/// <summary>
/// Searches for the specified query, returns a ranked List of Documents matching any word in the query.
/// </summary>
/// <param name="query">The query.</param>
/// <returns></returns>
public static List <Document> Search(String query)
{
//Separate words, remove punctiations,make lowercase
List <String> words = Semanter.Splitwords(query, ":").ToList();
//Obtains possible types searched by this query
HashSet <String> typesPossible = TypeChecker(words);
//Stem words and remove stopwords
//slower method words = words.Except(invt.Stopwords).ToList();
List <String> splitwords = new List <String>();
string stem;
foreach (string word in words)
{
stem = invt.Samantha.StemWord(word);
if (!(invt.Stopwords.Contains(stem)))
{
splitwords.Add(stem);
}
}
if (splitwords.Count == 0)
{
return(new List <Document>());
}
//search for documents
Dictionary <Document, Dictionary <string, List <int> > > Results = DocsFound(splitwords, typesPossible);
if (Results.Keys.Count < 2)
{
return(Results.Keys.ToList());
}
return(Ranker.RankQuery(splitwords, Results, invt.DocumentCount));
}
/// <summary>
/// Initializes a new instance of the <see cref="Inverter" /> class.
/// </summary>
/// <param name="StopWords">The path to a File Containing all the stop words.</param>
/// <param name="DictionaryPath">The path to the file holding all legal words.</param>
/// <param name="CommonWordsPath">The path to the file holding most coomonly used words.</param>
/// <param name="FormatsPath">The path to the file holding all supported file types.</param>
/// <param name="BooksPaths">Paths to Books, used for frequency nalysis of words.</param>
/// <exception cref="IOException">The specified path could not be Read</exception>
public Inverter(String StopWords, String DictionaryPath, String CommonWordsPath, String FormatsPath, List <String> BooksPaths)
{
Formats = new Dictionary <string, List <string> >();
AddToFormats(File.ReadAllLines(FormatsPath));
_samantha = new Semanter(DictionaryPath, CommonWordsPath);
//tomiwas idea about weight distribution based on file size
foreach (String BookPath in BooksPaths)
{
_samantha.AddToDictionary(BookPath, 1);
}
_stopwords = new HashSet <string>();
_documentCount = 0;
this.store = new Store();
_documentCount = Files.Count;
try {
foreach (String stp in File.ReadAllLines(StopWords))
{
_stopwords.Add(stp);
}
} catch (Exception ex) {
throw new IOException("The specified path for stopwords could not be Read", ex);
}
}
/// <summary>
/// Modifies the file in the inverted Index Table.
/// </summary>
/// <param name="doc">The document that was Modified.</param>
public static Document ModifyFile(Document doc)
{
String[] words = Semanter.Splitwords(x.Extract(doc.Address).Text);
return(invt.ModifyDocument(words, doc));
}
/// <summary>
/// Adds words from the Specified Document to the specified Inverted Index Table
/// </summary>
/// <param name="doc">The document to be Tokenized.</param>
/// <exception cref="TextExtractionException">Could not extract Files from the Document</exception>
public static void AddFileFrom(Document doc)
{
String[] words = Semanter.Splitwords(x.Extract(doc.Address).Text);
invt.AddDocument(words, doc);
}