Checking if a JavaScript value is of a given type can be very confusing:

• There are two operators typeof and instanceof which work very differently.
• JavaScript has some primitive types, like string literals, that are not objects (as opposed to Ruby, where every value is an object).
• Some values are sometimes a primitive value (e.g. "foo") and sometimes an object (new String("foo")) and each form requires different checks
• There are three different types for null (null, undefined and NaN) and each has different rules for...

RSpec allows you to mark a single Example/ExampleGroup so that only this will be run. This is very useful when using a test runner like guard.

Add the following config to spec/spec_helper.rb:

RSpec.configure do |config|
  # These two settings work together to allow you to limit a spec run
  # to individual examples or groups you care about by tagging them with
  # `:focus` metadata. When nothing is tagged with `:focus`, all examples
  # get run.
  config.filter_run_including :focus => true
  config.run_all_when_everything_filtere...

If you already selected an element and want to get its parent, you can call find(:xpath, '..') on it.

To get the grand-parent element, call find(:xpath, '../..').

Example

Find a link which contains a twitter icon and check that it links to the correct page:

<a href="http://twitter.com/">
  <i class="icon is-twitter"></i>
</a>

link = page.find("a .icon.is-twitter").find(:xpath, '..')
expect(link[:href]).to eq("http://twitter.com/")

About XPath

There is a good overview on XPath syntax on [w3schools](htt...

An end-to-end test (E2E test) is a script that remote-controls a web browser with tools like Selenium WebDriver. This card shows basic techniques for fixing a flaky E2E test suite that sometimes passes and sometimes fails.

Although many examples in this card use Ruby, Cucumber and Selenium, the techniques are applicable to all languages and testing tools.

Why tests are flaky

Your tests probably look like this:

When I click on A
And I click on B
And I click on C
Then I should see effects of C

A test like this works fine...

When your public-facing application has a longer downtime for server maintenance or long migrations, it's nice to setup a maintenance page to inform your users.

When delivering the maintenance page, be very careful to send the correct HTTP status code. Sending the wrong status code might get you kicked out of Google, or undo years of SEO work.

Popular footguns

Here are some ways to shoot yourself in the foot during maintenance:

• If all your routes send a "200 OK" with a HTML body "We're b...

Browsers support different types of redirects.

Be very careful with these status codes:

• 301 Moved Permanently
• 308 Permanent Redirect

Most browsers seem to cache these redirects forever, unless you set different Cache-Control headers. If you don't have any cache control headers, you can never change them without forcing users to empty their cache.

Note

By default Rails sends a ...

When you want to group rails models of a logical context, namespaces are your friend. However, if you have a lot of classes in the same namespace it might be tedious to specify the table name for each class seperately:

class Accounting::Invoice < ApplicationRecord
  self.table_name = 'accounting_invoices'
  ...
end

class Accounting::Payment < ApplicationRecord
  self.table_name = 'accounting_payments'
  ...
end

A replacement for the self.table_name-assignment is the table_name_prefix in the module definition:

modu...

RSpec allows defining methods inside describe/context blocks which will only exist inside them.
However, classes (or any constants, for that matter) will not be affected by this. If you define them in your specs, they will exist globally. This is because of how RSpec works (short story: instance_eval).

describe Notifier do
  class TestRecord
    # ...
  end
  
  let(:record) { TestRecord.new }
  
  it { ... }
end

# TestRecord will exist here, outside of the spec!

This will come bite you at least when you try to define a ...

When localizing model attributes via I18n you may run into errors like this:

I18n::InvalidPluralizationData: translation data { ... } can not be used with :count => 1. key 'one' is missing.

They seem to appear out of the blue and the error message is more confusing than helpful.

TL;DR A model (e.g. Post) is lacking an attribute (e.g. thread) translation.
Fix it by adding a translation for that model's attribute (attributes.post.thread). The error message reveals the (wrongly) located I18n data (from `attributes.thread...

Background

• before(:all) runs the block once before all of the examples.
• before(:each) runs the block once before each of your specs.

Summary

• around(:suite) does not exist.
• around(:all) runs after before(:all) and before after(:all).
• around(:each) runs before before(:each) and after after(:each).

As this is not 100% obvious (and not yet documented) it is written down in this card. In RSpec 3 :each has the alias :example and :all the alias :context.

Example

RSpec.configure do |config|
  ...

You can use local copies of gems in your Gemfile like this:

gem 'spreewald', path: '~/gems/spreewald'

As soon as you have bundled your project with the local copy of the gem, all code changes in the copy will be available on your project. So you can for example set a debugger or add console output in the gem and use it from your project.
If you checked out the gem with your versioning tool, you can easily reset your changes afterwards or make a pull request for the gem if you improved it.

Don't commit a Gemfile with local path...

I often see the use of || to set a default value for a variable that might be nil, null or undefined.

x = x || 'default-value'

This pattern should be avoided in all languages.

While using || works as intended when x is null or an actual object, it also sets the default value for other falsy values, such as false. false is a non-blank value that you never want to override with a default.

To make it worse, languages like JavaScript or Perl have [many more fal...

Nowadays it is fairly easy to intercept and modify mails globally before they are sent. All you have to do is register an interceptor class which responds to .delivering_email(message). This card will show you two common use cases.

Subject prefix:

Usually you want to prefix the subject line of emails with the current environment (except production) so you can differentiate between production mails and mails from other environments. Of course a...

Rails supports time zones, but there are several pitfalls. Most importantly because Time.now and Time.current are completely different things and code from gems might use one or the other.

Especially configuring an application that cares only about one time zone is a bit tricky.

The following was tested on Rails 5.1 but should apply to Rails 4.2 as well.

Using only local time

Your life will be easier if your application does not need to support time zones. Disable them like this:

config.time_zone = 'Berlin' # Your local ...

Database connections are not thread-safe. That's why ActiveRecord uses a separate database connection for each thread.

For instance, the following code uses 3 database connections:

3.times do
  Thread.new do
    User.first # first database access makes a new connection
  end
end

These three connections will remain connected to the database server after the threads terminate. This only affects threads that use ActiveRecord.

You can rely on Rails' various clean-up mechanisms to release connections, as outlined below. This may...

This card compares patterns to store trees in a relation database like MySQL or PostgreSQL. Implementation examples are for the ActiveRecord ORM used with Ruby on Rails, but the techniques can be implemented in any language or framework.

We will be using this example tree (from the acts_as_nested_set docs):

root
|
+-- Child 1
|   |
|   +-- Child 1.1
|   |
|   +-- Child 1.2
|
+-- ...

Some pseudo-elements need to be addressed with vendor prefixes. E.g. ::selection is not supported by Firefox, you need to use ::-moz-selection instead.

What you cannot do is to define a single CSS rule to address both the standard and vendor-prefixed form:

::selection, ::-moz-selection {
  background-color: red;
}

This rule will be ignored by all browsers. The reason is that if a browser doe...

There are various ways to run external commands from within Ruby, but the most powerful ones are Open3.capture3 and Open3.popen3. Since those can do almost everything you would possibly need in a clean way, I prefer to simply always use them.

Behind the scenes, Open3 actually just uses Ruby's spawn command, but gives you a much better API.

Open3.capture3

Basic usage is

require 'open3'

stdout_str, error_str, status = Open3.capture3('/some/binary', 'with', 'some', 'args')
if status.success?...

We generally use multiple application servers (at least two) and you have to search on all of them if you don't know which one handled the request you are looking for.

Rails application logs usually live in /var/www/<project-environment-name>/shared/log.
Web server logs usually live in /var/www/<project-environment-name>/log.

Searching through single logs with grep / zgrep

You can use grep in this directory to only search the latest logs or zgrep to also search older (already zipped) logs. zgrep is used just like grep ...

First find the reference for the entry you want through looking at the stash:

$ git stash list
stash@{0}: WIP on feature/foo
stash@{1}: WIP on feature/bar
stash@{2}: WIP on fix/baz

Now you can simply use that reference, but curly braces must be escaped:

git stash pop stash@\{1\}

or quoted:

git stash apply "stash@{1}"

Quick reminder to [not shoot yourself in the foot](https://makandracards.com/makandra/634-use-the-git-stash-withou...

Sometimes huge refactorings or refactoring of core concepts of your application are necessary for being able to meet new requirements or to keep your application maintainable on the long run. Here are some thoughts about how to approach such challenges.

Break it down

Try to break your refactoring down in different parts. Try to make tests green for each part of your refactoring as soon as possible and only move to the next big part if your tests are fixed. It's not a good idea to work for weeks or months and wait for all puzzle pieces ...

In general, you should not put a block element inside an inline element. So don't do this:

<span>
  <div>text</div>
</span>

The browser will think you wrote invalid HTML by accident, and will sometimes reorder elements silently.

There is one notable exception: It's OK to wrap block elements in a <a> tag in HTML5 (not 4). The spec says:

The a element may be wrapped around entire paragraphs, lists, tables, and so forth, even entire sections, so long ...

This card tries to summarize by example the different uses of heredoc.

• In Ruby << vs. <<- vs. <<~
• In Rails strip_heredoc vs. squish

strip_heredoc should be used for a text, where you want to preserve newlines. (multi-line -> multi-line)

squish should be used for a text, where you want to squish newlines. (multi-line -> one-line)

Ruby 2.3+

def foo
  bar = <<~TEXT
    line1
    line2
    line3
  TEXT
  puts bar.inspect
end
foo => "line1\nline2\nline3\n"

Read more: [Unindent HEREDOCs in Ruby 2.3](/m...

There is no CSS selector for matching elements that contains a given string ¹. Luckily, Capybara offers the :text option to go along with your selector:

page.find('div', text: 'Expected content')

You can also pass a regular expression!

page.find('div', text: /Expected contents?/i)

Note that if your CSS selector is as generic as div, you might get a lot more results than you expect. E.g. a <div class="container"> that surrounds your entire layout will probably also contain that text (in a descendant) and ...