badcode

# An example of bad code, that should really be refactored.

def tap_count_ok?(csv_file)
  # could put additional csv processing in here - currently just confirming tap count. 
  csv_data = CSV.read(csv_file)
 
  # count taps by counting how many of the first column are digits
  tap_count = 0
  csv_data.each_index do |row|
    tap_count += 1 if csv_data[row][0] =~ /\d+/
  end
 
  if tap_count == 27
    27
  elsif tap_count == 24
    24
  else  
    #puts  "#{csv_file} did not have 24 or 27 taps - can not process"
    false
  end
end
 
 
def pull_zxyz_data_from_csv(csv_file)
  tap_count = tap_count_ok?(csv_file)
  csv_mod_time = File.mtime(csv_file)
  csv_data = CSV.read(csv_file)
 
  # the taps
  z_x_taps = []
  z_y_taps = []
  z_z_taps = []
  xy_x_taps = []
  xy_y_taps = []
  xy_z_taps = []
 
  if csv_file.downcase.include?('dr1_data')
    z_tap_rows = (2..9).to_a + (18..25).to_a
    xy_tap_rows = (10..17).to_a
  
    x_columns = [1,4]
    y_columns = [2,5]
    z_columns = [3,6]
  elsif csv_file.downcase.include?('dr2_data')
    z_tap_rows = (3..10).to_a + (19..26).to_a
    xy_tap_rows = (11..18).to_a
  
    x_columns = [1,2]
    y_columns = [4,5]
    z_columns = [7,8]
  else
    # default to DR2, because that's the future. 
    z_tap_rows = (3..10).to_a + (19..26).to_a
    xy_tap_rows = (11..18).to_a
  
    x_columns = [1,2]
    y_columns = [4,5]
    z_columns = [7,8]
  end
 
  # Z Tap data
  # for all of the z rows, collect the absolute values of x,y,z max and min
  (z_tap_rows).each do |row|
    x_columns.each do |column|
      z_x_taps << csv_data[row][column].to_f.abs
    end
    y_columns.each do |column|
      z_y_taps << csv_data[row][column].to_f.abs
    end
    z_columns.each do |column|
      z_z_taps << csv_data[row][column].to_f.abs
    end
  end
 
  # XY Tap Data
  # for all of the xy rows, collect the absolute values of x,y,z max and min
  (xy_tap_rows).each do |row|
    x_columns.each do |column|
      xy_x_taps << csv_data[row][column].to_f.abs
    end
    y_columns.each do |column|
      xy_y_taps << csv_data[row][column].to_f.abs
    end
    z_columns.each do |column|
      xy_z_taps << csv_data[row][column].to_f.abs
    end
  end
 
  wafer = csv_file.split('/').last.split('.').first
 
  "#{wafer},#{csv_mod_time},#{tap_count},#{z_x_taps.max},#{z_x_taps.min},#{z_y_taps.max},#{z_y_taps.min}," \
  "#{z_z_taps.max},#{z_z_taps.min},#{xy_x_taps.max},#{xy_x_taps.min},#{xy_y_taps.max}," \
  "#{xy_y_taps.min},#{xy_z_taps.max},#{xy_z_taps.min},"
 
end
 
def pull_measurement_data_from_csv(csv_file)
  csv_data = CSV.read(csv_file)
 
  #Measurement Data - no ranges, just individual values
  meas2_x_max = csv_data[27][10]
  meas2_x_rms = csv_data[27][11]
  meas2_x_area = csv_data[27][12]
 
  meas3_x_max = csv_data[27][19]
  meas3_x_rms = csv_data[27][20]
  meas3_x_area = csv_data[27][21]
 
  meas2_y_max = csv_data[28][13]
  meas2_y_rms = csv_data[28][14]
  meas2_y_area = csv_data[28][15]
 
  meas3_y_max = csv_data[28][22]
  meas3_y_rms = csv_data[28][23]
  meas3_y_area = csv_data[28][24]
 
 
  meas2_z_max = csv_data[29][16]
  meas2_z_rms = csv_data[29][17]
  meas2_z_area = csv_data[29][18]
 
  meas3_z_max = csv_data[29][25]
  meas3_z_rms = csv_data[29][26]
  meas3_z_area = csv_data[29][27]
 
  "#{meas2_x_max},#{meas2_x_rms}," \
  "#{meas2_x_area},#{meas3_x_max},#{meas3_x_rms},#{meas3_x_area},#{meas2_y_max}," \
  "#{meas2_y_rms},#{meas2_y_area},#{meas3_y_max},#{meas3_y_rms},#{meas3_y_area}," \
  "#{meas2_z_max},#{meas2_z_rms},#{meas2_z_area},#{meas3_z_max},#{meas3_z_rms},#{meas3_z_area},\n"
 
end

Some unsolicited suggestions:

Instead of csv_data[29][27] use constants to reference field names.

ADDRESS_FIELD=30
csv_data[ADDRESS_FIELD]

Later, it will be easier to read.

Turn this into two methods. The numbers 24 and 27 are specific. But counting taps is generic. Break them out.

def tap_count(csv_file)
  # could put additional csv processing in here - currently just confirming tap count. 
  csv_data = CSV.read(csv_file)

  # count taps by counting how many of the first column are digits
  tap_count = 0
  csv_data.each_index do |row|
    tap_count += 1 if csv_data[row][0] =~ /\d+/
  end
  tap_count
end 

count = tap_count(input_csv_file)
if [27,24].include?(count)
  # good!
else
  # bad