string_diff 0.1.1 → 0.1.2

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 3f70c7948e72d79c0b55964ae736f1fc809fea94
-  data.tar.gz: 386023934c17d1e9537ac938b0d194291fd38f01
+  metadata.gz: 8b8695203741c40b206db357d9df4c02f2bc5787
+  data.tar.gz: c8c5ef569e59e6aa86862e860e2fc3bd29cc0c2d
 SHA512:
-  metadata.gz: 9270d80153e93f2cba4654aa2363f0a572d8d616526b313b5d342cb91aa66a150180b806a217498c0836c40830e3dd001e7980b29a62dbea76a3d2a0d868a143
-  data.tar.gz: 1cf252fc68bbaa2aa3fc64cd00b5c22c019b25d5f5fd7a032c505d9198d19b2eb550899009a61916740df48068e68c11404a06fb4ab0b9b14334d955bc9cbfd2
+  metadata.gz: 27e8f9967dcd916a1b8296274a805965266502c2aff01b113fd4aa649f9ee7f1cf71d6d21d4856ffae8c3b64e69d45b286be3428f7f74e2c527e38ac6f3c5995
+  data.tar.gz: 3f650464f981076bbea98a11e799df2613db1fff523ecd79a31dcbd89644fda1d64c7e645626eb05b885a1b1a00168f46db4534f328a6756240da1fbb82f2039

data/README.md

@@ -44,6 +44,9 @@ StringDiff::Diff.new(string_1, string_2).diff
 # => "hello <span class='deletion'>world</span>"
 ```
 
+## Known Bugs
+
+The gem at this point in time does not handle words that have simply changed position in the string but otherwise unchanged.
 
 ## Development
 

data/lib/string_diff.rb

@@ -14,14 +14,47 @@ module StringDiff
       a1 = PragmaticTokenizer::Tokenizer.new(downcase: false).tokenize(string1)
       a2 = PragmaticTokenizer::Tokenizer.new(downcase: false).tokenize(string2)
 
-      construct_string(compare(a1, a2))
+      construct_string(compare(process_parens(a1), process_parens(a2)))
     end
 
     private
 
+    def process_parens(array)
+      if array.include?('(') && array.include?(')')
+        array_open_parens_indexes = array.each_index.select{|i| array[i] == "("}
+        array_closed_parens_indexes = array.each_index.select{|i| array[i] == ")"}
+
+        if array_open_parens_indexes.count == array_closed_parens_indexes.count
+          removed_count = 0
+          array_open_parens_indexes.each do |i|
+            combined_string = ""
+            combined_string += (array[i-removed_count] + array[i+1-removed_count])
+            array.delete_at(i-removed_count)
+            array.delete_at(i-removed_count)
+            array.insert(i-removed_count, combined_string)
+            removed_count += 1
+          end
+
+          array_closed_parens_indexes.each do |i|
+            combined_string = ""
+            combined_string += (array[i-(removed_count+1)] + array[i-removed_count])
+            array.delete_at(i-(removed_count+1))
+            array.delete_at(i-(removed_count+1))
+            array.insert(i-(removed_count+1), combined_string)
+            removed_count += 1
+          end
+        end
+      else
+        array
+      end
+      array
+    end
+
     def compare(array1, array2)
       deletions = array1 - array2
+      puts "deletions: #{deletions.to_s}"
       insertions = array2 - array1
+      puts "insertions: #{insertions.to_s}"
 
       process_duplicates(array1, array2)
       annotate_deletions(deletions, array1)
@@ -46,12 +79,32 @@ module StringDiff
         end
 
         if index.nil?
-          array1.insert(-1, "<span class='insertion'>#{v}</span>")
+          # Check whether or not we're dealing with an annotated deletion/insertion, or plain token
+          contains_span = array1.last.include?("<span") ? true : false
+          contains_punct_in_span = !(array1.last.scan(/(?<='>).*(?=<\/)/)[0] =~ (/[[:punct:]]/)).nil? if contains_span
+          stand_alone_punct = array1.last =~ (/[[:punct:]]/) if !contains_span
+
+          # If there is punctuation after a deletion, we need to make sure the
+          # insertion is added before the punctuation.
+          if (contains_punct_in_span || stand_alone_punct) && array1[-2].include?("<span class='deletion'")
+            array1.insert(-2, "<span class='insertion'>#{v}</span>")
+          elsif array2.find_index(v) < (PragmaticTokenizer::Tokenizer.new(downcase: false).tokenize(string1).count)
+            # Count how many insertions up to the original position
+            insertions_count = 0
+            deletions_count = 0
+            for i in 0..(array2.find_index(v)+1) do
+              insertions_count += 1 if array1[i].include?("<span class='insertion'")
+              deletions_count += 1 if array1[i].include?("<span class='deletion'")
+            end
+            array1.insert(((find_correct_index(v, array1, array2) + insertions_count + deletions_count) - 1), "<span class='insertion'>#{v}</span>")
+          else
+            # Otherwise we put it on the end.
+             array1.insert(-1, "<span class='insertion'>#{v}</span>")
+          end
         else
           array1.insert(index + 1, "<span class='insertion'>#{v}</span>")
         end
       end
-
       array1
     end
 
@@ -60,6 +113,11 @@ module StringDiff
       dup2 = array2.find_all { |e| array2.count(e) > 1 }
 
       missing_words = (dup1 - dup2).uniq
+      additional_words = (dup2 - dup1).uniq
+
+      unless additional_words.empty?
+        set_additional_duplicates_indexes(array2, additional_words)
+      end
 
       duplicate_indexs_of_array1 = []
       duplicate_indexs_of_array2 = []
@@ -79,6 +137,28 @@ module StringDiff
       end
     end
 
+    def set_additional_duplicates_indexes(array, dup)
+      @additional_indexes = array.each_index.select{|i| array[i] == dup[0]}
+    end
+
+    def find_correct_index(token, array1, array2)
+      unless @additional_indexes.nil?
+        # We need to find if the word has already been added, if so, use a later index
+        appeared_count = 0
+        array1.each do |item|
+          appeared_count += 1 if item.include?("<span class='insertion'>#{token}")
+        end
+
+        if appeared_count == 0
+          @additional_indexes[0]
+        else
+          @additional_indexes[appeared_count]
+        end
+      else
+        array2.find_index(token)
+      end
+    end
+
     def construct_string(array1)
       string = ""
 
@@ -89,12 +169,40 @@ module StringDiff
           if token.include?("<span")
             if token.scan(/(?<='>).*(?=<\/)/)[0] !~ /[[:punct:]]/ || string1.include?(" #{token.scan(/(?<='>).*(?=<\/)/)[0]}")
               string += " #{token}"
+            elsif !( token.scan(/(?<='>).*(?=<\/)/)[0] =~ (/[']/) ).nil?
+              if string.scan(/[']/).empty? || string.scan(/[(]/).empty?
+                string += " #{token}#{array1[i+1]}"
+                array1.slice!(i+1)
+              else
+                string += token
+              end
+            elsif !( token.scan(/(?<='>).*(?=<\/)/)[0] =~ (/[(]/) ).nil?
+              if string.scan(/[(]/).empty?
+                string += " #{token}#{array1[i+1]}"
+                array1.slice!(i+1)
+              else
+                string += token
+              end
             else
               string += token
             end
           else
-            if token !~ /[[:punct:]]/ || string1.include?(" #{token.scan(/(?<='>).*(?=<\/)/)[0]}")
+            if token !~ /[[:punct:]]/
               string += " #{token}"
+            elsif !( token =~ (/[']/) ).nil?
+              if string.scan(/[']/).empty?
+                string += " #{token}#{array1[i+1]}"
+                array1.slice!(i+1)
+              else
+                string += token
+              end
+            elsif !( token =~ (/[(]/) ).nil?
+              if string.scan(/[(]/).empty?
+                string += " #{token}#{array1[i+1]}"
+                array1.slice!(i+1)
+              else
+                string += token
+              end
             else
               string += token
             end
@@ -103,6 +211,5 @@ module StringDiff
       end
       string
     end
-
   end
-end
+end

data/lib/string_diff/version.rb

@@ -1,3 +1,3 @@
 module StringDiff
-  VERSION = "0.1.1"
+  VERSION = "0.1.2"
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: string_diff
 version: !ruby/object:Gem::Version
-  version: 0.1.1
+  version: 0.1.2
 platform: ruby
 authors:
 - natanio
 autorequire: 
 bindir: exe
 cert_chain: []
-date: 2016-05-19 00:00:00.000000000 Z
+date: 2016-05-20 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: bundler