Say you have a file that's been duplicated a few and you want to get an overwiew of how each copy diverged, this gives you a TSV you can import in Excel

trace_files_git_history.rb

#!/usr/bin/env ruby

`for i in market*.sh.erb; do echo $i; git log --format="%ad%x09%h%x09%s" --date=iso -- $i; done > history1.log`

lines = File.readlines('history1.log')
commits_by_file = lines.reduce(['', Hash.new([])]) { |(current_file, commits_by_file), line| line.include?("\t") ? [current_file, commits_by_file.merge(current_file => (commits_by_file[current_file] + [line]))] : [line, commits_by_file] }.last
commits_by_file_parsed = Hash[commits_by_file.map{|file, commits| [file, commits.map{|c|c.split("\t")}.map{|s|{:date => s[0], :sha => s[1], :msg => s[2]}}]}]
flat_commits = commits_by_file_parsed.flat_map{|file, commits| commits.map {|commit| commit.merge(:file => file)}}
commits = flat_commits.reduce({}) { |mem, commit| new_commit = mem[commit[:sha]] ? mem[commit[:sha]].merge(:files => (mem[commit[:sha]][:files] + [commit[:file]])) : commit.select{|k| k != :file}.merge(:files => [commit[:file]]); mem.merge(commit[:sha] => new_commit) }.values.sort_by{|c| c[:date]}
files = commits_by_file.keys
commits_with_matrix = commits.map{ |c| c.merge(:line => files.map{|f| c[:files].include?(f) ? 'x' : ''}.join("\t")) }
header = "\t\t\t\t" + files.map(&:strip).join("\t")
output_lines = commits_with_matrix.map { |c| [c[:date], c[:sha], c[:msg].strip[0..20], "https://github.com/envato/marketplace-aws/commit/#{c[:sha]}", c[:line]].join("\t") }

STDOUT.puts header; output_lines.each { |l| STDOUT.puts l }; nil