Created
August 14, 2011 10:27
-
-
Save sriranggd/1144775 to your computer and use it in GitHub Desktop.
Ruby script to read a git index file and write the contents in human readable form
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
#!/usr/bin/env ruby | |
# ************************************************************************ | |
# A simple ruby script to read the index file of any git repository | |
# and dump its contents to a file in a readable format. | |
# | |
# This is totally a study tool, written when I was trying to understand | |
# the git internals. | |
# | |
# The file strucutre specification considered is this : | |
# http://git.rsbx.net/Documents/Git_Data_Formats.txt | |
# | |
# Note that this doesn't care about the INDEX_ENTRY_EXTENSIONS part. | |
# | |
# Author : Srirang G Doddihal (Brahmana) | |
# | |
# Feel free to use this and modify it, with no restrictions what-so-ever. | |
# | |
# ************************************************************************ | |
require 'pp' | |
EntryTime = Struct.new(:time, :nsec) do | |
def initialize(arr) | |
self.time = Time.at(arr[0]) | |
self.nsec = arr[1] | |
end | |
end | |
EntryStatInfo = Struct.new(:ctime, :mtime, :device, :inode, :mode, :uid, :gid, :size) do | |
def initialize(arr) | |
self.ctime = EntryTime.new(arr.slice(0, 2)) | |
self.mtime = EntryTime.new(arr.slice(2, 2)) | |
(4..9).each do |i| | |
self[members[i - 2]] = arr[i] | |
end | |
end | |
end | |
EntryFlags = Struct.new(:name_length, :stage) do | |
def initialize(num) | |
self.name_length = num & 0x0fff | |
self.stage = num & 0x3000 | |
end | |
end | |
IndexEntry = Struct.new(:stat_info, :id, :flags, :name) do | |
def initialize(arr) | |
self.stat_info = EntryStatInfo.new(arr.slice(0, 10)) | |
self.id = arr[10] | |
self.flags = EntryFlags.new(arr[11]) | |
end | |
end | |
def index_file_reader(path) | |
iFile = File.open(path, 'rb') | |
oFile = File.open(path + '_dechiphered_for_study', 'w') | |
result = {} | |
result[:header] = iFile.read(4) # Reads the string "DIRC" | |
result[:version] = iFile.read(4).unpack("N")[0] | |
index_entry_count = iFile.read(4).unpack("N")[0] | |
result[:index_entry_count] = index_entry_count | |
oFile.puts ("Header :") | |
oFile.puts (result.pretty_inspect) | |
oFile.puts("==========================================================================") | |
count = 0 | |
index_entry_count.times do |i| | |
# The following commented lines of code are elegant, but the pretty_inspect output | |
# of these strucutres involved isn't really pretty and writing custom pretty_print | |
# for each of those is a pain. So the non-elegant Hash based code gives better looking | |
# output. Conceptually they are still the same, though the hash based code involves | |
# more file reads. | |
=begin | |
entry = IndexEntry.new(iFile.read(62).unpack("N10H40n")) | |
entry.name = iFile.read(entry.flags.name_length) | |
excess_fill = entry[:flags][:name_length] % 8 | |
iFile.read(8 - excess_fill ) if (excess_fill > 0 ) | |
=end | |
entry = {} | |
############ STAT_INFO ############# | |
stat_info = {} | |
stat_info[:ctime] = {} | |
stat_info[:ctime][:time] = Time.at(iFile.read(4).unpack("N")[0]).to_s | |
stat_info[:ctime][:nsec] = iFile.read(4).unpack("N")[0] | |
stat_info[:mtime] = {} | |
stat_info[:mtime][:time] = Time.at(iFile.read(4).unpack("N")[0]).to_s | |
stat_info[:mtime][:nsec] = iFile.read(4).unpack("N")[0] | |
stat_info[:dev] = iFile.read(4).unpack("N")[0] | |
stat_info[:inode] = iFile.read(4).unpack("N")[0] | |
stat_info[:mode] = iFile.read(4).unpack("N")[0] | |
stat_info[:uid] = iFile.read(4).unpack("N")[0] | |
stat_info[:gid] = iFile.read(4).unpack("N")[0] | |
stat_info[:size] = iFile.read(4).unpack("N")[0] | |
entry[:stat_info] = stat_info | |
#entry[:id] = iFile.read(20).each_byte.map { |a| s = a.to_s(16); s = "0"+s if (s.length == 1); s }.join | |
entry[:id] = iFile.read(20).unpack("H40")[0] | |
entry[:flags] = {} | |
flags = iFile.read(2).unpack("n")[0] | |
entry[:flags][:name_length] = flags & 0x0fff | |
entry[:flags][:stage] = flags & 0x3000 | |
entry[:name] = iFile.read(entry[:flags][:name_length]) | |
# Skip the zero-padding, if any | |
excess_fill = entry[:flags][:name_length] % 8 | |
iFile.read(8 - excess_fill ) if (excess_fill > 0 ) | |
count+= 1 | |
oFile.puts("Entry #{count}") | |
oFile.puts (entry.pretty_inspect) | |
oFile.puts("==========================================================================") | |
end | |
iFile.close | |
oFile.close | |
rescue StandardError => e | |
puts e.message | |
exit | |
end | |
if (ARGV.length < 1) | |
puts "Usage : #{__FILE__ } <path-to-index-file>" | |
exit | |
end | |
if (ARGV.length > 1) | |
puts "Warning : Only the first argument is used as the path to the index file to be read. Others are discarded" | |
end | |
index_file_reader(ARGV[0]) | |
puts "The contents of the index file are saved at #{ARGV[0]}_deciphered_for_study" |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Header : | |
{:header=>"DIRC", :version=>2, :index_entry_count=>1} | |
========================================================================== | |
Entry 1 | |
{:stat_info=> | |
{:ctime=>{:time=>"2011-08-14 01:29:03 +0530", :nsec=>780046898}, | |
:mtime=>{:time=>"2011-08-14 01:29:03 +0530", :nsec=>780046898}, | |
:dev=>2049, | |
:inode=>937315, | |
:mode=>33188, | |
:uid=>1000, | |
:gid=>1000, | |
:size=>120}, | |
:id=>"d4639d8ddadbc592222efc3b42f705b7e369ef58", | |
:flags=>{:name_length=>8, :stage=>0}, | |
:name=>"file1.rb"} | |
========================================================================== |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment