https://github.com/embulk/embulk/issues/284

284.rb

# Taken from embulk/lib/embulk/guess/csv.rb                                                                                                                                                                                                                                                 
def array_avg(array)
  array.inject(0.0) {|r,i| r += i } / array.size
end

def array_variance(array)
  avg = array_avg(array)
  array.inject(0.0) {|r,i| r += (i - avg) ** 2 } / array.size
end

def column_length(row, column_index)
  column_index >= row.length ? 0 : row[column_index].length
end

def guess(rows)
  column_number = rows.map do |row|
    row.length
  end.max - 1
  headers = []
  for column_index in 0..(column_number) do
    var1 = array_variance(rows.map {|row| column_length(row, column_index) })
    var2 = array_variance(rows[1..-1].map {|row| column_length(row, column_index) })
    headers << (var1 < var2)
    puts "c#{column_index}: var1=#{var1} var2=#{var2}"
  end
  puts headers.inspect
  rows.each do |row|
    puts row.inspect
  end
end

dame = [['magic', 'description'],
        ['0XDEADBEEF', 'word-fill pattern'],
        ['0xCAFEBABE', 'java byte code'],
        ['0xDEAD10CC', 'dead lock'],
        ['0xDEADFEED', 'crash report']]

sample_01 = [['id', 'account',' time', 'purchase', 'comment'],
             ['1', '32864', '2015-01-27 19:23:49', '20150127', 'embulk'],
             ['2', '14824', '2015-01-27 19:01:23', '20150127', 'embulk jruby'],
             ['3', '27559', '2015-01-28 02:20:02', '20150128', "Embulk ""csv"" parser plugin"],
             ['4', '11270', '2015-01-29 11:54:36', '20150129', 'NULL']]

guess(dame)
guess(sample_01)

result.txt

c0: var1=4.0 var2=0.0
c1: var1=7.44 var2=8.5
[false, true]
["magic", "description"]
["0XDEADBEEF", "word-fill pattern"]
["0xCAFEBABE", "java byte code"]
["0xDEAD10CC", "dead lock"]
["0xDEADFEED", "crash report"]
c0: var1=0.16000000000000006 var2=0.0
c1: var1=0.6399999999999999 var2=0.0
c2: var1=31.360000000000003 var2=0.0
c3: var1=0.0 var2=0.0
c4: var1=51.839999999999996 var2=60.75
[false, false, false, false, true]
["id", "account", " time", "purchase", "comment"]
["1", "32864", "2015-01-27 19:23:49", "20150127", "embulk"]
["2", "14824", "2015-01-27 19:01:23", "20150127", "embulk jruby"]
["3", "27559", "2015-01-28 02:20:02", "20150128", "Embulk csv parser plugin"]
["4", "11270", "2015-01-29 11:54:36", "20150129", "NULL"]