Initial exercise at the Developing Project Guidance through Code History Mining workshop

code_event.rb

# This was a minor addition to the provided code_event.rb class
  
  def dir_name
    File.dirname(file_name)
  end

dpg_code_mining.rb

require './reader_csv.rb'
require './analytics.rb'
require 'ap'


# ap mutated_methods

# Task discover opportunities for refactoring
# Find 95th percentile of method size?
events = read_events("../data/pry/methodevents.csv")

def method_size(e)
  if e.status == :deleted
    0
  else
    e.method_length
  end
end

# large_methods = method_events(events).group_by(&:method_name)
#                        .map {|mn, es| [mn, ((es.map { |e| method_size(e)}).reduce(&:+) / es.size.to_f) ] }
#                        .select { |mn, line_count| line_count > 15 }
#                        .sort_by(&:second)

large_methods = method_events(events).group_by(&:method_name)
                                     .map {|mn, es| [mn, (es.map { |e| method_size(e) }).select {|s| s > 15 } ]} # remove small methods or method deletions
                                     .select { |mn, sizes| sizes.size > 3 } # look at the number of changes
                                     

ap large_methods

task.rb

require './reader_csv.rb'
require './analytics.rb'
require 'ap'

events = read_events "../data/rspec/methodevents.csv"

# ap events

end_date   = ((Time.parse("2010-01-01").to_date) - 90).to_time
start_date = ((end_date.to_date) - 90).to_time

# 
relevant_method_events = method_events(events).select { |e| e.date > start_date && e.date < end_date }
deleted_methods = relevant_method_events.select { |e| e.status == :deleted }.map(&:method_name)

mutated_methods = relevant_method_events.select { |e| !deleted_methods.include?(e.method_name) }
                     .group_by(&:method_name)
                     .map {|mn, es| [mn, es.size]}
                     .select { |mn, change_count| change_count > 2 }
                     .sort_by {|mn, change_count| change_count }

ap mutated_methods