rsrcmeter v2 (See related posts)

TextDrive: Simple disk usage and HTTP bandwidth counts
This script will allow a TextDrive account holder to easily see their disk usage and monthly HTTP bandwidth consumption, until TextPanel provides the information.
The first time you use rsrcmeter, it may take additional time, as it will be scanning every Apache/HTTP log file for usage information and caching it by month.
(Updated version from http://textsnippets.com/posts/show/632)

Quick install instructions and most up-to-date version is at http://rsrcmeter.textjoy.com/.

Download latest version at http://rsrcmeter.textjoy.com/rsrcmeter-latest.txt.
(Does not appear to work on Shared Accelerators, for disk quota, since the quota tools doesn't return any output currently and doesn't support the group option.)

Sample output:

Disk usage: 23.7227 MiB (Quota: 10.0000 GiB | 0.2% used)
Bandwidth:
Nov 2006: 0.5494 MiB (Month to Date)
Oct 2006: 0.5087 MiB
Sep 2006: 0.5523 MiB
Aug 2006: 1.1567 MiB

#!/usr/local/bin/ruby
# Unofficial TxD Disk & HTTP Bandwidth Usage Meter (rsrcmeter)
# Version 2.0.2
# Written by AJ Zmudosky
#
# THIS SOFTWARE IS PROVIDED BY THE AUTHOR "AS IS" AND WITHOUT WARRANTY OF
# ANY KIND. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE FOR ANY
# DAMAGES HOWEVER CAUSED, AND ON ANY THEORY OF LIABILITY, ARISING IN ANY
# WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY
# OF SUCH DAMAGE.

require 'date'
require 'optparse'

# E-mail regexp
module RFC2822
  EmailAddress = begin
    alpha = "a-zA-Z"
    digit = "0-9"
    atext = "[#{alpha}#{digit}\!\#\$\%\&\'\*+\/\=\?\^\_\`\{\|\}\~\-]"
    dot_atom_text = "#{atext}+([.]#{atext}*)*"
    dot_atom = "#{dot_atom_text}"
    qtext = '[^\\x0d\\x22\\x5c\\x80-\\xff]'
    text = "[\\x01-\\x09\\x11\\x12\\x14-\\x7f]"
    quoted_pair = "(\\x5c#{text})"
    qcontent = "(?:#{qtext}|#{quoted_pair})"
    quoted_string = "[\"]#{qcontent}+[\"]"
    atom = "#{atext}+"
    word = "(?:#{atom}|#{quoted_string})"
    obs_local_part = "#{word}([.]#{word})*"
    local_part = "(?:#{dot_atom}|#{quoted_string}|#{obs_local_part})"
    no_ws_ctl = "\\x01-\\x08\\x11\\x12\\x14-\\x1f\\x7f"   
    dtext = "[#{no_ws_ctl}\\x21-\\x5a\\x5e-\\x7e]"
    dcontent = "(?:#{dtext}|#{quoted_pair})"
    domain_literal = "\\[#{dcontent}+\\]"
    obs_domain = "#{atom}([.]#{atom})*"
    domain = "(?:#{dot_atom}|#{domain_literal}|#{obs_domain})"
    addr_spec = "#{local_part}\@#{domain}"
    pattern = /^#{addr_spec}$/
  end
end

# Default options
opt = {
  :base_dir => ENV['HOME'],
  :stat_file => 'histstat',
  :temp_file => 'temp-rsrcmeter-bwcalc',
  :bw_months => 5
}

# Our option flag parsing
OptionParser.new do |opts|
  opts.banner = "Unofficial TxD Disk & HTTP Bandwidth Usage Meter\nUsage: rsrcmeter [options]"
  
  opts.on("-b", "--base BASE", "Change the base directory, defaults to user's $HOME", String) {|base|
    if File.exists?(base)
      opt[:base_dir] = base
    else
      raise(ArgumentError, "Directory (#{base}) specified with -b does not exist")
    end
  }
  
  opts.on("--stat-file FILE", "Change default historical data file from '#{opt[:stat_file]}'", String) {|file|
    opt[:stat_file] = file
  }
  
  opts.on("--temp-file FILE", "Change default temporary file from '#{opt[:temp_file]}'", String) {|file|
    opt[:temp_file] = file
  }
  
  opts.on("-m NUM", "Number of months' bandwidth to show prior to current month (default: #{opt[:bw_months]})", Integer) {|mons|
    opt[:bw_months] = mons if mons > 0
  }
  
  opts.on("-e", "--email ADDRESS", "E-mail results to ADDRESS *instead* of outputting to the screen", String) {|addr|
    if RFC2822::EmailAddress =~ addr
      opt[:email_to] = addr
    else
      raise(ArgumentError, "Invalid e-mail address provided to -e")
    end
  }
  
  opts.on_tail("-h", "--help", "Show this message") do
    puts opts
    exit
  end
end.parse!

Dir.chdir(opt[:base_dir])
File.delete(opt[:temp_file]) if File.exists?(opt[:temp_file])

# Look for existing monthly statistics file
if File.exists?(opt[:stat_file])
  # Load and process existing stats
  @stats = Array.new
  lines = File.readlines(opt[:stat_file])
  lines.each {|l| (@stats << [l.chomp.split(',')[0], l.chomp.split(',')[1].to_f]) unless (l.strip)[0,1] == '#'}
  @stats.sort!
  # Determine if any months need to be added
  if Date.today.month == 1
    pm_year = Date.today.year - 1
    pm_month = 12
  else
    pm_year = Date.today.year
    pm_month = Date.today.month - 1
  end
  stop_month = pm_year.to_s + (pm_month < 10 ? '0' : '') + pm_month.to_s
  proc_year = @stats.last[0][0,4].to_i
  proc_month = @stats.last[0][4,2].to_i
  proc_year += 1 if proc_month == 12
  proc_month = (proc_month == 12) ? 1 : proc_month + 1
  proc_str = proc_year.to_s + (proc_month < 10 ? '0' : '') + proc_month.to_s
  # Add any months not already calculated
  while proc_str <= stop_month
    datestr = proc_str + "??"
    File.delete(opt[:temp_file]) if File.exists?(opt[:temp_file])
    `cat logs/access_log.#{datestr} 2>/dev/null > #{opt[:temp_file]}`
    `cat domains/*/logs/access_log.#{datestr} 2>/dev/null >> #{opt[:temp_file]}`
    `zcat logs/access_log.#{datestr}.gz 2>/dev/null >> #{opt[:temp_file]}`
    `zcat domains/*/logs/access_log.#{datestr}.gz 2>/dev/null >> #{opt[:temp_file]}`
    month_usage = `cat #{opt[:temp_file]} | awk '{sum += $10} END {print sum}'`.chomp.to_i
    @stats << [proc_str, month_usage]
    # advance counting
    proc_year += 1 if proc_month == 12
    proc_month = (proc_month == 12) ? 1 : proc_month + 1
    proc_str = proc_year.to_s + (proc_month < 10 ? '0' : '') + proc_month.to_s
  end
# Statistics files doesn't exist
else
  # Look for earliest log file
  file_list = Array.new
  Dir.foreach(opt[:base_dir] + "/logs") {|i| file_list << i if /^access_log.\d{8}/ =~ i}
  if File.exists?(opt[:base_dir] + "/domains")
    domains = Array.new
    Dir.foreach(opt[:base_dir] + "/domains") {|dom| domains << dom if /^\w+\.\w+/ =~ dom}
    domains.each {|dom| Dir.foreach(opt[:base_dir] + "/domains/" + dom + "/logs") {|i| file_list << i if /^access_log.\d{8}/  =~ i}}
  end
  file_list.sort!
  unless Date.today.year.to_s + Date.today.month.to_s == file_list.first[11,6]
    # Process from earliest_log year/month forward
    earliest_log_year = file_list.first[11,4]
    earliest_log_month = file_list.first[15,2]
    @stats = Array.new
    earliest_log_year.to_i.upto(Date.today.year) do |year|
      start_month = (earliest_log_year.to_i == year) ? earliest_log_month.to_i : 1;
      start_month.upto(12) do |month|
        unless (year == Date.today.year and month >= Date.today.month)
          month = "0" + month.to_s if month < 10
          datestr = year.to_s + month.to_s + '??'
          File.delete(opt[:temp_file]) if File.exists?(opt[:temp_file])
          `cat logs/access_log.#{datestr} 2>/dev/null > #{opt[:temp_file]}`
          `cat domains/*/logs/access_log.#{datestr} 2>/dev/null >> #{opt[:temp_file]}`
          `zcat logs/access_log.#{datestr}.gz 2>/dev/null >> #{opt[:temp_file]}`
          `zcat domains/*/logs/access_log.#{datestr}.gz 2>/dev/null >> #{opt[:temp_file]}`
          month_usage = `cat #{opt[:temp_file]} | awk '{sum += $10} END {print sum}'`.chomp.to_i
          @stats << [year.to_s + month.to_s, month_usage]
        end
      end
    end
  end
end

# Write @stats back to opt[:stat_file] to record any changes
File.open(opt[:stat_file], 'w') {|f|
  f.puts "# rsrcmeter historical statistics file"
  f.puts "# Contains only completed months"
  @stats.each {|i| f.puts i[0] + ',' + i[1].to_s}
}

# Calculate bandwidth consumption for current month
datestr = Date.today.year.to_s + Date.today.month.to_s + "??"
`cat logs/access_log 2>/dev/null > #{opt[:temp_file]}`
`cat domains/*/logs/access_log 2>/dev/null >> #{opt[:temp_file]}`
`cat logs/access_log.#{datestr} 2>/dev/null >> #{opt[:temp_file]}`
`cat domains/*/logs/access_log.#{datestr} 2>/dev/null >> #{opt[:temp_file]}`
`zcat logs/access_log.#{datestr}.gz 2>/dev/null >> #{opt[:temp_file]}`
`zcat domains/*/logs/access_log.#{datestr}.gz 2>/dev/null >> #{opt[:temp_file]}`
usage = `cat #{opt[:temp_file]} | awk '{sum += $10} END {print sum}'`.chomp.to_f / 1024 / 1024
File.delete(opt[:temp_file]) if File.exists?(opt[:temp_file])
# Determine disk usage
quotaline = `quota -g | tail -n 1`
disk_usage = `echo -n "#{quotaline}" | awk '{print $2}'`.to_f
disk_quota = `echo -n "#{quotaline}" | awk '{print $3}'`.to_f
disk_percent_used = (disk_usage / disk_quota) * 100

# Output results
results = ""
results << "Disk usage: " + sprintf("%.4f", disk_usage / 1024) + " MiB (Quota: " + sprintf("%.4f", disk_quota / 1024 / 1024) +" GiB | " + sprintf("%.1f", disk_percent_used) + "% used)\n"
results << "Bandwidth:\n" + Date::ABBR_MONTHNAMES[Date.today.month] + " " + Date.today.year.to_s + ": " + sprintf("%.4f", usage) + " MiB (Month to Date)\n"
hist_usage = @stats.last(opt[:bw_months]).reverse
hist_usage.each {|h| results << Date::ABBR_MONTHNAMES[Date.parse(h[0] + "01").month] + ' ' + h[0][0,4] + ": " + sprintf("%.4f", (h[1].to_f / 1024 / 1024)) + " MiB\n"}

# We're either e-mailing or outputting
if opt[:email_to]
  t = Time.now.strftime("%a %d %B %Y %H:%M:%S %Z")
  message = <<EOM
From: TxD Resource Meter <#{ENV['USER']}-noreply@#{`/bin/hostname`.chomp}>
To: #{opt[:email_to]}
Subject: [TxD Resource Meter] #{t} report for #{ENV['USER']}
X-Mailer: rsrcmeter | http://textsnippets.com/posts/show/842

Resource Report - #{t}
#{results}
---------------
Generated by rsrcmeter
EOM
  File.open("temp-emailresult", "w") { |file| file.print message }
  `cat temp-emailresult | /usr/sbin/sendmail -t`
  File.delete("temp-emailresult")
else
  puts results
end