Mist
A Git-powered, Gist-backed blogging engine for Rack / Ruby on Rails applications.
Quickest Possible Usage
To go from zero to running as quickly as humanly possible, follow these steps:
-
Generate rails project:
$ rails g my_project
-
Add mist
gem and install bundle:
$ echo "gem 'mist', :git => 'http://github.com/sinisterchipmunk/mist'" >>Gemfile
$ bundle install
-
Run the Mist installer:
$ rails g mist:setup
-
Visit http://localhost:3000/posts and enjoy!
Authorization
Mist doesn't automate authorization. Since it's designed to be dropped into an already-existing Rails application, it is assumed that you have already settled on an authentication / authorization scheme.
Mist provies the authorize
hook to allow you to run code when needed, returning a true or false value to indicate the user is or isn't authorized, respectively. If true, the user will be allowed to create new posts, edit existing ones, and delete them.
Here's an example of setting up basic authentication using Mist, assuming the logged-in user is in a controller method called current_user
:
Mist.authorize do |controller|
# Authorize only if the user is logged in and is an admin
controller.current_user and controller.current_user.admin?
end
See the documentation for Mist::Configuration for much more detailed information about how to authorize users.
Views
When you run the mist:setup
generator, Mist creates a layout file at app/views/layouts/mist/posts.html.erb
and some assets at app/assets/javascripts/mist.js
and app/assets/stylesheets/mist.css.scss
. Though not strictly required, you should seriously consider modifying these views, if only so you have a personalized site layout.
If you would like complete control over all of the Mist view templates, run the mist:views
generator, which will copy all of Mist's views and assets into app/
, where you can then customize them to your heart's content.
Tweaking
You can configure Mist with an initializer. Just create a file in config/initializers/mist.rb
. The configuration options you should think about right away are:
Mist.title = "Blog Title"
# The title of the blog
Mist.author.name = "Blog Owner's Name"
# The real (or false?) name of the blog owner, as visitors should see it
Mist.author.email = "Blog Owner's Email"
# A contact email address for the blog owner
ActiveGist::API.username = "Blog Owner's GitHub Username"
ActiveGist::API.password = "Blog Owner's GitHub Password"
# GitHub credentials aren't really required, but Gists will be posted
# anonymously and can't be edited unless credentials are given.
Post Format
Mist posts use Markdown format. There's not much more to say about that, except concerning code examples.
Let's take the following example post:
# Here is a header
Here is a paragraph of post content
def code_example_start
@code_counter += 1
end
The above example will be rendered using normal Markdown until the code example is encountered. At this point, mist will extract the code from the example, and send it off into a Gist. If there are multiple code examples, each code example is treated as a separate file within the same Gist.
When the blog is rendered as HTML, the Gist is embedded directly into it.
Gists are great because they allow other people to fork your code examples, recommend changes, comment on them, and so on. But they suddenly become really powerful if you adopt the convention of naming your code examples. Let's take the same example, with this change:
# Here is a header
Here is a paragraph of post content
file: code_snippet.rb
def code_example_start
@code_counter += 1
end
Note the file: ...
line. If Mist finds this line at the top of your code example, it will extract the filename and send it as the name of the example's particular file within the Gist. Then, GitHub will auto-detect the example's format from the file extension, and do all the syntax highlighting of your code when it is embedded!
Graceful Recovery
If the Gist cannot be found (GitHub is down or the Gist has been deleted), Mist will fall back to regular Markdown formatting for the code example. You'll lose syntax highlighting, but the code itself will still be formatted properly.
The Mist Repository
By default, Mist keeps its git repository at db/mist.repo.#{Rails.env}
. You can configure this by setting Mist.repository_location
in config/initializers/mist.rb
if you want.
If the repository doesn't exist the first time Mist tries to read from or write to it, Mist will create a new git repository out of thin air. Mist doesn't set up any remotes or whatever; it just build a bare-bones git repo so that it has something to commit to.
Because it's Git, you can jump into it and add remotes, create branches and check out, add tags -- and do pretty much whatever you like with it.
Pushing
There is a plan to add to Mist the ability to automatically push posts after committing. However, this isn't implemented today. Instead, if you need this functionality then you should look into setting up a scheduled task that simply pushes the git repository at a regular interval. Unix and Mac users, see cron.