// rank for eachDoc
private Dictionary <string, double> calculatingDocumentsBM25(Dictionary <string, LinkedList <WordsConnected> > dic)
{
Dictionary <string, double> array_doc = new Dictionary <string, double>();
double average_doc = m_average_lenght_doc / m_doc_dic.Count;
int numberOfDoc = m_doc_dic.Count;
foreach (string doc_name in dic.Keys)
{
LinkedList <WordsConnected> Word_D = dic[doc_name];
int countWord = 0;
double doc_final = 0;
double K = 0;
double k1 = 1.1;
double b = 0;
double k2 = 1000;
K = k1 * ((1 - b) + b * (m_doc_dic[doc_name].DOC_LENGTH));
double part1 = 0;
double part2 = 0;
double part3 = 0;
while (countWord < Word_D.Count)
{
WordsConnected first = Word_D.First();
Word_D.RemoveFirst();
part1 = (0.5 / 0.5) / ((m_terms_dictionary[first.NAME].DF + 0.5) / (numberOfDoc - m_terms_dictionary[first.NAME].DF + 0.5));
part1 = Math.Log(part1);
part2 = ((k1 + 1) * first.LOCATION) / (K + first.LOCATION);
part3 = ((k2 + 1) * m_words_in_query[first.NAME]) / (k2 + m_words_in_query[first.NAME]);
doc_final = doc_final + (part1 * part2 * part3);
part1 = 0;
part2 = 0;
part3 = 0;
countWord++;
Word_D.AddLast(first);
}
array_doc.Add(doc_name, doc_final);
doc_final = 0;
}
return(array_doc);
}
// CosSim
private Dictionary <string, double> calculatingDocumentsCosSim()
{
Dictionary <string, double> array_doc = new Dictionary <string, double>();
Dictionary <string, double> idf_words = new Dictionary <string, double>();
double wcalcu = 0;
int number_doc = m_doc_dic.Count();
foreach (string word in m_words_in_query.Keys)
{
if (m_terms_dictionary.ContainsKey(word))
{
wcalcu = number_doc / m_terms_dictionary[word].DF;
if (!idf_words.ContainsKey(word))
{
idf_words[word] = wcalcu;
}
}
}
double cosSimD = 0;
foreach (string doc in m_query_in_RelevantDoc.Keys)
{
LinkedList <WordsConnected> Word_D = m_query_in_RelevantDoc[doc];
int countWord = 0;
while (countWord < Word_D.Count)
{
WordsConnected first = Word_D.First();
Word_D.RemoveFirst();
cosSimD = cosSimD + idf_words[first.NAME];
Word_D.AddLast(first);
countWord++;
}
cosSimD = cosSimD / ((m_words_in_query.Count) * (m_doc_dic[doc].DOC_LENGTH));
// cosSimD = cosSimD / (Math.Sqrt(m_words_in_query.Count) * Math.Sqrt(m_doc_dic[doc].DOC_LENGTH));
array_doc.Add(doc, cosSimD);
cosSimD = 0;
}
return(array_doc);
}
// LInked list of - [ for each document : words + count ]
public Dictionary <string, LinkedList <WordsConnected> > getTheDocument(Dictionary <int, string[]> query_list, bool Dostemming, string path_posting)
{
//number of doc + average of the length of all doc
Dictionary <string, int> words_in_query = new Dictionary <string, int>();
int number_of_doc = m_doc_dic.Count;
double average = m_average_lenght_doc / number_of_doc;
Dictionary <int, List <string> > list_of_relevet = new Dictionary <int, List <string> >();
Dictionary <string, LinkedList <WordsConnected> > doc_to_calculate = new Dictionary <string, LinkedList <WordsConnected> >();
FileStream fs1, fs2;
if (Dostemming == true)/////////////////////////////////////////////
{
fs1 = new FileStream(path_posting + "/DictionaryStemming", FileMode.Open, FileAccess.Read);
fs2 = new FileStream(path_posting + "/PostingFileStemming", FileMode.Open, FileAccess.Read);
}
else
{
fs1 = new FileStream(path_posting + "/Dictionary", FileMode.Open, FileAccess.Read);
fs2 = new FileStream(path_posting + "/PostingFile", FileMode.Open, FileAccess.Read);
}
StreamReader sr1 = new StreamReader(fs1); //read dictionary
StreamReader sr2 = new StreamReader(fs2); //read posting
//get the parameters to the BM25
foreach (int query_number in query_list.Keys)
{
string[] wordsQuery = query_list[query_number];
int length_query = wordsQuery.Length;
Array.Sort(wordsQuery); // check if sort //////////////////////////////
//count the times of each word in the query
for (int i = 0; i < wordsQuery.Length; i++)
{
if (!words_in_query.ContainsKey(wordsQuery[i]))
{
words_in_query[(wordsQuery[i])] = 1;
}
else
{
words_in_query[(wordsQuery[i])]++;
}
}
m_words_in_query = words_in_query;
foreach (string word in words_in_query.Keys)
{
//int qfi = words_in_query[word];
//int word_df = dic_to_use[word].DF;
//find the posting of the word
string line1 = sr1.ReadLine(); //read dictionary
string line2 = sr2.ReadLine(); //read posting
int index;
string posting;
while (line1 != null && line2 != null)
{
index = line1.IndexOf("/");
line1 = line1.Substring(0, index);
//find the word
if (line1 == word)
{
posting = line2;
while (posting.Contains(";"))
{
index = posting.IndexOf(",");
string doc_number = posting.Substring(0, index);
posting = posting.Substring(index + 1);
index = posting.IndexOf(";");
string appear_number = posting.Substring(0, index);
int i3;
int appear = 0;
if (Int32.TryParse(appear_number, out i3))
{
appear = i3;
}
posting = posting.Substring(index + 1);
WordsConnected to_add = new WordsConnected(word, appear);
// check if the key alreday exist
if (doc_to_calculate.ContainsKey(doc_number))
{
doc_to_calculate[doc_number].AddFirst(to_add);
}
else // not exist
{
if (m_list_languages_pressed.Contains(m_doc_dic[doc_number].LANGUAGE.ToLower()) || m_list_languages_pressed.Count == 0)
{
doc_to_calculate.Add(doc_number, new LinkedList <WordsConnected>());
doc_to_calculate[doc_number].AddFirst(to_add);
}
}
}
line1 = null;
line2 = null;
}
else
{
line1 = sr1.ReadLine();
line2 = sr2.ReadLine();
}
}
// reset the postion of the line to the start
sr1.Close();
sr2.Close();
fs1.Close();
fs2.Close();
if (Dostemming == true)/////////////////////////////////////////////
{
fs1 = new FileStream(path_posting + "/DictionaryStemming", FileMode.Open, FileAccess.Read);
fs2 = new FileStream(path_posting + "/PostingFileStemming", FileMode.Open, FileAccess.Read);
}
else
{
fs1 = new FileStream(path_posting + "/Dictionary", FileMode.Open, FileAccess.Read);
fs2 = new FileStream(path_posting + "/PostingFile", FileMode.Open, FileAccess.Read);
}
sr1 = new StreamReader(fs1);
sr2 = new StreamReader(fs2);
}
}
sr1.Close();
sr2.Close();
fs1.Close();
fs2.Close();
// Dictionary<string, double> doc_rank = new Dictionary<string, double>();
// Dictionary<string, double> doc_rank_sim = new Dictionary<string, double>();
// m_query_in_RelevantDoc = doc_to_calculate; <= was before
return(doc_to_calculate);
// doc_rank_sim = calculatingDocumentsCosSim();
//doc_rank = calculatingDocumentsBM25(average_lenght_doc);
// return doc_rank;
}
// parse the terms
public void parse(Dictionary <int, string[]> d, string path_stopWords, string path_newFile, string path_docs, bool Dostemming, bool Doindexer, string path_wordsConnected, int total_lenght_doc, Dictionary <string, Document> doc_dic, Dictionary <string, Term> terms_dictionary, ArrayList list_languages_pressed, string path_queries_ranked)
{
Dictionary <int, string[]> dicOfTerms = new Dictionary <int, string[]>();
Dictionary <string, ArrayConnectedwords> final_connectWords = new Dictionary <string, ArrayConnectedwords>();
LinkedList <WordsConnected> list_wordsCOnnected = new LinkedList <WordsConnected>();
ArrayList documents = new ArrayList(); // create a list of documnets
char[] delimiters = { ' ' };
HashSet <string> stopWords = new HashSet <string>();
stopWords = StopWords(path_stopWords);
//create month dictionary
Dictionary <string, string> MonthDic = new Dictionary <string, string>();
MonthDic = CreateDic(MonthDic);
//int how = 1;
foreach (int NumOfFile in d.Keys)
{
string[] words = d[NumOfFile][1].Split(delimiters);
string[] AfterParse = new String[words.Length * 2];
//delete from the end the ; . , : ) *
for (int i = 0; i < words.Length; i++)
{
if (((words[i] != "") && (words[i].ToLower() != "u.s.")) && ((words[i].Substring(words[i].Length - 1, 1) == "-" || words[i].Substring(words[i].Length - 1, 1) == "!" || words[i].Substring(words[i].Length - 1, 1) == "?" || words[i].Substring(words[i].Length - 1, 1) == "\"" || words[i].Substring(words[i].Length - 1, 1) == ";" || words[i].Substring(words[i].Length - 1, 1) == "." || words[i].Substring(words[i].Length - 1, 1) == "," || words[i].Substring(words[i].Length - 1, 1) == ":" || words[i].Substring(words[i].Length - 1, 1) == ")" || words[i].Substring(words[i].Length - 1, 1) == "*")))
{
words[i] = words[i].Substring(0, words[i].Length - 1);
}
}
int counterWords = 0;
int counterAfterParse = 0;
double i1;
double i2;
bool CheckIFEnter = false;
while (counterWords < words.Length)
{
//delete the , and the . , ; * ) : from the end of the string
while ((words[counterWords] != "" && (words[counterWords].ToLower() != "u.s.")) && (words[counterWords].Substring(words[counterWords].Length - 1, 1) == "]" || words[counterWords].Substring(words[counterWords].Length - 1, 1) == "{" || words[counterWords].Substring(words[counterWords].Length - 1, 1) == "(" || words[counterWords].Substring(words[counterWords].Length - 1, 1) == "," || words[counterWords].Substring(words[counterWords].Length - 1, 1) == "'" || words[counterWords].Substring(words[counterWords].Length - 1, 1) == "|" || words[counterWords].Substring(words[counterWords].Length - 1, 1) == "`" || words[counterWords].Substring(words[counterWords].Length - 1, 1) == "\"" || words[counterWords].Substring(words[counterWords].Length - 1, 1) == "." || words[counterWords].Substring(words[counterWords].Length - 1, 1) == ";" || words[counterWords].Substring(words[counterWords].Length - 1, 1) == "*" || words[counterWords].Substring(words[counterWords].Length - 1, 1) == "?" || words[counterWords].Substring(words[counterWords].Length - 1, 1) == ")" || words[counterWords].Substring(words[counterWords].Length - 1, 1) == ":"))
{
words[counterWords] = words[counterWords].Substring(0, words[counterWords].Length - 1);
}
//delete * " ( from the beginning
while ((words[counterWords] != "") && ((words[counterWords].Substring(0, 1) == "(") || words[counterWords].Substring(0, 1) == "\\" || words[counterWords].Substring(0, 1) == "}" || words[counterWords].Substring(0, 1) == ")" || words[counterWords].Substring(0, 1) == "@" || words[counterWords].Substring(0, 1) == ";" || words[counterWords].Substring(0, 1) == ":" || words[counterWords].Substring(0, 1) == "!" || words[counterWords].Substring(0, 1) == "%" || words[counterWords].Substring(0, 1) == "=" || words[counterWords].Substring(0, 1) == "]" || words[counterWords].Substring(0, 1) == ")" || words[counterWords].Substring(0, 1) == "+" || words[counterWords].Substring(0, 1) == "|" || words[counterWords].Substring(0, 1) == "'" || (words[counterWords].Substring(0, 1) == ".") || words[counterWords].Substring(0, 1) == "`" || (words[counterWords].Substring(0, 1) == ",") || (words[counterWords].Substring(0, 1) == "?") || (words[counterWords].Substring(0, 1) == "&") || (words[counterWords].Substring(0, 1) == "[") || words[counterWords].Substring(0, 1) == "/" || words[counterWords].Substring(0, 1) == "_" || (words[counterWords].Substring(0, 1) == "\"") || (words[counterWords].Substring(0, 1) == "-") || (words[counterWords].Substring(0, 1) == "*")))
{
words[counterWords] = words[counterWords].Substring(1, words[counterWords].Length - 1);
}
// delete foe the next word . , ; * ) : from the end of the string
if ((counterWords + 1 < words.Length))
{
counterWords++;
//delete the , and the . , ; * ) : from the end of the string
while ((words[counterWords] != "" && (words[counterWords].ToLower() != "u.s.")) && (words[counterWords].Substring(words[counterWords].Length - 1, 1) == "]" || words[counterWords].Substring(words[counterWords].Length - 1, 1) == "," || words[counterWords].Substring(words[counterWords].Length - 1, 1) == "'" || words[counterWords].Substring(words[counterWords].Length - 1, 1) == "|" || words[counterWords].Substring(words[counterWords].Length - 1, 1) == "`" || words[counterWords].Substring(words[counterWords].Length - 1, 1) == "\"" || words[counterWords].Substring(words[counterWords].Length - 1, 1) == "." || words[counterWords].Substring(words[counterWords].Length - 1, 1) == ";" || words[counterWords].Substring(words[counterWords].Length - 1, 1) == "*" || words[counterWords].Substring(words[counterWords].Length - 1, 1) == "?" || words[counterWords].Substring(words[counterWords].Length - 1, 1) == ")" || words[counterWords].Substring(words[counterWords].Length - 1, 1) == ":"))
{
words[counterWords] = words[counterWords].Substring(0, words[counterWords].Length - 1);
}
counterWords--;
}
if (Dostemming == true)
{
Stemmer s = new Stemmer();
words[counterWords] = s.stemTerm(words[counterWords]);
}
// check if it is date - [16TH]
if ((words[counterWords] != "") && (words[counterWords].Length == 4 && words[counterWords].Substring(2).ToLower() == "th") && (Double.TryParse(words[counterWords].Substring(0, 2), out i1)))
{
words[counterWords] = words[counterWords].Substring(0, 2);
}
// check if it is date - [9TH]
if ((words[counterWords] != "") && (words[counterWords].Length == 3 && words[counterWords].Substring(1).ToLower() == "th") && (Double.TryParse(words[counterWords].Substring(0, 1), out i1)))
{
words[counterWords] = words[counterWords].Substring(0, 2);
}
//check if the term is number in the end
if ((words[counterWords] != "") && (Double.TryParse(words[counterWords], out i2)) && (words.Length == counterWords + 1))
{
AfterParse[counterAfterParse] = words[counterWords];
CheckIFEnter = true;
counterAfterParse++;
} //check if it is NUMBER
else if ((words[counterWords] != "") && (Double.TryParse(words[counterWords], out i2) && !MonthDic.ContainsKey(words[counterWords + 1])))
{
double number = Double.Parse(words[counterWords]);
// if the string is up to miliion
if (number > 999999)
{
//the last word in the array
if ((counterWords + 1) == words.Length)
{
AfterParse[counterAfterParse] = IfNumber(Double.Parse(words[counterWords]), "");
CheckIFEnter = true;
}
else
{
// word in the middle
AfterParse[counterAfterParse] = IfNumber(Double.Parse(words[counterWords]), words[counterWords + 1]);
counterAfterParse++;
CheckIFEnter = true;
if (Dostemming == true)
{
Stemmer s = new Stemmer();
words[counterWords + 1] = s.stemTerm(words[counterWords + 1]);
}
if (words[counterWords + 1].ToLower() == "million" || words[counterWords + 1].ToLower() == "billion" || words[counterWords + 1].ToLower() == "trillion")
{
counterWords++;
}
}
}
else // the string is less than million [ num < MILLION]
{
//the last word in the array
if ((counterWords + 1) == words.Length)
{
AfterParse[counterAfterParse] = words[counterWords];
CheckIFEnter = true;
}
//if the string is SHEVER [ LIKE 3/4 ]
else if (words[counterWords + 1].Contains("/"))
{
int i = words[counterWords + 1].IndexOf("/");
double i3, i4;
if (Double.TryParse(words[counterWords + 1].Substring(0, i), out i3) && Double.TryParse(words[counterWords + 1].Substring(i + 1, words[counterWords + 1].Length - i - 1), out i4))
{
double new_num = i3 / i4;
i2 = i2 + new_num;
AfterParse[counterAfterParse] = i2.ToString();
counterWords++;
counterAfterParse++;
CheckIFEnter = true;
}
//AfterParse[counterAfterParse] = words[counterWords] + " " + words[counterWords + 1];
}
else // the string is in the middle
{
//check if there is enough words to end
if (counterWords + 1 < words.Length)
{
//this month
if (MonthDic.ContainsKey(words[counterWords + 1].ToLower()))
{
// check if the string is not in the end
if (counterWords + 2 < words.Length)
{
// this is NUMBER MONTH NUMBER [ LIKE 14 MAY 1991]
if (Double.TryParse(words[counterWords + 2], out i1))
{
// this is 2 DIGIT NUMBER [LIKE 91]
if (words[counterWords + 2].Length == 2)
{
if (i2 < 10)
{
words[counterWords] = "0" + words[counterWords];
}
AfterParse[counterAfterParse] = "19" + words[counterWords + 2] + "-" + MonthDic[words[counterWords + 1].ToLower()] + "-" + words[counterWords];
CheckIFEnter = true;
counterAfterParse++;
counterWords = counterWords + 2;
}
// this is 4 DIGITS NUMBER [LIKE 1991]
else if (words[counterWords + 2].Length == 4)
{
if (i2 < 10)
{
words[counterWords] = "0" + words[counterWords];
}
AfterParse[counterAfterParse] = words[counterWords + 2] + "-" + MonthDic[words[counterWords + 1].ToLower()] + "-" + words[counterWords];
CheckIFEnter = true;
counterAfterParse++;
counterWords = counterWords + 2;
}
}
// this is NUMBER MONTH [ 14 MAY ]
else
{
double ii;
// check if the number is small than 10. than change to 03
if (Double.TryParse(words[counterWords], out ii))
{
// check if the number is small than 10. than change to 03
if (ii < 10)
{
words[counterWords] = "0" + words[counterWords];
}
}
AfterParse[counterAfterParse] = MonthDic[words[counterWords + 1].ToLower()] + "-" + words[counterWords];
CheckIFEnter = true;
counterAfterParse++;
counterWords++;
}
}
// this is only NUMBER MONTH in the end of the string! [14 MAY]
else
{
double ii;
// check if the number is small than 10. than change to 03
if (Double.TryParse(words[counterWords], out ii))
{
// check if the number is small than 10. than change to 03
if (ii < 10)
{
words[counterWords] = "0" + words[counterWords];
}
}
AfterParse[counterAfterParse] = MonthDic[words[counterWords + 1].ToLower()] + "-" + words[counterWords];
CheckIFEnter = true;
counterAfterParse++;
counterWords++;
}
}
else //check if this is a Million/Trillion/billion
{
if (Dostemming == true)
{
Stemmer s = new Stemmer();
words[counterWords + 1] = s.stemTerm(words[counterWords + 1]);
}
if (words[counterWords + 1].ToLower() == "million" || words[counterWords + 1].ToLower() == "billion" || words[counterWords + 1].ToLower() == "trillion")
{
AfterParse[counterAfterParse] = IfNumber(i2, words[counterWords + 1]);
counterAfterParse++;
counterWords++;
}
}
}
else // the number is at the end of the array
{
AfterParse[counterAfterParse] = words[counterWords];
CheckIFEnter = true;
counterAfterParse++;
}
}
}
}
//check if in the phrase is BETWEEN
if ((words[counterWords] != "") && words[counterWords].ToLower() == "between")
{
if (counterWords + 3 < words.Length)
{
//check if it BETWEEN NUMBER and NUMBER
if (Double.TryParse(words[counterWords + 1], out i1) || double.TryParse(words[counterWords + 3], out i1))
{
AfterParse[counterAfterParse] = words[counterWords] + " " + words[counterWords + 1] + " " + words[counterWords + 2] + " " + words[counterWords + 3];
CheckIFEnter = true;
counterAfterParse++;
counterWords = counterWords + 3;
}
else // not contain a numberic variable
{
AfterParse[counterAfterParse] = words[counterWords];
CheckIFEnter = true;
counterAfterParse++;
}
}
}
//check if it is percenct and add %
if ((words[counterWords] != "") && (words[counterWords].ToLower() == "percent" || words[counterWords].ToLower() == "percentage"))
{
if (counterWords != 0)
{
// this in number in front of him [ like 9 % ]
if ((words[counterWords - 1] != "") && (Double.TryParse((AfterParse[counterAfterParse - 1]), out i1)))
{
AfterParse[counterAfterParse - 1] = AfterParse[counterAfterParse - 1] + "%";
CheckIFEnter = true;
}
}
}
//check if it is dollar and add DOLLARS WITH STEMMING
if (Dostemming == true)
{
if (words[counterWords] != "" && (words[counterWords].ToLower() == "dollar"))
{
if (Double.TryParse(words[counterWords - 1], out i2) || words[counterWords - 1].Contains("/"))
{
AfterParse[counterAfterParse - 1] = AfterParse[counterAfterParse - 1] + " Dollars";
CheckIFEnter = true;
}
}
}
//check if it is dollar and add DOLLARS
if (Dostemming == false)
{
if (words[counterWords] != "" && (words[counterWords].ToLower() == "dollars"))
{
if (Double.TryParse(words[counterWords - 1], out i2) || words[counterWords - 1].Contains("/"))
{
AfterParse[counterAfterParse - 1] = AfterParse[counterAfterParse - 1] + " Dollars";
CheckIFEnter = true;
}
}
}
//check if it is contain $ and add DOLLARS
if (words[counterWords] != "" && words[counterWords].Substring(0, 1) == ("$"))
{
words[counterWords] = words[counterWords].Substring(1);
//check that it is a NUMBER
if (Double.TryParse(words[counterWords], out i1))
{
// more than 999999
if (i1 > 999999)
{
AfterParse[counterAfterParse] = IfNumber(i1, "") + " Dollars";
CheckIFEnter = true;
counterAfterParse++;
}
else // less than 999999 ( i1 < 999999 )
{
if (Dostemming == true)
{
Stemmer s = new Stemmer();
words[counterWords + 1] = s.stemTerm(words[counterWords + 1]);
}
// the next word is million or billion or trillion
if (words[counterWords + 1].ToLower() == "million" || words[counterWords + 1].ToLower() == "trillion" || words[counterWords + 1].ToLower() == "billion")
{
AfterParse[counterAfterParse] = IfNumber(i1, words[counterWords + 1]) + " Dollars";
CheckIFEnter = true;
counterWords++;
counterAfterParse++;
}
else // NO next word
{
AfterParse[counterAfterParse] = words[counterWords] + " Dollars";
CheckIFEnter = true;
counterAfterParse++;
}
}
}
}
//check for tha MONTH
if (words[counterWords] != "" && MonthDic.ContainsKey(words[counterWords].ToLower()))
{
if (counterWords + 2 < words.Length)
{
// LIKE APRIL 28, 1990
if (Double.TryParse(words[counterWords + 2], out i2) && (Double.TryParse(words[counterWords + 1], out i1)))
{
// ADD 0 if the number is under 10
if (i1 < 10)
{
words[counterWords + 1] = "0" + words[counterWords + 1];
}
AfterParse[counterAfterParse] = words[counterWords + 2] + "-" + MonthDic[words[counterWords].ToLower()] + "-" + words[counterWords + 1].Substring(0, words[counterWords + 1].Length - 1);
CheckIFEnter = true;
counterAfterParse++;
counterWords = counterWords + 2;
}
// LIKE APRIL 28
else
{
if (Double.TryParse(words[counterWords + 1], out i2))
{
// MONTH DD [ APRIL 4 ]
if (i2 < 32)
{
// ADD 0 if the number is under 10
if (i2 < 10)
{
words[counterWords + 1] = "0" + words[counterWords + 1];
}
AfterParse[counterAfterParse] = MonthDic[words[counterWords].ToLower()] + "-" + words[counterWords + 1];
CheckIFEnter = true;
counterAfterParse++;
counterWords = counterWords++;
}
// if it is MONTH YEAR [ APRIL 1991]
else
{
// ADD 0 if the number is under 10
if (words[counterWords + 1] != "")
{
AfterParse[counterAfterParse] = words[counterWords + 1] + "-" + MonthDic[words[counterWords].ToLower()];
CheckIFEnter = true;
counterAfterParse++;
counterWords++;
}
}
}
}
}
// the string is in the end! [LIKE APRIL 28]
else if (counterWords + 1 < words.Length)
{
if (Double.TryParse(words[counterWords + 1], out i2))
{
// MONTH DD [ APRIL 4 ]
if (i2 < 32)
{
// ADD 0 if the number is under 10
if (i2 < 10)
{
words[counterWords + 1] = "0" + words[counterWords + 1];
}
AfterParse[counterAfterParse] = MonthDic[words[counterWords].ToLower()] + "-" + words[counterWords + 1];
CheckIFEnter = true;
counterAfterParse++;
counterWords = counterWords++;
}
// if it is MONTH YEAR [ APRIL 1991]
else
{
AfterParse[counterAfterParse] = words[counterWords + 1] + "-" + MonthDic[words[counterWords].ToLower()];
CheckIFEnter = true;
counterAfterParse++;
counterWords++;
}
}
}
else // the MONTH is in the end of the string
{
AfterParse[counterAfterParse] = words[counterWords];
CheckIFEnter = true;
counterAfterParse++;
}
}
// if the case is 100bn DOLLARS
if ((words[counterWords] != "" && words[counterWords].Length > 1))
{
if (words[counterWords].Substring(words[counterWords].Length - 2, 2).ToLower() == "bn")
{
if (Double.TryParse(words[counterWords].Substring(0, words[counterWords].Length - 2), out i1))
{
double num = i1 * 1000;
if (Dostemming == true)
{
Stemmer s = new Stemmer();
words[counterWords + 1] = s.stemTerm(words[counterWords + 1]);
if (words[counterWords + 1].ToLower() == "dollar")
{
AfterParse[counterAfterParse] = num.ToString() + " M Dollars";
CheckIFEnter = true;
}
}
else if (words[counterWords + 1].ToLower() == "dollars")
{
AfterParse[counterAfterParse] = num.ToString() + " M Dollars";
CheckIFEnter = true;
}
}
}
}
// for case 100 billion U.S dollars
if ((words[counterWords] != "" && counterWords + 2 < words.Length))
{
if (Dostemming == true)
{
Stemmer s = new Stemmer();
words[counterWords + 1] = s.stemTerm(words[counterWords + 1]);
if (words[counterWords].ToLower() == "u.s." && words[counterWords + 1].ToLower() == "dollar")
{
AfterParse[counterAfterParse - 1] = AfterParse[counterAfterParse - 1] + " Dollars";
CheckIFEnter = true;
}
}
else if (words[counterWords].ToLower() == "u.s." && words[counterWords + 1].ToLower() == "dollars")
{
AfterParse[counterAfterParse - 1] = AfterParse[counterAfterParse - 1] + " Dollars";
CheckIFEnter = true;
}
}
// case of Price UP to Million = 20.6m Dollars
if ((words[counterWords] != "" && counterWords + 1 < words.Length))
{
if (Dostemming == true)
{
Stemmer s = new Stemmer();
words[counterWords + 1] = s.stemTerm(words[counterWords + 1]);
if (words[counterWords].Substring(words[counterWords].Length - 1, 1).ToLower() == "m" && Double.TryParse(words[counterWords].Substring(0, words[counterWords].Length - 1), out i1) && words[counterWords + 1].ToLower() == "dollar")
{
AfterParse[counterAfterParse] = i1.ToString() + " M Dollars";
CheckIFEnter = true;
counterAfterParse++;
counterWords++;
}
}
else if (words[counterWords].Substring(words[counterWords].Length - 1, 1).ToLower() == "m" && Double.TryParse(words[counterWords].Substring(0, words[counterWords].Length - 1), out i1) && words[counterWords + 1].ToLower() == "dollars")
{
AfterParse[counterAfterParse] = i1.ToString() + " M Dollars";
CheckIFEnter = true;
counterAfterParse++;
counterWords++;
}
}
// if it is just a REGULAR STRING
if (words[counterWords] != "" && CheckIFEnter == false)
{
if (words[counterWords].Contains("/"))
{
int index = words[counterWords].IndexOf("/");
if (!Double.TryParse(words[counterWords].Substring(0, index), out i1))
{
AfterParse[counterAfterParse] = words[counterWords].Substring(0, index);
counterAfterParse++;
//contain more than 1 /
string temp_string = words[counterWords].Substring(index + 1, words[counterWords].Length - index - 1);
//delete * " ( from the beginning
while ((temp_string != "") && ((temp_string.Substring(0, 1) == "(") || temp_string.Substring(0, 1) == "}" || temp_string.Substring(0, 1) == ")" || temp_string.Substring(0, 1) == "[" || temp_string.Substring(0, 1) == "/" || temp_string.Substring(0, 1) == "|" || temp_string.Substring(0, 1) == "]" || temp_string.Substring(0, 1) == ";" || temp_string.Substring(0, 1) == ":" || temp_string.Substring(0, 1) == "_" || temp_string.Substring(0, 1) == "@" || temp_string.Substring(0, 1) == "=" || temp_string.Substring(0, 1) == "+" || temp_string.Substring(0, 1) == "!" || temp_string.Substring(0, 1) == "%" || temp_string.Substring(0, 1) == "|" || temp_string.Substring(0, 1) == "'" || (temp_string.Substring(0, 1) == ".") || temp_string.Substring(0, 1) == "`" || (temp_string.Substring(0, 1) == ",") || (temp_string.Substring(0, 1) == "?") || (temp_string.Substring(0, 1) == "&") || (temp_string.Substring(0, 1) == "[") || (temp_string.Substring(0, 1) == "\"") || (temp_string.Substring(0, 1) == "-") || (temp_string.Substring(0, 1) == "*")))
{
temp_string = temp_string.Substring(1, temp_string.Length - 1);
}
while (temp_string.Contains("/"))
{
index = temp_string.IndexOf("/");
AfterParse[counterAfterParse] = temp_string.Substring(0, index);
counterAfterParse++;
temp_string = temp_string.Substring(index + 1, temp_string.Length - index - 1);
}
AfterParse[counterAfterParse] = temp_string;
counterAfterParse++;
}
}
else
{
// check if it is not contain the stop words
if (!stopWords.Contains(words[counterWords].ToLower()))
{
if (words[counterWords].Substring(0, 1) != "-")
{
AfterParse[counterAfterParse] = words[counterWords];
counterAfterParse++;
CheckIFEnter = false;
// add the list the words connected
list_wordsCOnnected.AddLast(new WordsConnected(words[counterWords], counterAfterParse - 1));
}
}
}
}
counterWords++;
CheckIFEnter = false;
}
if (Doindexer == true)
{
Document doc = new Document(d[NumOfFile][2], AfterParse, counterAfterParse - 1, d[NumOfFile][0], d[NumOfFile][3]);
documents.Add(doc);
while (list_wordsCOnnected.Count >= 2)
{
WordsConnected obj1 = ((WordsConnected)list_wordsCOnnected.First());
list_wordsCOnnected.RemoveFirst();
WordsConnected obj2 = ((WordsConnected)list_wordsCOnnected.First());
list_wordsCOnnected.RemoveFirst();
// check if the string is connected - one string is after the other
if (obj1.LOCATION + 1 == obj2.LOCATION)
{
string key = obj1.NAME + " " + obj2.NAME;
// check if it is in the dictionary
if (final_connectWords.ContainsKey(key))
{
if (final_connectWords[key].DICTIONARY.ContainsKey(d[NumOfFile][2])) // if contains name of file
{
final_connectWords[key].DICTIONARY[d[NumOfFile][2]]++; // add the num of apperance
}
}
//not in the dictionary
else
{
final_connectWords.Add(key, new ArrayConnectedwords(d[NumOfFile][2]));
final_connectWords[key].DICTIONARY.Add(d[NumOfFile][2], 1);
}
}
list_wordsCOnnected.AddFirst(obj2);
}
}
else // enter the string after the parsing
{
d[NumOfFile] = AfterParse;
ArrayList arrayList = new ArrayList();
for (int i = 0; i < AfterParse.Length; i++)
{
if (AfterParse[i] != null)
{
arrayList.Add(AfterParse[i]);
}
}
string[] final_after_parsing = new string[arrayList.Count];
int j = 0;
foreach (string val in arrayList)
{
final_after_parsing[j] = val;
j++;
}
// create the final dictionary for queries
Dictionary <int, string[]> final_dic_query = new Dictionary <int, string[]>();
// get the num of the query
foreach (int num_query in d.Keys)
{
final_dic_query.Add(num_query, final_after_parsing);
}
//sent to the Ranker!
Ranker r = new Ranker(terms_dictionary, doc_dic, final_dic_query, total_lenght_doc, Dostemming, path_newFile, list_languages_pressed, path_queries_ranked);
ArrayList list_rel = new ArrayList();
m_rel_doc = new ArrayList();
Dictionary <string, double> temp_dic = new Dictionary <string, double>();
temp_dic = r.TOP50DOCS;
foreach (string doc in temp_dic.Keys)
{
list_rel.Add(doc);
}
m_rel_doc = list_rel;
m_query_num = r.QUERY_NUM; ////////////////////////////////
return;
}
}
//for the documents
if (Doindexer == true)
{
// string chaining = "";
Dictionary <string, ArrayConnectedwords> Dic_connect_new = new Dictionary <string, ArrayConnectedwords>();
foreach (var name in final_connectWords.OrderBy(i => i.Key))
{
Dic_connect_new[name.Key] = name.Value;
}
FileStream fs = new FileStream(path_wordsConnected, FileMode.Create);
StreamWriter sw = new StreamWriter(fs);
foreach (string term in Dic_connect_new.Keys)
{
sw.Write(term + "/");
foreach (string nameDoc in Dic_connect_new[term].DICTIONARY.Keys)
{
sw.Write(nameDoc + "," + Dic_connect_new[term].DICTIONARY[nameDoc] + ";");
}
sw.WriteLine();
}
sw.Close();
fs.Close();
Indexer indexer = new Indexer(documents, path_newFile, path_docs);
}
// for the queries
else
{
}
}