Simple ruby heap dump json navigator

heap-navigator.rb

#!/usr/bin/env ruby
# encoding: utf-8

require 'pry'
require 'json'
require 'awesome_print'

AP_OPTIONS = {indent: 2, index: false, multiline: false}

def load_heap(file)
  file.collect do |l|
    process_dump_line(l)
  end.compact
end

def process_dump_line(line)
  symbolize_keys(JSON.load(line))
rescue JSON::ParserError => e
  puts "Error parsing line: #{e.message}"
  nil
end

def symbolize_keys(hash)
  hash.inject({}) do |result, (key, value)|
    result.update(key.to_sym => value)
  end
end

def class_names(heap_classes)
  heap_classes.inject({}) do |result, elem|
    result.update(elem[:class] => elem[:name])
  end
end

def heap_by_type(heap)
  heap.inject(Hash.new{[]}) do |result, elem|
    type = elem[:type].downcase.to_sym
    result.update(type => result[type] << elem)
  end
end

def address_map(heap)
  heap.inject({}) do |result, elem|
    result.update(elem[:address] => elem)
  end
end

def sort_by_memsize(heap_objects)
  heap_objects.sort_by { |elem| -(elem[:memsize] || 0) }
end

def print_elem(elem)
  ap(elem, AP_OPTIONS)
end

puts "Heap navigator loading #{ARGV[0]}"

file = if ARGV[0].end_with?('.gz')
    Zlib::GzipReader.open(ARGV[0])
  else
    File.open(ARGV[0])
  end

HEAP = load_heap(file)
HEAP_BY_TYPE = heap_by_type(HEAP)
CLASS_NAMES = class_names(HEAP_BY_TYPE[:class])
HEAP_BY_SIZE = sort_by_memsize(HEAP)
HEAP_BY_ADDRESS = address_map(HEAP)

puts "Heap loaded"
puts
puts "Largest instances:"
HEAP_BY_SIZE[0...10].each { |e| print_elem(e) }

Pry.start