http://htwins.net/scale2/
I love how much information is communicated in such a simple interface. I also like the fact that data appears when it is needed, but doesn't clutter up the screen when it is not needed.
Setting up a Rails3 application quickly, I ran in to this error...
no such file to load -- spec/expectations
easy fix... I had gem 'rspec' but needed to add gem 'rspec-rails'
From
Use Mercurial, you Git!
"If I'd wanted building blocks for rolling my own, I'd have gone to Home Depot and bought a 1 and a 0."
Here's how I typically setup my rails environments. NOTE: I don't include stored procedure or view permissions because Rails doesn't ever use those. Wouldn't it be great if Rails used views or stored procedures? Actually, I kinda like it the way it is. Stored Procedures are nice but the tradeoff, IE having business logic in the app and in the database, is a high price to pay. As for views, actually, that might be really nice for some reporting things. Maybe that'll be a new gem... :) but for now, stick with these.
DEVELOPMENT MACHINE
I create a dev user and a test machine. For each app, I create app_dev and app_test.
1
2
3
4
5
|
CREATE USER test IDENTIFIED BY 'somepass'
CREATE USER test IDENTIFIED BY 'somepass'
GRANT ALL ON app_dev.* to dev;
GRANT ALL ON app_test.* to test;
|
In production I typically have three environments: edge, staging, production. Each has its own database and user.
1
2
3
4
5
6
7
8
9
10
|
CREATE USER 'edge'@'localhost' IDENTIFIED BY 'somepass'
CREATE USER 'staging'@'localhost' IDENTIFIED BY 'somepass'
CREATE USER 'production'@'localhost' IDENTIFIED BY 'somepass'
/* note that, counterintuitively, you do NOT have to */
/* create these databases first. You can let */
/* the Rails app do that with rake db:create or whatever. */
GRANT SELECT, INSERT, UPDATE, DELETE, CREATE, DROP, INDEX ON app_edge.*
GRANT SELECT, INSERT, UPDATE, DELETE, CREATE, DROP, INDEX ON app_staging.*
GRANT SELECT, INSERT, UPDATE, DELETE, CREATE, DROP, INDEX ON app_production.*
|
Typically, you define your path in /etc/environment. Here's a simple example:
1
|
PATH="/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games:/usr/local/ruby/bin"
|
That last bit tells the system where Ruby lives. If Ruby isn't in the path, you won't be able to run it or any gem scripts, like bundle.
Unfortunately, neither /etc/environment nor /etc/profile is respected under certain situations, like the way Capistrano works by default.
The solution for me was to edit /etc/bash.bashrc, adding this as the first line:
1
|
source /etc/environment
|
Note: Adding it as the first line allows / encourages it to be overwritten as needed. If you're still having problems, here's one way to get started with troubleshooting:
edit /etc/bash.bashrc to include this
1
|
export USES_BASHRC='bashrc'
|
edit /etc/profile to include this line:
1
|
export USES_PROFILE = 'profile'
|
then see whether either of those environmental variables is around when capistrano connects. From your dev machine, run "cap shell" and you will end up with a "cap>" prompt. type "echo $USES_BASHRC && echo $USES_PROFILE. Here's what I got:
1
2
3
4
5
|
cap> echo $USES_BASHRC && echo $USES_PROFILE
[establishing connection(s) to www.verdacom.com]
Password:
** [out :: www.verdacom.com] bashrc
** [out :: www.verdacom.com]
|
Thanks to the SliceHost support team for pointing me in the right direction on this one.
REFERENCES
============
http://stefaanlippens.net/bashrc_and_others
I received this message when I ran cucumber tests but not when I used script/console. Kinda strange. Perhaps this is because I didn't configure cucumber for my dev environment?? Anyway, the fix:
1
2
|
gem uninstall rubygems-update
gem update --system 1.4.3
|
In the words of Yoda, "Unfortunate this is, and unexpected."
References
=======
Can not install redmine
Gist -- Undefined local variable dep
RubyGems Developers Mailing List
StackOverflow -- How do you downgrade RubyGems
Ruby 1.9 assumes that everyone has been broken of the bad habit of illogical date formats.
in the console:
1
2
3
4
5
6
7
8
|
>> Date.parse('2/15/2011')
ArgumentError: invalid date
>> Date.parse('2/15/11')
ArgumentError: invalid date
>> '2/15/2011'.to_date
ArgumentError: invalid date
|
While I'm driving everyone crazy by writing the date as YYYY-MM-DD, my clients aren't as OCD. They insist on using the American date format mm/dd/yyyy and it's lazy alternate, mm/dd/yy. Here's the code I wrote to help them with their bad habits.
For my rails apps, in config/initializers/american_date_format.rb:
1
2
3
4
5
6
7
8
9
|
def Date.parse(value = nil)
if value =~ /^(\d{1,2})\/(\d{1,2})\/(\d{2})$/
::Date.civil($3.to_i + 2000, $1.to_i, $2.to_i)
elsif value =~ /^(\d{1,2})\/(\d{1,2})\/(\d{4})$/
::Date.civil($3.to_i, $1.to_i, $2.to_i)
else
::Date.new(*::Date._parse(value, false).values_at(:year, :mon, :mday))
end
end
|
Now:
1
2
3
4
5
6
7
8
9
10
11
12
13
14
|
>> Date.parse('2/15/11')
=> Tue, 15 Feb 2011
>> Date.parse('2/15/2011')
=> Tue, 15 Feb 2011
>> Date.parse('2011-02-15')
=> Tue, 15 Feb 2011
>> '2011-02-15'.to_date
=> Tue, 15 Feb 2011
>> '2/4/11'.to_date
=> Fri, 04 Feb 2011
|
Twilio apparently offers a
telephone api and also an SMS API. I get how you could easily set up an SMS API. Pretty cool stuff. But what amazed me what the idea that someone could call a telephone number, which would result in a post to my website... and I would respond with instructions?!?!!
Ladies and gentlemen, the future is now. My rails app just answered the phone.
I recently received the following errors when running capybara with envjs. Strangely, there was a kind of double stack trace. Here are the exeptions:
===
WARNIING: [Thu Mar 10 2011 09:54:41 GMT-0600 (CST)] {ENVJS} Exception while dispatching events: a.ownerDocument is null
a.ownerDocument is null (Johnson::Error)
===
===
WARNIING: [Thu Mar 10 2011 09:54:41 GMT-0600 (CST)] {ENVJS} Exception while dispatching events: a.ownerDocument is null
oopse a.ownerDocument is null
undefined
inline:129 [JavaScript]
===
===diff
--- a/features/support/env.rb
+++ b/features/support/env.rb
@@ -1,4 +1,4 @@
-ENV["RAILS_ENV"] ||= "cucumber"
+ENV["RAILS_ENV"] ||= "test"
require File.expand_path(File.dirname(__FILE__) + '/../../config/environment')
@@ -20,7 +20,6 @@ require 'capybara/session'
Capybara.default_selector = :css
require 'capybara/envjs'
-Capybara.javascript_driver = :envjs
Before do
User.test = Factory.create(:user)
===
and for reference:
===
diff config/environments/test.rb config/environments/cucumber.rb
< config.gem 'factory_girl'
< config.gem 'rspec', :lib => false
< config.gem 'rspec-rails', :lib => false
< config.gem 'capybara-envjs', :lib => 'capybara/envjs'
---
> config.gem 'capybara-envjs', :lib => 'capybara/envjs'
> config.gem 'factory_girl'
> config.gem 'cucumber', :lib => false
> config.gem 'cucumber-rails', :lib => false, :version => '>=0.3.2' unless File.directory?(File.join(Rails.root, 'vendor/plugins/cucumber-rails'))
> config.gem 'database_cleaner', :lib => false, :version => '>=0.5.0' unless File.directory?(File.join(Rails.root, 'vendor/plugins/database_cleaner'))
===
my complete features/support/env.rb file:
===
ENV["RAILS_ENV"] ||= "test"
require File.expand_path(File.dirname(__FILE__) + '/../../config/environment')
puts "========================================"
puts "RAILS_ENV is #{RAILS_ENV} "
puts "========================================"
require File.expand_path(File.dirname(__FILE__) + '/../../test/helpers/user.rb')
require 'cucumber/formatter/unicode' # Remove this line if you don't want Cucumber Unicode support
require 'cucumber/rails/world'
require 'cucumber/rails/active_record'
require 'cucumber/web/tableish'
require 'capybara/rails'
require 'capybara/cucumber'
require 'capybara/session'
Capybara.default_selector = :css
require 'capybara/envjs'
Before do
User.test = Factory.create(:user)
Given 'I am logged in'
System.destroy_all
System.load
end
ActionController::Base.allow_rescue = false
Cucumber::Rails::World.use_transactional_fixtures = true
if defined?(ActiveRecord::Base)
begin
require 'database_cleaner'
DatabaseCleaner.strategy = :truncation
rescue LoadError => ignore_if_database_cleaner_not_present
end
end
===
Two very useful cucumber steps:
1
2
3
4
5
6
7
|
Then 'display the page' do
puts "\n\n\n#{page.body.to_s}\n\n\n"
end
Then /^display "([^"]*)"$/ do |selector|
puts "\n\n\n#{find(selector).native.to_html}\n\n\n"
end
|
When you're remoting in to your dev machine, you can't use "Then show me the page" because there is no desktop in which to open the browser... so I created "Then display the page". It outputs the HTML to the screen for your examination. This becomes very tedious when pages get long... after some help on the capybara mailing list, I created the second entry, which accepts a CSS selector.
I've been exposed to bundler on a couple of projects recently.
My impression is that it does a poor job of solving a rare problem, while making common tasks more difficult.
Spinner (in german... Chrome translates)
post
example
"Ajaxian: Canvas Loader":http://ajaxian.com/archives/canvas-loading-indicator-for-the-iphone-and-beyond
Apple's Tutorial
Somehow, Rails3 was installed in my 1.8 rvm gemset. rails3 depends on i18n v >= 0.4. Uninstalling i18n (and rails3, since it doesn't run in 1.8 anyway) corrected this issue.
Issue:
*"Missing these required gems: pdf-writer = 1.1.8"*
config/environment.rb:
Issue:
*"Missing these required gems: pdf-writer = 1.1.8"*
config/environment.rb:
1
|
config.gem "pdf-writer", :lib => "pdf/writer", :version => '1.1.8'
|
That looks write... what's the issue? Turns out my new server didn't have the required dependencies.
I added this above the config/gem line:
1
|
require "#{RAILS_ROOT}/vendor/gems/pdf-writer-1.1.8/lib/pdf/writer.rb"
|
and saw:
1
2
|
no such file to load -- color
no such file to load -- transaction-simple # after fixing color
|
the fixes:
1
2
|
gem install color
gem install transaction-simple
|
remove the require line, and voila!
Rails is fortunate to have ActiveMerchant, a gem that provides a standard interface for billing. Unfortunately, there are a wide variety of gateways out there and each has their own "uniqueness." I experienced PayFlowPro's "uniqueness" today and thought I would share my now-working spike code.
THE PARTNER FIELD
The partner field below is optional. If you leave it blank, the default will be PayPal. The other option, apparently, is VeriSign. I have included it here because magical, undocumented parameters confuse people.
MERCHANT LOGIN
in my case, this was the company name without any spaces. It's the same as the merchant login required to log in to the PayflowPro website.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
|
gateway = ActiveMerchant::Billing::PayflowGateway.new(
:login => 'my_company_name', # 'Merchant Login from the login page
:user => 'my_username',
:password => 'ComPliCatedP4sSw0rd',
:partner => 'PayPal'
)
credit_card = ActiveMerchant::Billing::CreditCard.new(
:number => '5105105105105100',
:month => '9',
:year => '2011',
:first_name => 'Longbob',
:last_name => 'Longsen',
:verification_value => '123',
:type => 'master'
)
# Make a $1 purchase (100 cents)
response = gateway.purchase(100, credit_card)
puts response.success?
puts response.message
|
MOVING FROM A SPIKE TO REAL CODE
Copy the above code, replace your company name, and get that to say "Accepted"... once that's done, check back for my blog entry on the easiest way to set up ActiveMerchant in a Rails app.
ISSUES I EXPERIENCED
PayFlowPro response 26 - Invalid Vendor Account
I was using my username where my merchant login should have been.
PayflowPro response 1 - User authentication failed
Here I was submitting the company name as my :login but was not including my username under :user
REFERENCES
com.googlegroups.activemerchant - Payflow Pro Integration
Payflow Gateway - User authentication failed
337 326 3121
jw@mustmodify.com
