timrwood

manager	dist	manifest	export	ignore	include	entry
npm	npm	package.json	cjs	.npmignore		#main
bower	git	bower.json	umd	#ignore		#main
component	git	component.json	umd		#files	#main
spm	git	package.json#spm	umd	.spmignore	#output	#main
packagist	git	composer.json	umd
ender	npm	package.json	cjs	.npmignore		#ender
dojo	???	package.json#dojoBuild	umd			#dojoBuild
jspm	npm/git	package.json#jspm	any	#ignore	#files	#main

chrisk

# config/initializers/cucumber_fakeweb_hook.rb

# If this looks like a test server process, make its FakeWeb available to other
# processes via DRb. That way the test runner (which runs in a separate process)
# can dynamically register stubs for this server to use.

if Rails.env.cucumber? && defined?(PhusionPassenger)
  require 'drb'
  require 'fakeweb'
  DRb.start_service("druby://localhost:30010", FakeWeb)

crofty

Steps required for getting tests running

git clone git://github.com/sproutcore/sproutcore20.git
cd sproutcore20
echo "rvm use 1.9.2@sproutcore" > .rvmrc
cd -
cd sproutcore20
gem update --system 1.7.2
bundle install
gem uninstall spade # Version 0.1.0 isn't working for me yet

dudleyf

require "json"
require "rake-pipeline-web-filters"

input "assets"
output "public"

class HandlebarsFilter < Rake::Pipeline::Filter
  def initialize(&block)
    block ||= proc { |input| input.sub(/\.handlebars$/, '.js') }
    super(&block)

garth

var fs = require('fs')
var vm = require('vm')

var handlebarsjs = fs.readFileSync('path/to/handlebars.js', 'utf8')
var emberjs = fs.readFileSync('path/to/ember.js', 'utf8')
var templatesDir = 'path/to/template/dir'

desc('Compile all .handlebars templates')
task({ 'handlebars': [] }, function () {
  process.stdout.write('Compiling .handlebars templates')

jpoz

require 'goliath'
require 'em-synchrony/em-http'

class UploadProxy < Goliath::API

  def on_headers(env, headers)
    env.logger.info 'received headers: ' + headers.inspect
    env['async-headers'] = headers
  end

ghempton

// This file contains utilities for creating bound helpers

// For reference: https://github.com/wagenet/ember.js/blob/ac66dcb8a1cbe91d736074441f853e0da474ee6e/packages/ember-handlebars/lib/views/bound_property_view.js
Ember.Handlebars.BoundHelperView = Ember.View.extend(Ember._Metamorph, {

  context: null,
  options: null,
  property: null,
  // paths of the property that are also observed
  propertyPaths: [],

robharper

require "rake-pipeline-web-filters"
require "neuter_plus_filter"

input 'src' do
  output 'build'
  match '**/main.js' do
    filter(Rake::Pipeline::Web::ExtraFilters::NeuterPlusFilter,
      # Possibly depend on JS files within same or subdirectories
      :additional_dependencies => proc { |input|
        Dir.glob(File.join(File.dirname(input.fullpath),'**','*.js'))

ghempton

Ember.js Controller Brainstorm

Currently, there are several awkward points to Ember.js controllers. Namely:

1. Passing controllers around to view's not instantiated by the routing system is hard.
2. Non-singleton controllers are very difficult to manage (e.g. #each where each view has it's own controller).
3. Sharing data between controllers is difficult.

Solution for 1 and 2: Controllers are lazily created by views

ahawkins

Iridium: The Next Step Forward

Proposed changes/goals/musings

• Make an iridium command
• Enable: iridium app your_app_name so it generates a complete application structure
• Enable: iridium server # starts a stupid development server serving /site or whatever the application config is
• Enable: iridium generate model|view|template|controller|template|state
• Make testing work with casperjs for integration tests and unit tests
• Tests written in Coffescript by default (casper automatically compiles cs so no extra work needed)