public List<WordOccurrenceNode> GetWordOccurrencies(Word word)
{
invertedfileName = GetFileName(word.WordID);
List<WordOccurrenceNode> result = new List<WordOccurrenceNode>();
try
{
//open the file
br = new BinaryReader(new FileStream(invertedfileName, FileMode.Open));
//reading the file
for (int i = 0; (i < conf.MaxResultList) && (br.BaseStream.Position < br.BaseStream.Length); i++)
{
int tempDocumentHashOne = br.ReadInt32();
int hitsCount = br.ReadInt32();
WordOccurrenceNode node = new WordOccurrenceNode();
node.Hits = new List<WordHit>();
for (int y = 0; y < hitsCount; y++)
{
WordHit hit = new WordHit();
hit.Position = br.ReadInt32();
node.Hits.Add(hit);
}
node.Word = word;
node.QuantityHits = hitsCount;
node.Doc = this.docIndex.Search(tempDocumentHashOne);
result.Add(node);
}
return result;
}
catch (IOException e)
{
throw e;
}
finally
{
br.Close();
}
}
public Hashtable GetPostingList()
{
Hashtable postingList = new Hashtable();
string text = this.GetText();
string[] splitWords = text.Split(' ');
this.WordQuantity = splitWords.Length + 1;
//index words
for (int i = 0; i < splitWords.Length; i++)
{
string wordTmp = QueryParser.GetCleanQuery(splitWords[i]);
wordTmp = wordTmp.Replace(" ", string.Empty);
int key = wordTmp.GetHashCode();
//get frequency for each document word
if (postingList.ContainsKey(key))
{
WordOccurrenceNode node = postingList[key] as WordOccurrenceNode;
WordHit newhit = new WordHit();
newhit.Position = i;
node.Hits.Add(newhit);
}
else if(!string.IsNullOrEmpty(wordTmp))
{
WordOccurrenceNode newNode = new WordOccurrenceNode();
newNode.Word = new Word();
newNode.Word.WordID = key;
newNode.Word.Text = wordTmp;
newNode.Doc = this;
WordHit newhit = new WordHit();
newhit.Position = i;
//define frequency
newNode.Hits.Add(newhit);
postingList.Add(key, newNode);
}
}
GC.ReRegisterForFinalize(text);
GC.ReRegisterForFinalize(splitWords);
GC.Collect();
return postingList;
}
