Adds a Danger Process command that outputs the DSL as a JSON object

[orta]

Re: #340

This will now take the results of the Danger PR and pass that into the STDIN of a child process. For example this Ruby Script:

#!/usr/bin/env ruby

puts 'Hello world'

str = STDIN.tty? ? 'Cannot read from STDIN' : $stdin.read
puts str

With this command danger process ./scripts/danger_runner.rb would echo first "Hello world" then output the JSON in the above screenshot. You could make a trivial version of Ruby Danger with this:

#!/usr/bin/env ruby

require 'json'
dsl_json = STDIN.tty? ? 'Cannot read from STDIN' : $stdin.read
danger = JSON.parse(dsl_json)

if danger.github.pr.body.include? "Hello world"
  # TODO: sending messages back
end

Now I need to handle sending messages back to Danger JS.

[DangerCI]

New dependencies added: .

Generated by 🚫 dangerJS

[zdnk]

Is it somehow possible to do this for danger Ruby version and then use the output of Ruby version be parsed and worked with in danger-js? Because we have private Bitbucket server which cannot be used with danger-js yet.

[orta]

It's possible, as with all programming anything is possible but it's not something I plan on doing maintaining three processes that communicate with each other for this feature sounds complex. If you're looking to make Bitbucket work with Danger Process, Bitbucket support will need to be added to Danger JS.

[bdotdub]

Minor clarification: JSON file may be more accurate as JSON document if I'm understanding correctly.

[orta]

You're correct, I've just updated the changelog: d0fc1eb