diff --git a/Gemfile b/Gemfile
new file mode 100644
index 0000000..3dfb5dd
--- /dev/null
+++ b/Gemfile
@@ -0,0 +1,4 @@
+source 'https://rubygems.org'
+
+# Specify your gem's dependencies in mysql_dump_split.gemspec
+gemspec
diff --git a/Gemfile.lock b/Gemfile.lock
new file mode 100644
index 0000000..6d942ab
--- /dev/null
+++ b/Gemfile.lock
@@ -0,0 +1,17 @@
+PATH
+  remote: .
+  specs:
+    mysql_dump_split (0.0.1)
+
+GEM
+  remote: https://rubygems.org/
+  specs:
+    rake (10.3.2)
+
+PLATFORMS
+  ruby
+
+DEPENDENCIES
+  bundler (>= 1.6)
+  mysql_dump_split!
+  rake (~> 10.0)
diff --git a/LICENSE.txt b/LICENSE.txt
new file mode 100644
index 0000000..76a413f
--- /dev/null
+++ b/LICENSE.txt
@@ -0,0 +1,22 @@
+Copyright (c) 2014 R.I. Pienaar
+
+MIT License
+
+Permission is hereby granted, free of charge, to any person obtaining
+a copy of this software and associated documentation files (the
+"Software"), to deal in the Software without restriction, including
+without limitation the rights to use, copy, modify, merge, publish,
+distribute, sublicense, and/or sell copies of the Software, and to
+permit persons to whom the Software is furnished to do so, subject to
+the following conditions:
+
+The above copyright notice and this permission notice shall be
+included in all copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE
+LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
+OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
+WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
diff --git a/README.markdown b/README.markdown
index 1d0a4d6..08aa944 100644
--- a/README.markdown
+++ b/README.markdown
@@ -4,13 +4,19 @@ What is it?
 A simple script that splits a MySQL dump into lots of smaller files.
 It works both with data definitions and data only dumps.
 
+Install:
+
+```shell
+gem install mysq-dump-split
+```
+
 Usage:
 ------
 
 First you need a mysqldump file, put it into the directory you want
 all the split files in:
 
-<pre>
+```shell
 $ ruby split-mysql-dump.rb db.sql
 Found a new db: app
 Found a new table: administrator_log
@@ -18,34 +24,34 @@ Found a new table: administrator_log
 
     Found a new table: auth_strings
         writing line: 239 205.482MB in 6 seconds 34.247MB/sec
-</pre>
+```
 
 Alternatively, you can pipe in via STDIN in using '-s'. Great
 for working with large gzipped backups:
 
-<pre>
+```shell
 $ gunzip -c db.sql.gz | ruby split-mysql-dump.rb -s
-</pre>
+```
 
-You can also limit the dump to particular tables using '-t' 
-or exclude tables using '-i'. 
+You can also limit the dump to particular tables using '-t'
+or exclude tables using '-i'.
 
-<pre>
+```shell
 $ ruby split-mysql-dump.rb -t auth_strings, administrator_log db.sql
-</pre>
+```
 
 and
 
-<pre>
+```shell
 $ ruby split-mysql-dump.rb -i auth_strings
-</pre>
+```
 
 When you're done you should have lots of files like this:
 
-<pre>
+```
 -rw-r--r-- 1 rip rip   210233252 May 17 18:06 administrator_log.sql
 -rw-r--r-- 1 rip rip   215463582 May 17 18:06 auth_strings.sql
-</pre>
+```
 
 The first bit of the files will be the database that the tables are in
 based on the _USE_ statements in the dump.
diff --git a/Rakefile b/Rakefile
new file mode 100644
index 0000000..809eb56
--- /dev/null
+++ b/Rakefile
@@ -0,0 +1,2 @@
+require "bundler/gem_tasks"
+
diff --git a/bin/mysql_dump_split b/bin/mysql_dump_split
new file mode 100755
index 0000000..b624509
--- /dev/null
+++ b/bin/mysql_dump_split
@@ -0,0 +1,45 @@
+#!/usr/bin/env ruby
+
+require 'optparse'
+
+require 'mysql_dump_split'
+
+dump_split = MysqlDumpSplit.new
+
+OptionParser.new do |opts|
+  opts.banner = "Usage: split-mysql-dump.rb [options] [FILE]"
+
+  opts.on("-s", "Read from stdin") do
+    dump_split.dumpfile = $stdin
+  end
+
+  opts.on("-t", '--tables TABLES', Array, "Extract only these tables") do |t|
+    dump_split.tables = t
+  end
+
+  opts.on("-i", '--ignore-tables TABLES', Array, "Ignore these tables") do |i|
+    dump_split.ignore = i
+  end
+
+  opts.on("-u", '--use-database NAME', String, "Assume NAME as database name") do |n|
+    dump_split.use_database = n
+  end
+
+  opts.on_tail("-h", "--help") do
+    puts opts
+    exit
+  end
+
+end.parse!
+
+dump_split.dumpfile ||= ARGV.shift
+
+unless dump_split.dumpfile
+  puts "Nothing to do"
+  exit
+end
+
+STDOUT.sync = true
+
+
+dump_split.split!
diff --git a/lib/mysql_dump_split.rb b/lib/mysql_dump_split.rb
new file mode 100644
index 0000000..1831c83
--- /dev/null
+++ b/lib/mysql_dump_split.rb
@@ -0,0 +1,151 @@
+require 'mysql_dump_split/version'
+
+class Numeric
+  def bytes_to_human
+    units = %w{B KB MB GB TB}
+    e = self > 0 ? (Math.log(self)/Math.log(1024)).floor : 0
+    s = "%.3f" % (to_f / 1024**e)
+    s.sub(/\.?0*$/, units[e])
+  end
+end
+
+class MysqlDumpSplit
+  attr_accessor :tables, :ignore, :use_database
+  attr_reader :dumpfile
+
+  NEW_TABLE = [/\A-- Table structure for table .(.+)./,
+               /\A-- Dumping data for table .(.+)./,
+               /\A# Dump of table.(.+)/]
+  NEW_DB = [/\A-- Current Database: .(.+)./]
+  NEW_POSITION = [/\A-- Position to start replication or point-in-time recovery from/]
+
+  def initialize
+    @tables = []
+    @ignore = []
+    @use_database = nil
+    @dumpfile = nil
+    @print_each = 10
+  end
+
+  def dumpfile=(dumpfile)
+    @dumpfile = open(dumpfile)
+  end
+
+  def split!
+    return unless File.exist?(@dumpfile)
+    @current_db = @use_database
+
+    @line_count = @table_count = @start_time = 0
+
+    @current_table = @outfile = nil
+
+    while (line = @dumpfile.gets) do
+
+      case line
+        when *NEW_TABLE
+          is_new_table = (@current_table != $1)
+          table = $1
+
+          new_table(table) if is_new_table
+
+        when *NEW_DB
+          @current_db = $1
+          @current_table = nil
+
+          close_outfile
+
+          Dir.mkdir(@current_db)
+          Dir.mkdir("#{@current_db}/tables")
+          self.outfile = "#{@current_db}/create.sql"
+          puts("\n\nFound a new db: #{@current_db}")
+
+        when *NEW_POSITION
+          @current_db = nil
+          @current_table = nil
+
+          close_outfile
+
+          self.outfile = '1replication.sql'
+          puts("\n\nFound replication data")
+
+        else
+          write(line)
+      end
+    end
+
+    puts
+  end
+
+  def outfile=(name)
+    @outfile = File.new(name, 'w')
+  end
+
+  def write(line)
+    return unless @outfile
+    return if @outfile.closed?
+
+    @outfile.syswrite(line)
+    @line_count += 1
+
+    if @line_count % @print_each == 0
+      print_status
+    end
+  end
+
+  def print_status
+    elapsed = Time.now.to_i - @start_time.to_i + 1
+    print("    writing line: #{@line_count} #{@outfile.stat.size.bytes_to_human} in #{elapsed} seconds #{(@outfile.stat.size / elapsed).bytes_to_human}/sec                 \r")
+  end
+
+  def close_outfile
+    @outfile.close if @outfile and !@outfile.closed?
+    @outfile = nil
+  end
+
+  def included?(table)
+    @tables.empty? or @tables.include?(table)
+  end
+
+  def ignored?(table)
+    ! @ignore.empty? && @ignore.include?(table)
+  end
+
+  def new_table(table)
+    close_outfile
+
+    puts("\n\nFound a new table: #{table}")
+
+    if not included?(table)
+      puts "`#{table}` not in list, ignoring"
+    elsif ignored?(table)
+      puts "`#{table}` will be ignored"
+    else
+      @start_time = Time.now
+      @line_count = 0
+      @table_count += 1
+      path = tables_path
+      Dir.mkdir(path) unless File.exists?(path)
+      self.outfile = "#{path}/#{table}.sql"
+
+      if db = @current_db
+        @outfile.syswrite("USE `#{db}`;\n\n")
+      end
+    end
+
+    @current_table = table
+  end
+
+  def tables_path
+    [@current_db, 'tables'].compact.join('/')
+  end
+
+  private
+
+  def open(dumpfile)
+    if dumpfile == $stdin
+      $stdin
+    elsif dumpfile
+      File.new(dumpfile, "r:binary")
+    end
+  end
+end
diff --git a/lib/mysql_dump_split/version.rb b/lib/mysql_dump_split/version.rb
new file mode 100644
index 0000000..72a042d
--- /dev/null
+++ b/lib/mysql_dump_split/version.rb
@@ -0,0 +1,3 @@
+class MysqlDumpSplit
+  VERSION = "0.0.1"
+end
diff --git a/mysql_dump_split.gemspec b/mysql_dump_split.gemspec
new file mode 100644
index 0000000..f1c305c
--- /dev/null
+++ b/mysql_dump_split.gemspec
@@ -0,0 +1,23 @@
+# coding: utf-8
+lib = File.expand_path('../lib', __FILE__)
+$LOAD_PATH.unshift(lib) unless $LOAD_PATH.include?(lib)
+require 'mysql_dump_split/version'
+
+Gem::Specification.new do |spec|
+  spec.name          = 'mysql_dump_split'
+  spec.version       = MysqlDumpSplit::VERSION
+  spec.authors       = ['R.I. Pienaar']
+  spec.email         = ['rip@devco.net']
+  spec.summary       = %q{Splits a MySQL dump into lots of smaller files.}
+  spec.description   = %q{It works both with data definitions and data only dumps.}
+  spec.homepage      = 'https://github.com/ripienaar/mysql-dump-split'
+  spec.license       = 'MIT'
+
+  spec.files         = `git ls-files -z`.split("\x0")
+  spec.executables   = spec.files.grep(%r{^bin/}) { |f| File.basename(f) }
+  spec.test_files    = spec.files.grep(%r{^(test|spec|features)/})
+  spec.require_paths = ['lib']
+
+  spec.add_development_dependency 'bundler', '>= 1.6'
+  spec.add_development_dependency 'rake', '~> 10.0'
+end