have different names and ids.

April 8, 2009 - (v2.3.2)
6 thanks

Superclass of OrderedHash

Note that in Rails 2.3, OrderedHash changed from being a subclass of Array to a subclass of Hash. This is contrary to what the documentation says above.

April 6, 2009
5 thanks

HTML entities in options

Unfortunately everything is escaped with ERB::Util#html_escape. Your only option is either manually construct options or compeletely overwrite this method.

April 6, 2009
6 thanks

Array clustering

Sometimes you don’t want to mangle sequence of an array and just want to group adjacent values. Here’s a nice method to do so (drop it in your initializers directory or something):

module Enumerable
  # clumps adjacent elements together
  # >> [2,2,2,3,3,4,2,2,1].cluster{|x| x}
  # => [[2, 2, 2], [3, 3], [4], [2, 2], [1]]
  def cluster
    cluster = []
    each do |element|
      if cluster.last && yield(cluster.last.last) == yield(element)
        cluster.last << element
      else
        cluster << [element]
      end
    end
    cluster
  end
end

Similarly you can do the clustering on more complex items. For instance you want to cluster Documents on creation date and their type:

Document.all.cluster{|document| [document.created_on, document.type]}
April 6, 2009
3 thanks

Take care when writing regex

When you want to validate a field for a continuous string you’d probably write something like this (if it’s really early in the morning and you didn’t have your coffee yet):

validates_format_of :something => /\w/

At the first sight it looks like it’s working because something = “blahblahblah” is valid. However, so is this: something = “blah meh 55”. It’s just that your regex matched a substring of the value and not the whole thing. The proper regex you’re looking for is actually:

validates_format_of :something => /^\w$/
April 6, 2009
3 thanks

Assets hosts

You can also setup assets hosts in enviroments:

config.action_controller.asset_host = "http://your-assets-server.com"
April 3, 2009
12 thanks

The docs are in AR::Base

The docs you’re looking for are in ActiveRecord::Base

April 1, 2009
5 thanks

Ordering of format blocks is important

The order in which your format blocks appear, like:

format.html { } format.js { }

are used to infer priority in cases where the appropriate format is ambiguous.

March 31, 2009
3 thanks

Sorting Hashes with Symbol Keys

To sort a hash with symbol keys, use Enumerable#sort_by:

h = { :a => 20, :b => 30, :c => 10  }
h.sort                       # => NoMethodError: undefined method `<=>' for :a:Symbol
h.sort_by { |k,v| k.to_s }   # => [[:a, 20], [:b, 30], [:c, 10]]
March 31, 2009 - (v2.0.0 - v2.3.2)
3 thanks

Override fieldWithErrors markup in Rails > v2

The code posted by @hosiawak will still work in recent versions of Rails, but maybe a more current, idiomatic way to do it is to stick this inside the Rails::Initializer block in environment.rb (obviously you’ll also need to restart your server to pick up the config change):

config.action_view.field_error_proc = Proc.new {|html_tag, instance| 
  %(<span class="fieldWithErrors">#{html_tag}</span>)}
March 27, 2009
4 thanks

Hour with/without preceding zero

One gotcha is the difference between the hour in 12 hour time with and without a preceding zero. In some fonts they look the same.

With preceding zero (capital I)

Time.now.strftime("%I:%M") # => 05:21

Without preceding zero (lowercase L)

Time.now.strftime("%l:%M") # => 5:21
March 27, 2009
4 thanks

Usage

Here’s how to use it, just so it’s perfectly clear:

skip_before_filter :method_to_skip, :only => [:method_name]