From cbb7ccb59b340089e33c9ffd4d0f6489844ed423 Mon Sep 17 00:00:00 2001
From: inga-lovinde <52715130+inga-lovinde@users.noreply.github.com>
Date: Wed, 26 Apr 2017 16:50:47 +0300
Subject: [PATCH] Refactored vector-to-words conversion to lower-level code

---
 dotnet/WhiteRabbit/PhraseSet.cs        | 10 ++++----
 dotnet/WhiteRabbit/StringsProcessor.cs | 34 +++++++++++++++++---------
 dotnet/WhiteRabbit/Word.cs             |  9 +++++++
 3 files changed, 36 insertions(+), 17 deletions(-)

diff --git a/dotnet/WhiteRabbit/PhraseSet.cs b/dotnet/WhiteRabbit/PhraseSet.cs
index 964f62c..1bad330 100644
--- a/dotnet/WhiteRabbit/PhraseSet.cs
+++ b/dotnet/WhiteRabbit/PhraseSet.cs
@@ -7,20 +7,20 @@
     {
         public long[] Buffer;
 
-        public unsafe PhraseSet(Word[] words, ulong[] permutations, int offset, int numberOfCharacters)
+        public unsafe PhraseSet(Word[] allWords, int[] wordIndexes, ulong[] permutations, int permutationOffset, int numberOfCharacters)
         {
-            Debug.Assert(numberOfCharacters + words.Length - 1 < 27);
+            Debug.Assert(numberOfCharacters + wordIndexes.Length - 1 < 27);
 
             this.Buffer = new long[4 * Constants.PhrasesPerSet];
 
             fixed (long* bufferPointer = this.Buffer)
             {
                 long* longBuffer = (long*)bufferPointer;
-                int numberOfWords = words.Length;
+                int numberOfWords = wordIndexes.Length;
 
                 fixed (ulong* permutationsPointer = permutations)
                 {
-                    var currentPermutationPointer = permutationsPointer + offset;
+                    var currentPermutationPointer = permutationsPointer + permutationOffset;
                     for (var i = 0; i < Constants.PhrasesPerSet; i++, currentPermutationPointer++)
                     {
                         var permutation = *currentPermutationPointer;
@@ -32,7 +32,7 @@
                         var cumulativeWordOffsetX4 = 0;
                         for (var j = 0; j < numberOfWords; j++)
                         {
-                            var currentWord = words[permutation & 15];
+                            var currentWord = allWords[wordIndexes[permutation & 15]];
                             permutation = permutation >> 4;
                             longBuffer[0] |= currentWord.Buffers[cumulativeWordOffsetX4 + 0];
                             longBuffer[1] |= currentWord.Buffers[cumulativeWordOffsetX4 + 1];
diff --git a/dotnet/WhiteRabbit/StringsProcessor.cs b/dotnet/WhiteRabbit/StringsProcessor.cs
index da244c0..fd82e10 100644
--- a/dotnet/WhiteRabbit/StringsProcessor.cs
+++ b/dotnet/WhiteRabbit/StringsProcessor.cs
@@ -20,18 +20,26 @@
             this.NumberOfCharacters = filteredSource.Length;
             this.VectorsConverter = new VectorsConverter(filteredSource);
 
-            // Dictionary of vectors to array of words represented by this vector
-            var vectorsToWords = words
+            var allWordsAndVectors = words
                 .Where(word => word != null && word.Length > 0)
                 .Select(word => new { word, vector = this.VectorsConverter.GetVector(word) })
                 .Where(tuple => tuple.vector != null)
-                .Select(tuple => new { tuple.word, vector = tuple.vector.Value })
+                .Select(tuple => tuple.word)
+                .Distinct(new ByteArrayEqualityComparer())
+                .Select(word => word)
+                .ToArray();
+
+            // Dictionary of vectors to array of words represented by this vector
+            var vectorsToWords = allWordsAndVectors
+                .Select((word, index) => new { word, index, vector = this.VectorsConverter.GetVector(word).Value })
                 .GroupBy(tuple => tuple.vector)
-                .Select(group => new { vector = group.Key, words = group.Select(tuple => tuple.word).Distinct(new ByteArrayEqualityComparer()).Select(word => new Word(word)).ToArray() })
+                .Select(group => new { vector = group.Key, words = group.Select(tuple => tuple.index).ToArray() })
                 .ToList();
 
             this.WordsDictionary = vectorsToWords.Select(tuple => tuple.words).ToArray();
 
+            this.AllWords = allWordsAndVectors.Select(word => new Word(word)).ToArray();
+
             this.VectorsProcessor = new VectorsProcessor(
                 this.VectorsConverter.GetVector(filteredSource).Value,
                 maxWordsCount,
@@ -40,10 +48,12 @@
 
         private VectorsConverter VectorsConverter { get; }
 
+        private Word[] AllWords { get; }
+
         /// <summary>
-        /// WordsDictionary[vectorIndex] = [word1, word2, ...]
+        /// WordsDictionary[vectorIndex] = [word1index, word2index, ...]
         /// </summary>
-        private Word[][] WordsDictionary { get; }
+        private int[][] WordsDictionary { get; }
 
         private VectorsProcessor VectorsProcessor { get; }
 
@@ -61,7 +71,7 @@
             // converting sequences of vectors to the sequences of words...
             return from sum in sums
                    let filter = ComputeFilter(sum)
-                   let wordsVariants = this.ConvertVectorsToWords(sum)
+                   let wordsVariants = this.ConvertVectorsToWordIndexes(sum)
                    from wordsArray in Flattener.Flatten(wordsVariants)
                    from phraseSet in this.ConvertWordsToPhrases(wordsArray, filter)
                    select phraseSet;
@@ -88,10 +98,10 @@
             return result;
         }
 
-        private Word[][] ConvertVectorsToWords(int[] vectors)
+        private int[][] ConvertVectorsToWordIndexes(int[] vectors)
         {
             var length = vectors.Length;
-            var words = new Word[length][];
+            var words = new int[length][];
             for (var i = 0; i < length; i++)
             {
                 words[i] = this.WordsDictionary[vectors[i]];
@@ -111,13 +121,13 @@
             return Tuple.Create(vectors.Length, result);
         }
 
-        private IEnumerable<PhraseSet> ConvertWordsToPhrases(Word[] words, uint filter)
+        private IEnumerable<PhraseSet> ConvertWordsToPhrases(int[] wordIndexes, uint filter)
         {
-            var permutations = PrecomputedPermutationsGenerator.HamiltonianPermutations(words.Length, filter);
+            var permutations = PrecomputedPermutationsGenerator.HamiltonianPermutations(wordIndexes.Length, filter);
             var permutationsLength = permutations.Length;
             for (var i = 0; i < permutationsLength; i += Constants.PhrasesPerSet)
             {
-                yield return new PhraseSet(words, permutations, i, this.NumberOfCharacters);
+                yield return new PhraseSet(this.AllWords, wordIndexes, permutations, i, this.NumberOfCharacters);
             }
         }
     }
diff --git a/dotnet/WhiteRabbit/Word.cs b/dotnet/WhiteRabbit/Word.cs
index 0ccf3b1..9ddf46d 100644
--- a/dotnet/WhiteRabbit/Word.cs
+++ b/dotnet/WhiteRabbit/Word.cs
@@ -8,6 +8,13 @@
 
         public int LengthX4 { get; }
 
+        private Word()
+        {
+            this.Original = new byte[0];
+            this.Buffers = new long[128];
+            this.LengthX4 = 0;
+        }
+
         public unsafe Word(byte[] word)
         {
             var tmpWord = new byte[word.Length + 1];
@@ -37,5 +44,7 @@
             this.Buffers = buffers;
             this.LengthX4 = tmpWord.Length * 4;
         }
+
+        private static Word Empty { get; } = new Word();
     }
 }