/// <summary>
/// Encode list of integers to bytes
/// </summary>
/// <param name="postings">a posting list</param>
/// <returns>encoded bytes stream</returns>
public byte[] Encoding(List <Posting> postings)
{
List <int> concat = new List <int>();
//1. Write document frequency
concat.Add(postings.Count);
int previousDocID = 0;
foreach (Posting p in postings)
{
//2. Write docID using gap
concat.Add(p.DocumentId - previousDocID); //4byte integer per docID
List <int> positions = p.Positions;
//3. Write term frequency (# of positions)
concat.Add(positions.Count); //4byte integer per term frequency
//4. Write positions using gap
int previousPos = 0;
foreach (int pos in positions)
{
concat.Add(pos - previousPos); //4byte integer per position
previousPos = pos;
}
previousDocID = p.DocumentId;
}
return(VariableBytes.Compress(concat));
}
/// <summary>
/// Encode list of integers to bytes
/// </summary>
/// <param name="value">List of integers</param>
/// <returns>bytes array</returns>
public byte[] Encoding(List <MaxPriorityQueue.InvertedIndex> queue)
{
queue.Sort(
delegate(MaxPriorityQueue.InvertedIndex Item1, MaxPriorityQueue.InvertedIndex Item2)
{
int docID1 = Item1.GetDocumentId();
int docID2 = Item2.GetDocumentId();
if (docID1 < docID2)
{
return(-1);
}
if (docID2 < docID1)
{
return(1);
}
else
{
return(0);
}
});
List <int> concat = new List <int>();
//1. Write the document frequency
concat.Add(queue.Count);
int previousDocId = 0;
int documentID;
int termFreq;
foreach (MaxPriorityQueue.InvertedIndex item in queue)
{
documentID = item.GetDocumentId();
//1. Write the document id using gaps
concat.Add(documentID - previousDocId);
termFreq = item.GetTermFreq();
//3.Write the term frequency
concat.Add(termFreq);
previousDocId = documentID;
}
return(VariableBytes.Compress(concat));
}
/// <summary>
/// Encode list of integers to bytes
/// </summary>
/// <param name="value">List of integers</param>
/// <returns>bytes array</returns>
public byte[] Encoding(List <int> value)
{
return(VariableBytes.Compress(value));
}