gistfile1.rb

# SMELL I don't like the name gemfile_lock_file, but I don't know
# what else to call it. I want to distinguish it from Gemfile.
def extract_gems_from_gemfile_lock_file(contents)
  gems = []
  # Extract all the non-empty lines between and excluding 'specs:' and 'PLATFORMS'
  contents.each_line do |line|
    line = line.strip
    # It takes a few seconds to understand this algorithm,
    # but I can't yet justify replacing it with a functional
    # approach. Again, it depends what the reader understands
    # better.
    # If we're going to let iterating and filtering stay
    # intertwined like this, I'm glad we've separated it from
    # the rest of the code.
    break if line.include?("PLATFORMS")
    next if line.include?("GEM")
    next if line.include?("remote:")
    next if line.include?("specs:")
    next if line.empty?
    gems.push(line.split(' ').first)
  end
  return gems
end

# Guess what? Now independent of "gem" concept. More reusable. MAGIC!
# Guess what? Almost dead-simple wrapper around command line tool. Cohesive. SRP. Nice.
def count_lines_in_file(file)
  output = `wc -l #{file}`
  # ASSERT: output is of the form <whitespace><line count><whitespace><filename>
  # We could turn this comment into code by matching with a regex, which some
  # would find clearer and others less clear. I could do either.
  line_count_text = output.strip.split(' ').first
  return line_count_text.to_i
end

# SMELL Depends on globals like 'puts' and `` (execute process).
def count_lines_for_gem(gem)
  puts "Processing #{gem}"
  gem_filenames = `gem contents #{gem}`.split
  # If you prefer to 'inject', then feel free to 'inject'.
  # Either way, the name duplication disappears.
  line_count_for_this_gem = gem_filenames.map { |each| count_lines_in_file(each) }.reduce(:+)
  puts "  LOC: #{line_count_for_this_gem}"
  return line_count_for_this_gem
end

total = extract_gems_from_gemfile_lock_file(File.read("Gemfile.lock"))
        .map { |each| count_lines_for_gem(each) }.reduce(:+)

puts "Total Lines: #{total}"

If you don't care about the interim line count in each gem, then we can remove the duplicate count_lines concept, replacing the whole thing with this, in approximate Haskell, of course:

sum ((count_lines . filenames_in_gem . gems_in_gemfile_lock) gemfile_lock)