smart_diff 0.0.1

data/README.md

@@ -0,0 +1,38 @@
+# SmartDiff
+
+This is a port to Ruby of part of @yinwang0's [psydiff](https://github.com/yinwang0/psydiff). Specifically, I've ported the htmlizer part of his app, which converts a semantic diff into an html page which can be viewed in a webbrowser to make visual the information contained in the diff.  @yinwang0
+has written extensively about semantic diffing [on his blog](http://yinwang0.wordpress.com/2012/01/03/ydiff/) and also about [psydiff](http://yinwang0.wordpress.com/2013/07/06/psydiff/).
+
+I created this port to demonstrate the use of the NodeDiff classes
+I recently added to [JRubyParser](https://github.com/jruby/jruby-parser) and
+to serve as an example of using it. I should make clear that JRubyParser and
+said NodeDiff classes are entirely unrelated to psydiff, except that they do
+something quite similar (create a diff of source code based on the AST rather than comparing the text directly).
+
+This application should especially serve as an example of using the subdiff's
+produced by NodeDiff to improve the quality of the output. See `bin/smart_diff`
+.
+
+## Install
+
+```
+$ jruby -S gem install smart_diff
+```
+
+## Usage
+
+Run `smart_diff` along with the paths to two versions of a ruby source file.
+
+    smart_diff example/bar.rb example/foo.rb
+
+This will output an html file you can open in your browser. Check out `foo.rb-bar.rb.html` for a sample of the output.
+
+LICENSE
+
+Copyright (C) 2011-2013 Yin Wang
+
+This program is free software: you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
+
+This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.
+
+You should have received a copy of the GNU General Public License along with this program. If not, see http://www.gnu.org/licenses/.

data/Rakefile

@@ -0,0 +1,13 @@
+require "rspec/core/rake_task"
+require 'bundler'
+Bundler::GemHelper.install_tasks
+
+desc "Run specs"
+RSpec::Core::RakeTask.new do |r|
+  r.ruby_opts = "-J-ea -Ilib"
+end
+
+desc "Build gem"
+task :build_gem => [:spec, :build] do
+  puts "Building gem with Bundler"
+end

data/bin/smart_diff

@@ -0,0 +1,69 @@
+#!/usr/bin/env ruby
+
+require_relative "../lib/smart_diff"
+require_relative "../lib/smart_diff/htmlize"
+
+file_path1 = ARGV[0]
+file_path2 = ARGV[1]
+
+puts "Creating Diff from #{file_path1} and #{file_path2}"
+
+rd = SmartDiff.new(file_path1, file_path2)
+
+nd = rd.node_diff.to_a
+
+# Subdiffs use levenschtein distances to match up nodes. This won't work on
+# the full diff, because it has too many chances to create false positives.
+# However, within a subdiff, it can be pretty accurate because its less
+# likely to have many highly similar nodes. Here's an example of using those
+# subdiffs to improve the diffing results.
+
+puts "Examining subdiffs to improve results."
+
+# Gather up subdiffs
+
+subs = []
+
+nd.each do |d|
+  if d.subdiff
+    d.subdiff.each do |s|
+      subs << s
+    end
+  end
+end
+
+# Go through our diff, removing any diff which matches a subdiff
+# We create a few nils in the process, so we strip those out at the end
+
+complete = nd.map do |dd|
+  subs.each do |c|
+    if dd.old_node && c.old_node and !dd.new_node && c.new_node
+      if dd.old_node.isSame(c.old_node)
+        dd.old_node = nil
+      end
+    elsif dd.new_node && c.new_node and !dd.old_node && c.old_node
+      if dd.new_node.isSame(c.new_node)
+        dd.new_node = nil
+      end
+    end
+  end
+  dd unless dd.new_node == nil && dd.old_node == nil
+end
+
+complete.compact!
+
+# Add all the subdiffs onto the end of our list of diffs
+# We need them at the end so later we can make sure we are
+# not creating nested anchor tags.
+
+subs.each do |c|
+  complete << c
+end
+
+
+puts "Creating html output."
+
+html = Htmlize.new
+output = html.create_html(complete, file_path2, file_path1, rd.code_two, rd.code_one)
+
+puts "Output complete. Try opening #{output} in your browser."