capsule-sync-analyze.rb

capsule-sync-analyze.rb

#!/usr/bin/env ruby
require 'rubygems'
require 'nokogiri'         
require 'set'
require 'yaml'
require 'time'

# Usage:
#
# # run inside a directory with extracted task export
# ruby capsule-sync-analyze.rb | tee capsule_sync_analysis.csv
#

puts "file, capsule_name, started_at, ended_at, duration, status, result, number_of_repos, orgs, envs"

files = ARGV
if files.empty?
  files = Dir.glob('*.html')
end

def repo_id_to_org_env(repo_id)
  parts = repo_id.split('-')
  return if parts.first =~ /^\d*$/ # skip malformed formats
  org, env = parts
  return org, env
end

def parse_time(str)
  Time.strptime(str, '%F %T %Z')
end

files.reverse.each do |filename|
  file = File.read(filename)
  page = Nokogiri::HTML(open(filename))
  ps = page.css('body').css('p')
  headers = ps.map(&:text)[0..5]
  task_data = {}
  headers.each do |line| 
    field, value = line.strip.split("\n").map(&:strip)
    task_data[field.downcase.tr(':','').tr(' ','_')] = value
  end

  if (proxy_data = page.css('.lang-yaml').first)
    data = YAML.load(proxy_data.text)
    if data['smart_proxy']
      task_data['capsule'] = data['smart_proxy']['name']
    end
  end

  next unless task_data['label'] == "Actions::Katello::CapsuleContent::Sync"

  data = YAML.load(proxy_data.text)


  run = page.css('#run').css("table.flow")

  repo_ids = Set.new
  orgs = Set.new
  envs = Set.new

  run.css('td:not(.flow)').each do |tr|
    data = YAML.load(tr.css('.lang-yaml').first.text)
    if data.key?('repo_id')
      repo_ids << data['repo_id']
    end
  end

  repo_ids.each do |repo_id|
    org, env = repo_id_to_org_env(repo_id)
    if org
      orgs << org
      envs << env
    end
  end
  if task_data['ended_at']
    duration = parse_time(task_data['ended_at']) - parse_time(task_data['started_at'])
  end

  # might be useful to group the repos by environments
  # repo_by_env = repo_ids.group_by do |repo_id|
  #   _, env = repo_id_to_org_env(repo_id)
  #   env
  # end

  puts [filename,
        task_data['capsule'],
        task_data['started_at'],
        task_data['ended_at'],
        duration,
        task_data['status'],
        task_data['result'],
        repo_ids.size,
        orgs.to_a.join(' '),
        envs.to_a.join(' '),
       ].map(&:to_s).join(',')
end