How do I programmatically set a content_security_policy? - ruby

I'm configuring the Content Security Policy for our Rails 5.2 app. I need to whitelist some domains in our CSP. I'd like to put the list of domains elsewhere so I can reference them in other places in the application, then generate the CSP headers programmatically from that list.
Looking at the source code for the Content Security Policy configuration mechanisms in Rails 5, it looks like there's some magic metaprogramming going on, so it's not clear to me how to accomplish what I need to do. It looks like the functions I need to call to set headers might be picky about how exactly they want to be called. In particular, it's not clear to me if I can pass them arrays or safely call them multiple times, or if they do some metaprogramming magic that only works if the domains are passed in as individual function arguments.
Can I pass in an array to the header I want to set, like this?
whitelisted_domains = ['https://example.com', 'self']
Rails.application.configure do
config.content_security_policy do |csp|
csp.child_src whitelisted_domains
end
end
Or can I call the same function multiple times, like this?
whitelisted_domains = ['https://example.com', 'self']
Rails.application.configure do
config.content_security_policy do |csp|
whitelisted_domains.each {|domain| csp.child_src domain}
end
end
If neither of those will work, what's the best way of accomplishing what I want to do?

From what I can tell from sourcecode and documentation, it takes an array. From the edgeguides at rails, posting following
Rails.application.config.content_security_policy do |policy|
policy.default_src :self, :https
...
end
and the sourcecode, using *sources as param; it believe it takes any number of arguments, meaning you could do something along the lines of;
whitelisted_domains = ['https://example.com', 'self']
Rails.application.configure do
config.content_security_policy do |csp|
csp.child_src(*whitelisted_domains)
end
end
https://blog.sqreen.io/integrating-content-security-policy-into-your-rails-applications-4f883eed8f45/
https://edgeguides.rubyonrails.org/security.html#content-security-policy
Sourcecode of define_method for each directive
https://github.com/rails/rails/blob/master/actionpack/lib/action_dispatch/http/content_security_policy.rb#L151
(note: None of this has been tested in a Rails app, simple looking guides and source code of Rails)

Related

How to extend redcarpet to support auto linking user mentions?

On my rails3 application I want to use redcarpet to handle user's posts and the user comment section. As such I'd like to extend redcarpet to support turning #username into a link to a user on my site. I know redcarpet is written in C but is there anyway easy way to extend it in ruby? How hard would it be to write it in C? Should I just do this outside of redcarpet?
Also I'm intrested in some other extensions of redcarpet that would be shorthand for linking to other models in my app. I'm not sure the syntax yet but I'm guessing it would be similar to how github handles linking to issues.
I found it pretty easy to extend redcarpet's parser in Ruby for my rails 3 app. It wasn't scary at all.
First, start by deriving a class from Redcarpet's HTML renderer and override the preprocess method as recommended in the docs. In Rails 3.2 and Rails 4, this file can go anywhere and you don't need to require it. I use a 'services' folder to hold code like this.
# app/services/my_flavored_markdown.rb
class MyFlavoredMarkdown < Redcarpet::Render::HTML
def preprocess(text)
text
end
end
Next step is to add methods that do text substitutions you want. Here I use regex to wrap text that looks like "#mention" in an HTML span tag with a css class 'mention'.
# app/services/my_flavored_markdown.rb
class MyFlavoredMarkdown < Redcarpet::Render::HTML
def preprocess(text)
wrap_mentions(text)
end
def wrap_mentions(text)
text.gsub! /(^|\s)(#\w+)/ do
"#{$1}<span class='mention'>#{$2}</span>"
end
text
end
end
You could just as easily look up a user's profile page and wrap the #mention in an anchor tag instead. In my case, I also made methods for emoticons and hashtags that worked the same way and chained the methods together.
The last step is to add a helper that accepts some text, creates an instance of your Redcarpet-derived class, passes the text into that for processing, and returns the html result.
# app/helpers/application_helper.rb
def flavored_markdown_to_html(text)
renderer = MyFlavoredMarkdown.new()
# These options might be helpful but are not required
options = {
safe_links_only: true,
no_intra_emphasis: true,
autolink: true
}
Redcarpet::Markdown.new(renderer, options).render(text)
}
In your views you can call it like this:
<%= flavored_markdown_to_html("This is something worth #mentioning") %>
The output would then be:
This is something worth <span class='mention'>#mentioning</span>.
I once tried to extend redcarpet, but found it very difficult. If there are no other dependencies on redcarpet I'd recommend you try rpeg-markdown which is a (somewhat outdated) Ruby gem providing bindings to the excellent peg-markdown.
peg-markdown is a markdown interpreter written as a formal grammar. This means that it is very easy to extend it with own syntax. I've successfully extended peg-markdown for my own projects (see my fork here) and I found it to be much simpler than fiddling with redcarpet's custom parser code.
I also found peg-markdown to have fewer bugs.
The Ruby bindings may have to be made current by updating the git submodule. (I'm planning to submit a pull request to update rpeg-markdown to the latest version of peg-markdown.)

render individual file in middleman

I am writing a helper and I need to get a rendered file as String.
I see that the method that I need exists in the middleman's library: http://rubydoc.info/github/middleman/middleman/Middleman/CoreExtensions/Rendering/InstanceMethods#render_individual_file-instance_method
How do I call this function from my helper class?
I tried:
require "middleman-core/lib/middleman-core/core_extensions/rendering.rb"
...
puts Middleman::CoreExtensions::Rendering::InstanceMethods.render_individual_file(filepath)
But it does not seem to find the file, any idea?
I'm not sure 3.0 beta is quite ready for primetime.
That said, it does sound like the partial method is what you're looking for.
Unless I'm missing something, the Middleman method seems like an overly-complex solution. For one of my sites I wanted to load entire text files into my templates, so I wrote this helper:
# Shortcut for loading raw text files. Obviously assumes that given file is in a valid format.
# #return [String] File contents
def load_textfile(filename)
File.read filename.to_s
end
Also, you should clarify if you are intending to use this within a template, or within Ruby code. It's not clear to me based on your question.
Here is an example of how one would use above helper:
Currently of note, Middleman is in the process of transitioning to version 4, and the conventions for loading helpers will change. The most straightforward way to define a helper is within a helper block in your config.rb file, as follows:
helpers do
# Define helper functions here to make them available in templates
end
I use Slim for templating. It really is the best. In slim you would appply helper as thus:
= load_textfile 'path'
p You can embed helper output in your page with interpolation, too: #{load_textfile 'path'}

Partial HAML templating in Ruby without Rails

I really don’t need the overhead of Rails for my very small project, so I’m trying to achieve this just using just plain Ruby and HAML.
I want to include another HAML file inside my HAML template. But I haven’t found a good—or really usable—way of doing this.
For example, I have these two HAML files:
documents.haml
%html
%body
= include(menu.haml) body
%article …
menu.haml
%ul
%li
%a whatever …
Include is obviously not the way to go here. But it does a nice job describing what I’m trying to achieve in this example.
I totally recommend the Tilt gem for these things. It provides a standard interface for rendering many different template langages with the same API, lets you set custom scope and locals, lets you use yield, and is robust and fast. Sinatra is using it for templates.
Example:
require 'haml'
require 'tilt'
template = Tilt.new('path/to/file.haml')
# => #<Tilt::HAMLTemplate #file="path/to/file.haml" ...>
layout = Tilt.new('path/to/layout.haml')
output = layout.render { template.render }
This lets you yield inside the layout to get the rendered template, just like Rails. As for partials, David already described a simple and nice way to go.
But actually, if what you're writing is going to be served over HTTP, i suggest you take a look at Sinatra, which already provides templating, and has the simplest request routing you could imagine.
I've done this before, just for a quick-and-dirty template producer. The easiest way is to just render the HAML inside the parent object:
%p some haml that's interesting
= Haml::Engine.new('%p this is a test').render
%p some more template
You'll more than likely want to build some methods to make this easier--a couple of helper methods. Maybe you write one called render_file that takes a filename as an argument. That method might look something like:
def render_file(filename)
contents = File.read(filename)
Haml::Engine.new(contents).render
end
Then, your template would look more like:
%p some template
= render_file('some_filename.haml')
%p more template
Note, you will probably need to pass self to the original Haml::Engine render so that it knows how to find your render_file method.
I've had great success just using the instructions posted by David Richards in a concatenated way, without variables, like this:
= Haml::Engine.new(File.read('/Volumes/Project/file_to_include.haml')).render
There's obviously a more elegant way. But for someone who just wants to include one or two files, this should work nicely. It's a drawback that all base files using these includes have to be recompiled after some changes to the latter. It might be worthwile to just use php include if the environment allows that.
def render_file(filename)
contents = File.read('views/'+filename)
Haml::Engine.new(contents).render
end
(Adding this semi-redundant answer to show how one might incorporate the techniques from other answers.)
Include something like this in your setup code to monkey-patch the Haml library.
module Haml::Helpers
def render_haml(filename)
contents = File.read(Rails.root.join("app", "assets", "templates", filename))
Haml::Engine.new(contents).render
end
end
Then in your Haml file
.foo
= render_haml('partial.haml')
.bar
Obviously this is a Rails-ish example (because I wanted to use HAML in my asset pipeline instead of as views)... You will want to replace Rails.root.join(...) with a way to find filename in your project's templates or partials directory.

Get absolute (base) url in sinatra

Right now, I do a
get '/' do
set :base_url, "#{request.env['rack.url_scheme']}://#{request.env['HTTP_HOST']}"
# ...
haml :index
end
to be able to use options.base_url in the HAML index.haml.
But I am sure there is a far better, DRY, way of doing this. Yet I cannot see, nor find it. (I am new to Sinatra :))
Somehow, outside of get, I don't have request.env available, or so it seems. So putting it in an include did not work.
How do you get your base url?
You can get it using request.base_url too =D (take a look at rack/request.rb)
A couple things.
set is a class level method, which means you are modifying the whole app's state with each request
The above is a problem because potentially, the base url could be different on different requests eg http://foo.com and https://foo.com or if you have multiple domains pointed at the same app server using DNS
A better tactic might be to define a helper
helpers do
def base_url
#base_url ||= "#{request.env['rack.url_scheme']}://#{request.env['HTTP_HOST']}"
end
end
If you need the base url outside of responding to queries(not in a get/post/put/delete block or a view), it would be better to set it manually somewhere.

Generate XML with soap4r without invoking the web service?

I have set up a soap4r client for a web service, and it's working fairly well. We're using it to send data from one database to another (don't bother asking about that... I know it's not optimal), but we're not entirely sure the mapping is correct, so it's often very handy to get the XML that a particular record would generate.
Of course, that's possible - if you set $DEBUG, soap4r will supply you with a nice dump of the XML going over the wire. You can even set the "device" (file) that you would like to send it to.
However, I'd like to be able to get the XML that it's going to generate without having to actually call the web service.
Is there a way to do this? Grepping around, I've found a variety of obj2soap and similar methods, but none of them seems to be quite the one I want.
An indirect answer: you might want to look at handsoap. It's faster and tries to be more Ruby-like. It uses builder-style XML generation - but you have to generate everything yourself. It's more like a toolbox to write your client in a clean way. This way you know what was generated (and can inspect it easily).
Another option is to set $DEBUG and restore it afterwards:
$REMEMBER_DEBUG_STATE = $DEBUG
$DEBUG = true
# call soap (and have your XML generated)
$DEBUG = $REMEMBER_DEBUG_STATE
This could be extracted to a nice function like this:
def with_debug_output
remember = $DEBUG
$DEBUG = true
yield if block_given?
$DEBUG = remember
end
and then use it:
with_debug_output do
# call soap
end

Resources