Created
July 26, 2016 18:29
-
-
Save ignasio/bcb7c4f7d14a9afd0e3438af77ede179 to your computer and use it in GitHub Desktop.
diff between two files based on lcs
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
class Diff | |
def initialize(_old,_new) | |
@old = _old | |
@new = _new | |
end | |
def results | |
@global_index = {} | |
@old.each_with_index{|v,i| @global_index[v] = i} | |
@new.each_with_index{|v,i| @global_index[v] = i} | |
result = diff(@old, @new).group_by(&:first).map{|k, v| {k => v.map(&:last).flatten}}.reduce Hash.new, :merge | |
both = @old | @new | |
ignore = [] | |
delta = 0 | |
both.each_with_index do |val, i| | |
if ignore.include? val | |
delta = delta + 1 | |
next | |
end | |
variant = result.select{|k,v| v.include?(val)} | |
show_both = variant.values.flatten.map{|v| @global_index[v]}.uniq.count == 1 | |
p "#{i+1-delta} #{variant.keys.first} #{show_both ? variant.values.flatten.map(&:strip).join('|') : val.strip} " | |
if show_both | |
variant.values.flatten.each do |x| | |
ignore.push x | |
end | |
end | |
end | |
result | |
end | |
def diff(_old, _new) | |
_old ||= [] | |
_new ||= [] | |
old_index = {} | |
_old.each_with_index do |val, i| | |
(old_index[val] ||= []).push(i) | |
end | |
conjunction = {} | |
start_old = 0 | |
start_new = 0 | |
length = 0 | |
_new.each_with_index do |val,i| | |
_conjunction = {} | |
Array(old_index[val]).each do |j| | |
_conjunction[j] = (conjunction[j-1] || 0) + 1 | |
if (_conjunction[j] > length) | |
length = _conjunction[j] | |
start_old = j - length + 1 | |
start_new = i - length + 1 | |
end | |
end | |
conjunction = _conjunction | |
end | |
if length == 0 | |
modified = [] | |
_old.each{|i| _new.each{|j| modified.push(_old.delete(i),_new.delete(j)) if @global_index[i]==@global_index[j] }} | |
return [[_old.empty? ? [] : ['-',_old]],[_new.empty? ? [] : ['+',_new]], [['*',modified]]].flatten(1).reject { |c| c.empty? } | |
else | |
return diff(_old[0,start_old],_new[0,start_new]) + [['=',_new[start_new, start_new + length -1]]] + diff(_old[start_old+length .. -1],_new[start_new+length..-1]) | |
end | |
end | |
end | |
@file1 = File.readlines "file1.txt" | |
@file2 = File.readlines "file2.txt" | |
Diff.new(@file1,@file2).results |
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
Some | |
Simple | |
Text | |
File |
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
Another | |
Text | |
File | |
With | |
Additional | |
Lines |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment