Lucky v0.23 is out, and it's huge!

Lucky v0.23 is out now and works with the newest version of Crystal (v0.35.0)

Be sure to upgrade your version of Crystal, and take a look at our UPGRADE NOTES for help with migrating your app.

# How to upgrade

See the UPGRADE_NOTES to learn how to upgrade to Lucky 0.23.0.

Special thanks to community member @stephendolan for creating LuckyDiff. This tool helps see what has changed between versions and will be included in upgrade notes from now on.

# What happened to 0.22.0?

Don’t worry, you didn’t miss anything. If you’re using Lucky 0.21.0, you can upgrade to 0.22.0 just by updating your Crystal version to 0.35.0. This release didn’t change anything other than compatibility with the latest Crystal. To get all of the latest goodies, you’ll want to go up to 0.23.0.

# Here’s what’s new

# `mount` is now `m`

We have deprecated the use of mount in favor of a shorter method m. The structure of this is a little bit different.

# Before
mount MyComponent.new(arg: 1)

# After
m MyComponent, arg: 1

You still have all the flexibility, but in a more compact version. The new m method also opens the doors for more functionality that we’ll be adding to components in future releases of Lucky.

# Trim extra whitespace in params

Leading and trailing whitespace is now stripped automatically when accessing params. If you need to get the unstripped value, you can use params.get_raw(:the_key).

# the user sends the email param as " email@email.com "
params.get(:email) #=> "email@email.com"
params.get_raw(:email) #=> " email@email.com"

# Revamped memoization

The Memoizable module got a nice little upgrade. It now supports arguments, as well as false and nil values.

memoize user(user_id : Int64) : User?
  UserQuery.new.find(user_id)
end

Now when you call this method, it will run the first time, and return the value even if that value is nil. If you need access to the uncached version you can use the __uncached version of the method.

# run the query once, and always return that value
user(4)

# always run the query and return the value
user__uncached(4)

# Security and more

Some of the HTML page helpers like highlight, and simple_format would allow for unescaped HTML. This could potentially be a security issue, so that’s been fixed. All of these methods will escape the HTML by default now. If you need the unescaped version, you can pass escape: false.

highlight("<p>This is a beautiful morning, but also a beautiful day</p>", "beautiful", escape: false)

We’ve also fixed a few issues related to cookies.

You can now set a samesite default.
Exceptions aren’t raised when trying to delete a cookie that doesn’t exist

# Routing Updates

We have added a new route_prefix macro that alllows you to prefix your routes with some path. This really helps with APIs where you may want to prefix all your routes with something like /api/v1.

abstract class ApiAction < Lucky::Action
  accepted_formats [:json], default: :json

  route_prefix "/api/v1"
end

class Posts::Index < ApiAction
  # GET /api/v1/posts`
  get "/posts" do
    #...
  end
end

On top of all the neat routing features, we were also able to squeeze a little more performance out thanks to @matthewmcgarvey!

# CLI Task Args

Writing CLI tasks for your application are pretty common, but they can also be a bit complicated. Lucky now makes it easier to write custom CLI tasks with arguments.

We break CLI arguments in to 3 types:

arg - This is the most standard CLI argument type. e.g. -m User or --model=User.
switch - Just like arg, but without a value. Returns true if this flag is passed. e.g. -v, -h
positional_arg - When you want to pass args that are not flags. Lucky uses this with generators like lucky gen.page About where About is a positional arg.

# lucky search.reindex User
class Search::Reindex < LuckyCli::Task
  summary "Reindex records for a model"

  positional_arg :model, "Specify which model to reindex", required: true

  def call
     # made up code to reindex User
     Elasticsearch.reindex(model: model)
  end
end

# lucky import_data --dry
class ImportData < LuckyCli::Task
  summary "Import the latest data"

  switch :dry, "Perform a dry run before the actual import", shortcut: "-d"

  def call
    if dry?
      # ...
    else
      # ...
    end
  end
end

# lucky generate_sitemaps --environment=dev
class GenerateSitemaps < LuckyCli::Task
  summary "Generate some sitemaps"
  arg :environment, "Specify the ENV for the sitemaps", format: /(dev|prod)/

  def call
    # ...
   end
end

# Lucky Flow

The underlying shard that LuckyFlow wraps has been replaced with the selenium shard by @matthewmcgarvey. This update allows us to support all browsers and not just Chrome. On top of that, we can now stay up-to-date with the new W3C WebDriver standard going forward.

It also automatically installs the correct driver for your version of Chrome so browser testing is easier than ever to get started with.

For the most part, everything still works as it did before, just with a shiny new engine!

# Parting words

This release has seen a huge surge in community contributions. It’s tough maintaining open source projects, so we just want to say how much we appreciate all of the hard work the community has put in to Lucky making this framework so amazing!

We’re really excited about getting closer to a 1.0 release, and we can’t do it without your support. Please give it a spin and help us find bugs so our next release is even more solid. If you find any issues, don’t hesitate to report the issue. If you’re unsure, just hop on gitter chat so we can help you out.

Thanks so much for the support!

# Follow and spread the word

If you haven’t already, give us a star on github, and be sure to follow us on Twitter.

For questions, or just to chat, come say hi on gitter.