Hourly request counts on an Nginx log file

gistfile1.txt

# Usage: ruby nginx-requests.rb logfilename 31 Jul 2014

logfilename = ARGV[0]
day = ARGV[1]
month = ARGV[2]
year = ARGV[3]

hours = (0..24).to_a

hours.each do |hour|
  hour_str = sprintf("%02d", hour)
  cmd = %{grep "#{day}/#{month}/#{year}:#{hour_str}:" #{logfilename} | wc -l}
  puts cmd
  result = `#{cmd}`
  puts result
end

The output here is different, and this isn't filtering for the day (although you could add that to the grep), but it only takes one pass over the input file:

cat <nginx logfile> | grep -oE '\[.*\]' | cut -c 14-15 | uniq -c