18F/hash-joiner

#! /usr/bin/env ruby
# hash-joiner - Pruning, promoting, deep-merging, and joining Hash data
#
# Written in 2015 by Mike Bland (michael.bland@gsa.gov)
# on behalf of the 18F team, part of the US General Services Administration:
# https://18f.gsa.gov/
#
# To the extent possible under law, the author(s) have dedicated all copyright
# and related and neighboring rights to this software to the public domain
# worldwide. This software is distributed without any warranty.
#
# You should have received a copy of the CC0 Public Domain Dedication along
# with this software. If not, see
# <https://creativecommons.org/publicdomain/zero/1.0/>.
#
# ---
#
# Script to generate a 'public' Array version of 'private' YAML files.
#
# The command line flags support stripping object properties other than
# `private:`, or promoting the data associated with the property rather than
# stripping it out.
#
# Author: Mike Bland (michael.bland@gsa.gov)
# Date:   2015-02-18

require 'hash-joiner'
require 'optparse'
require 'safe_yaml'

options = {
  :property => 'private',
}

opt_parser = OptionParser.new do |opts|
  opts.banner = "Usage: #{$0} [-hp] [--promote] [file ...]"

  opts.separator ''
  opts.separator <<EOF
By default, reads a collection of YAML files containing "private" data, i.e.
objects containing a `private:` property, and prints a single YAML Array to
standard output with all the private data stripped out.

Using the option flags, other properties besides `private:` can be removed, or
the data associated with the target PROPERTY can be promoted rather than
stripped.

Options:
EOF

  opts.on('-h', '--help', "Show this help") do
    puts opts
    exit
  end

  opts.on('-p', '--property PROPERTY',
    'Property to strip/promote ' +
    "(default: #{options[:property]})") do |property|
    options[:property] = property
  end

  opts.on('--promote',
    'Promote the PROPERTY rather than strip it') do |promote|
    options[:promote] = promote
  end
end
opt_parser.parse!

if ARGV.length < 1
  STDERR.puts 'No input files specified'
  exit 1
end

input_files = []
errors = []

ARGV.each do |input_file|
  if !File.exists? input_file
    errors << "File does not exist: #{input_file}"
  elsif !File.readable? input_file
    errors << "File not readable: #{input_file}"
  else
    input_files << input_file
  end
end

unless errors.empty?
  STDERR.puts errors.join "\n"
  STDERR.puts "Aborting; no files processed."
  exit 1
end

FILTERED_PROPERTY = options[:property]
FILTER_OPERATION = options[:promote] ? :promote_data : :remove_data

result = []

input_files.each do |input_file|
  data = SafeYAML.load_file(input_file, :safe=>true)
  if data
    result << HashJoiner.send(FILTER_OPERATION, data, FILTERED_PROPERTY)
  else
    errors << "Failed to parse #{source}" 
  end
end

unless errors.empty?
  STDERR.puts "\n*** Errors:"
  STDERR.puts errors.join "\n"
  exit 1
end

puts result.to_yaml