public void Decode(string fileNameSource, string fileNameDecode)
{
BitArray bitArray = null;
try
{
bitArray = new BitArray((File.ReadAllBytes(fileNameSource)));
}
catch (Exception exc)
{
throw exc;
}
//считываем длину битов декодировочной таблицы
BitArray numBitCodeTable = new BitArray(32);
for (int i = 0; i < 32; i++)
{
numBitCodeTable[i] = bitArray[i];
}
byte[] num = new byte[4];
numBitCodeTable.CopyTo(num, 0);
int lenTable = BitConverter.ToInt32(num, 0);
Frequencies = GetFrequencyDictionaryFromFile(bitArray, 32, lenTable + 32);
BuildTree(Frequencies);
Decode(bitArray, lenTable + 32, fileNameDecode);
}
public HuffmanTree(FrequencyDictionary frequencyDictionary)
{
Frequencies = new FrequencyDictionary();
Frequencies = frequencyDictionary;
var startTime = System.Diagnostics.Stopwatch.StartNew();
BuildTree(Frequencies);
startTime.Stop();
var resultTime = startTime.Elapsed;
string elapsedTime = String.Format("{0:00}:{1:00}:{2:00}.{3:000}",
resultTime.Hours,
resultTime.Minutes,
resultTime.Seconds,
resultTime.Milliseconds);
Console.WriteLine("Build tree: " + elapsedTime);
startTime = System.Diagnostics.Stopwatch.StartNew();
BuildCodeTable(Frequencies);
startTime.Stop();
resultTime = startTime.Elapsed;
elapsedTime = String.Format("{0:00}:{1:00}:{2:00}.{3:000}",
resultTime.Hours,
resultTime.Minutes,
resultTime.Seconds,
resultTime.Milliseconds);
Console.WriteLine("Build code table: " + elapsedTime);
//BuildCodeTable(Frequencies);
}
public void BuildCodeTable(FrequencyDictionary frequencyDictionary)
{
CodeTable = new Dictionary <string, List <bool> >();
foreach (var wordNode in nodesDictionary)
{
List <bool> encodedWord = GetCode(wordNode.Value); //Root.Traverse(word.Key, new List<bool>());
CodeTable.Add(wordNode.Key, encodedWord);
}
}
/// <summary>
/// Получить вектор с честотами для блока слов
/// </summary>
/// <param name="frequencyDictionary">Словарь всех слов текста</param>
/// <param name="words">Блок слов</param>
/// <returns>Вектор, с частотами слов в блоке</returns>
public static Dictionary <string, double> GetVectorFrequencyFromBlock(FrequencyDictionary frequencyDictionary, List <string> words)
{
//Получаем нулевой вектор
Dictionary <string, double> vector = GetVectorZerosFromFrequencyDictionary(frequencyDictionary);
//подсчитываем количество слов
foreach (var word in words)
{
vector[word] += 1;
}
//получаем частоту
foreach (var item in words)
{
vector[item] /= words.Count;
}
return(vector);
}
public void BuildTree(FrequencyDictionary frequencyDictionary)
{
foreach (KeyValuePair <string, int> word in frequencyDictionary.Dictionary)
{
nodes.Add(new NodeHT()
{
Word = word.Key, Frequency = word.Value
});
nodesDictionary.Add(word.Key, nodes.Last());
}
while (nodes.Count > 1)
{
List <NodeHT> orderedNodes = nodes.OrderBy(node => node.Frequency).ToList <NodeHT>();
if (orderedNodes.Count >= 2)
{
// Take first two items
List <NodeHT> taken = orderedNodes.Take(2).ToList();
// Create a parent node by combining the frequencies
NodeHT parent = new NodeHT()
{
Word = "",
Frequency = taken[0].Frequency + taken[1].Frequency,
Left = taken[0],
Right = taken[1]
};
taken[0].Parent = parent;
taken[1].Parent = parent;
nodes.Remove(taken[0]);
nodes.Remove(taken[1]);
nodes.Add(parent);
}
Root = nodes.FirstOrDefault();
Root.Parent = null;
}
}
/// <summary>
/// Получение нулевого вектора, для текста
/// </summary>
/// <param name="frequencyDictionary">Частотный словарь по всему тексту</param>
/// <returns>Нулевой вектор, где каждое слово ассоциируется с конкретной позицией в векторе</returns>
public static Dictionary <string, double> GetVectorZerosFromFrequencyDictionary(FrequencyDictionary frequencyDictionary)
{
Dictionary <string, double> vector = new Dictionary <string, double>();
//Отсортируем по частоте встречания слова
frequencyDictionary.SortDescValue();
//Вормируем ассоциативный 0-вектор
foreach (var item in frequencyDictionary.Dictionary)
{
vector.Add(item.Key, 0.0);
}
return(vector);
}
public HuffmanTree()
{
nodes = new List <NodeHT>();
Root = null;
Frequencies = new FrequencyDictionary();
}
public List <bool> GetBitArrayTable(FrequencyDictionary frequencyDictionary)
{
List <bool> res = new List <bool>();
byte lenBitWord = 0x00;
foreach (var item in frequencyDictionary.Dictionary)
{
lenBitWord = (byte)item.Key.Length;
for (int i = 0; i < 8; i++)
{
if (((lenBitWord << i) & maskForEncode) == 0x00)
{
res.Add(false);
}
else
{
res.Add(true);
}
}
byte[] wordByteArray = Encoding.Default.GetBytes(item.Key);
foreach (var symbol in wordByteArray)
{
for (int i = 0; i < 8; i++)
{
if ((byte)((symbol << i) & maskForEncode) == 0x00)
{
res.Add(false);
}
else
{
res.Add(true);
}
}
}
//выполняем реверс
byte[] buf = BitConverter.GetBytes(item.Value);
byte[] freq = new byte[4];
freq = buf;
//for (int i = 0; i < buf.Length; i++)
//{
// freq[i] = buf[buf.Length - 1 - i];
//}
foreach (var b in freq)
{
for (int i = 0; i < 8; i++)
{
if (((b << i) & maskForEncode) == 0x00)
{
res.Add(false);
}
else
{
res.Add(true);
}
}
}
}
//if(codeTableBit.Count % 8 == 0)
return(res);
}
