github_repo_statistics 2.0.8

data/lib/github_repo_statistics/github_repo_statistics.rb

@@ -0,0 +1,290 @@
+# frozen_string_literal: true
+
+require 'pry'
+require 'date'
+
+class GithubRepoStatistics
+  def initialize(directory_path:, duration_in_days:, begin_time:, debug: nil, steps: 1)
+    @directory_path = directory_path
+    @duration_in_days = duration_in_days
+    @begin_time = begin_time
+    @debug = debug
+    @steps = steps
+  end
+
+  def true?(obj)
+    obj.to_s.downcase == 'true'
+  end
+
+  def read_codeowners_file
+    raise "CODEOWNERS file does not exist under #{CODEOWNERS_PATH}" unless File.exist?(CODEOWNERS_PATH)
+
+    codeowners = {}
+    File.readlines(CODEOWNERS_PATH).each do |line|
+      next if line.strip.empty? || line.start_with?('#') # Skip comments and empty lines
+
+      parts = line.split(/\s+/)
+      directory_pattern = parts[0]
+      owner = parts[1..].map { |o| o.start_with?('@') ? o[1..] : o }.join(' ') # Remove leading '@' from team names
+      codeowners[directory_pattern] = owner
+    end
+    codeowners
+  end
+
+  def find_owners(file_path, codeowners)
+    matching_patterns = codeowners.keys.select do |pattern|
+      pattern_regex = Regexp.new("^#{Regexp.escape(pattern.sub(%r{^/+}, '').chomp('/')).gsub('\*', '.*').gsub('**',
+                                                                                                              '.*?')}")
+      file_path =~ pattern_regex
+    end
+
+    return ['unknown'] if matching_patterns.empty?
+
+    # Sort patterns by length in descending order
+    sorted_patterns = matching_patterns.sort_by(&:length).reverse
+
+    # Find the most specific matching pattern
+    best_match = sorted_patterns.find do |pattern|
+      pattern_regex = Regexp.new("^#{Regexp.escape(pattern.sub(%r{^/+}, '').chomp('/')).gsub('\*', '.*').gsub('**',
+                                                                                                              '.*?')}")
+      file_path =~ pattern_regex
+    end
+
+    codeowners[best_match].split(' ')
+  end
+
+  def handle_codeowners(file_team_map:)
+    output = "\n   *Code ownership data:*\n"
+    codeowners = read_codeowners_file
+
+    owners_data = Hash.new do |hash, key|
+      hash[key] = { directories: Hash.new do |h, k|
+        h[k] = { files: [] }
+      end, churn_count: 0 }
+    end
+
+    file_team_map.each do |file, count|
+      owners = find_owners(file, codeowners)
+      owners.each do |owner|
+        owners_data[owner][:churn_count] += count.last
+
+        dir_path = File.dirname(file)
+        owners_data[owner][:directories][dir_path][:files] << { name: File.basename(file), count: }
+      end
+    end
+
+    # Sort owners_data by total count in descending order
+    sorted_owners_data = owners_data.sort_by { |_, data| -data[:churn_count] }
+    converted_team_map = file_team_map.transform_keys { |key| File.basename(key) }
+
+    sorted_owners_data.each do |owner, data|
+      output += "\n      #{owner.split('/').last}:\n         Total Count: #{data[:churn_count]}\n"
+      data[:directories].each do |dir, dir_data|
+        output += "            Directory: #{dir}\n               Top files:\n"
+        dir_data[:files].each do |file_data|
+          next if converted_team_map[File.basename(file_data[:name])].nil?
+
+          contributors = converted_team_map[file_data[:name]]&.first&.empty? ? ['Excluded contributor'] : converted_team_map[file_data[:name]].first
+          output += "                  #{File.basename(file_data[:name])} - #{file_data[:count].last} #{contributors}}\n"
+        end
+      end
+    end
+
+    if FILE_OUTPUT
+      File.open('codeowners.txt', 'w') do |f|
+        f.puts output
+      end
+    else
+      puts output
+    end
+  end
+
+  def find_owner(file:)
+    codeowners = read_codeowners_file
+    find_owners(file, codeowners)
+  end
+
+  def count_big_files(directory_path, size: BIG_FILE_SIZE)
+    size = size.to_i
+    # Get a list of all files in the specified directory
+    files = Dir.glob(File.join(directory_path, '**', '*')).select { |file| File.file?(file) }
+
+    code_files = files.select do |f|
+      extension = File.extname(f)
+      valid_extensions = CODE_EXTENSIONS
+      valid_extensions.include?(extension)
+    end
+
+    # Initialize a counter for files that meet the criteria
+    count = 0
+    # Iterate through each file and check the line count
+    code_files.each do |file|
+      lines_count = File.foreach(file).reject { |line| line.match(%r{^\s*(//|/\*.*\*/|\s*$)}) }.count
+
+      count += 1 if lines_count > size
+    end
+
+    puts "     *Current total number of code files longer than #{size} lines:* #{count}"
+  end
+
+  def count_hotspot_lines(files)
+    code_files = files.select do |f|
+      extension = File.extname(f)
+      valid_extensions = CODE_EXTENSIONS
+      valid_extensions.include?(extension)
+    end
+
+    count = 0
+
+    code_files.each do |file|
+      lines_count = File.foreach(file).reject { |line| line.match(%r{^\s*(//|/\*.*\*/|\s*$)}) }.count
+
+      count += lines_count
+    end
+
+    puts "     *Total lines of hotspot code:* #{count}"
+  end
+
+  def filter_existing_code_files(files)
+    files.select do |f|
+      next unless File.exist?(f)
+
+      if EXCLUDED_FILES
+        excluded_patterns = EXCLUDED_FILES.split(',')
+        next if excluded_patterns.any? { |pattern| f.include?(pattern) }
+      end
+
+      extension = File.extname(f)
+      valid_extensions = CODE_EXTENSIONS
+      valid_extensions.include?(extension)
+    end
+  end
+
+  def git_files(directory_path:)
+    `git ls-tree -r --name-only $(git rev-list -1 HEAD) -- "#{directory_path}"`
+  end
+
+  def files_with_changes(directory_path:, start_date:, end_date:)
+    `git log --name-only --pretty=format:"" --since="#{start_date}" --until="#{end_date}" "#{directory_path}"`
+  end
+
+  def git_commit_count(file:, start_date:, end_date:)
+    `git log --since="#{start_date}" --until="#{end_date}" --follow -- "#{file}" | grep -c '^commit'`
+  end
+
+  def git_commit_info(file:, start_date:, end_date:)
+    `git log --pretty=format:"%s" --since="#{start_date}" --until="#{end_date}" --follow -- "#{file}"`
+  end
+
+  def analyze_changed_files(uniq_code_files_with_changes:, start_date:, end_date:)
+    all_teams = []
+    cross_teams_count = 0
+    single_ownership_teams_count = 0
+    files_changed_by_many_teams = 0
+    total_changes = 0
+    file_team_map = {}
+    uniq_code_files_with_changes.each do |file|
+      filename = File.basename(file)
+      commit_count = git_commit_count(file:, start_date:, end_date:).to_i
+      git_log = git_commit_info(file:, start_date:, end_date:).split("\n")
+      teams = git_log.map do |team|
+        team.match(/#{TEAM_REGEX}/)[0].upcase
+      end.reject { |e| EXCLUSIONS&.include?(e) }
+
+      total_changes += commit_count
+      all_teams << teams
+      teams = teams.uniq
+
+      if teams.count > 1
+        files_changed_by_many_teams += 1
+        file_team_map.merge!(file.to_s => [teams, commit_count])
+        cross_teams_count += teams.count
+      else
+        single_ownership_teams_count += 1
+      end
+
+      puts "\n#{filename} [#{commit_count}]:#{teams}\n" if @debug
+    end
+    [all_teams, cross_teams_count, single_ownership_teams_count, files_changed_by_many_teams, total_changes, file_team_map]
+  end
+
+  def filter_files(file_team_map:)
+    file_team_map.select do |file_path|
+      next unless File.exist?(file_path)
+
+      # Check if the file size is more than BIG_FILE_SIZE lines (excluding empty and commented lines)
+      File.foreach(file_path).reject { |line| line.match(%r{^\s*(//|/\*.*\*/|\s*$)}) }.count > BIG_FILE_SIZE.to_i
+    end
+  end
+
+  def contribution_message
+    duration_in_days = @duration_in_days.to_i
+    start_date = @begin_time.to_time.to_i - duration_in_days * 86_400 - 30 * 86_400
+    end_date = @begin_time.to_time.to_i - 30 * 86_400
+    git_ls = git_files(directory_path: @directory_path)
+    file_count = filter_existing_code_files(git_ls.split).count
+    all_files_with_changes = files_with_changes(directory_path: @directory_path, start_date:, end_date:).split.sort
+    code_files_with_changes = filter_existing_code_files(all_files_with_changes)
+    uniq_code_files_with_changes = code_files_with_changes.uniq
+    all_teams, cross_teams_count, single_ownership_teams_count, files_changed_by_many_teams, total_changes, file_team_map = analyze_changed_files(uniq_code_files_with_changes:, start_date:, end_date:)
+    occurrences = all_teams.flatten.compact.tally
+    sorted_occurrences = occurrences.sort_by { |element, count| [-count, element] }
+    contributors = Hash[sorted_occurrences]
+    churn_count = file_team_map.values.map { |value| value[1] }.sum
+    hotspot_changes_percentage = (churn_count.to_f / total_changes) * 100
+    # Filter files based on extension, existence and size
+    filtered_files = filter_files(file_team_map:)
+    filtered_top_touched_files = filtered_files.sort_by { |element, count| [-count.last, element] }
+
+    puts ''
+    puts "*Timeframe:* #{(@begin_time - duration_in_days).strftime('%Y-%m-%d')} to #{@begin_time.strftime('%Y-%m-%d')}"
+    puts "     *Code files with a single contributor:* #{(100 - ((files_changed_by_many_teams.to_f / file_count) * 100)).round(2)}%"
+    puts "        *Existing files changed by many teams:* #{files_changed_by_many_teams}"
+    puts "        *Current existing #{CODE_EXTENSIONS} files:* #{file_count}"
+    puts '     *Cross-Squad Dependency:*'
+    puts "        *Contributions by multiple squads to the same files:* #{cross_teams_count}"
+    puts "        *Contributions by single squads contributing to single files:* #{single_ownership_teams_count}"
+    puts "     *Hotspot Code Changes:* #{hotspot_changes_percentage.round(2)}%"
+    puts "        *Churn count(commits to files by multiple teams):* #{churn_count}"
+    puts "        *Total amount of commits:* #{total_changes}"
+    count_hotspot_lines(filtered_files.keys)
+    puts "     *#{CODE_EXTENSIONS} files with multiple contributors:* #{file_team_map.count}"
+    puts "     *#{CODE_EXTENSIONS} files exceeding #{BIG_FILE_SIZE} lines with multiple contributors:* #{filtered_top_touched_files.count}"
+    puts "     *Total amount of commits to #{CODE_EXTENSIONS} files:* #{total_changes}"
+    puts "     *Total #{CODE_EXTENSIONS} files changed:* #{uniq_code_files_with_changes.count}"
+    count_big_files(@directory_path)
+    puts "     *Current total of #{CODE_EXTENSIONS} files in the folder:* #{file_count}"
+    puts "     *Contributors:* #{contributors}"
+
+    if HOTSPOT
+      hotspot_output = "\n   *Hotspot files(#{filtered_top_touched_files.count}):*\n"
+
+      filtered_top_touched_files.each do |line|
+        hotspot_output += "\n"
+        file = line.first
+        contributors = line.last.first
+        commits = line.last.last
+        hotspot_output += "     #{file.gsub(@directory_path, '')} Contributors: #{contributors} Commits: #{commits} Owner: #{find_owner(file:)}\n"
+      end
+
+      if FILE_OUTPUT
+        File.open('hotspot.txt', 'w') do |f|
+          f.puts hotspot_output
+        end
+      else
+        puts hotspot_output
+      end
+    end
+
+    handle_codeowners(file_team_map:) if CODEOWNERS
+
+    @steps -= 1
+
+    return unless @steps.positive?
+
+    system("git checkout `git rev-list -1 --before='#{(@begin_time - duration_in_days).strftime('%B %d %Y')}' HEAD`",
+           %i[out err] => File::NULL)
+    @begin_time -= duration_in_days
+    contribution_message
+  end
+end

data/lib/github_repo_statistics/version.rb

@@ -0,0 +1,5 @@
+# frozen_string_literal: true
+
+class GithubRepoStatistics
+  VERSION = '2.0.8'
+end

data/lib/github_repo_statistics.rb

@@ -0,0 +1,8 @@
+# frozen_string_literal: true
+
+require_relative 'github_repo_statistics/version'
+
+class GithubRepoStatistics
+  class Error < StandardError; end
+  # Your code goes here...
+end

data/sig/github_repo_statistics.rbs

@@ -0,0 +1,4 @@
+class GithubRepoStatistics
+  VERSION: String
+  # See the writing guide of rbs: https://github.com/ruby/rbs#guides
+end

data/spec/fixtures/file1.swift

@@ -0,0 +1,333 @@
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+line
+// comment
+
+
+line
+line
+line
+line
+line
+line
+line
+line
+line