/// <summary>
/// convers line from posting file to Term object
/// </summary>
/// <param name="line">string of Posting's data</param>
/// <returns>Term, constructed from the posting-line received</returns>
public Term convertPostingStringToTerm(string line)
{
Term result = new Term();
string termString;
//string type;
Dictionary<string, StringBuilder> dLocations = new Dictionary<string, StringBuilder>();
Dictionary<string, int> dDocTf = new Dictionary<string, int>();
Dictionary<string, bool> dDocHeader = new Dictionary<string, bool>();
//extract term string
int termEndIndex = line.IndexOf('@');
termString = line.Substring(0, termEndIndex);
//cut the term string
line = line.Substring(termEndIndex + 1);
//extract documents and positions
string[] docsDivisionArray = line.Split('|');
foreach (string docString in docsDivisionArray)
{
if (docString.Length < 1)
continue;
int currLoc, docTf;//, tfCount = 0;
string docName;
Boolean isHeader;
StringBuilder sb = new StringBuilder();
//extracting the docName
currLoc = docString.IndexOf(';');
docName = docString.Substring(0, currLoc);
//removing docName
string docLocationsString = docString.Substring(currLoc + 1);
//extracting isHeader
currLoc = docLocationsString.IndexOf('&');
string tempHeader = docLocationsString.Substring(0, currLoc);
if (tempHeader == "H")
isHeader = true;
else
isHeader = false;
//removing isHeader
docLocationsString = docLocationsString.Substring(currLoc + 1);
//extracting docTf
currLoc = docLocationsString.IndexOf(':');
docTf = Int32.Parse(docLocationsString.Substring(0, currLoc));
//removing docTf
docLocationsString = docLocationsString.Substring(currLoc + 1);
string[] locationsDivisionArray = docLocationsString.Split(',');
//iterating over locations
foreach (string locString in locationsDivisionArray)
{
if (locString.Length < 1)
continue;
sb.Append(locString + ",");
//tfCount++;
}
//updating dLocations and dDocTf with data
dLocations[docName] = sb;
dDocTf[docName] = docTf;
dDocHeader[docName] = isHeader;
}
result.termString = termString;
result.d_locations = dLocations;
result.d_docTf = dDocTf;
result.d_docHeader = dDocHeader;
return result;
}
/// <summary>
/// adds a term to dictionary
/// </summary>
/// <param name="d_terms">dictionary to add to</param>
/// <param name="term">term string to add</param>
/// <param name="docName">docName</param>
/// <param name="index">index of term</param>
/// <param name="numOfTerms">numOfTerms in doc</param>
/// <param name="type">term type</param>
private static void addTermToDic(SortedDictionary<string, Term> d_terms, string term, string docName, int index, ref int numOfTerms, string type)
{
lock (d_terms)
{
if (!d_terms.ContainsKey(term))
{
d_terms[term] = new Term(type, term);
numOfTerms++;
}
if (type[0] != 'H')
{
d_terms[term].addPosition(docName, index);
}
else
{
d_terms[term].addPosition(docName, index);
d_terms[term].d_docHeader[docName] = true;
}
}
int tfDoc;
// change max tf in doc if needed
Doc doc = d_docs[docName];
if (doc.d_TermsCount.ContainsKey(term))
{
tfDoc = ++doc.d_TermsCount[term];
}
else
{
doc.d_TermsCount[term] = 1;
tfDoc = 1;
}
if (doc.maxtfCount < tfDoc)
{
doc.maxtfString = term;
doc.maxtfCount = tfDoc;
}
}
