public IEnumerable <SampleHit> Search(string query_str)
{
List <SampleHit> result_hits = new List <SampleHit>();
using (Lucene.Net.Store.Directory luceneIndexDirectory = Lucene.Net.Store.FSDirectory.Open(index_folder))
{
Lucene.Net.Analysis.Analyzer analyzer = new Lucene.Net.Analysis.Ru.RussianAnalyzer(Lucene.Net.Util.Version.LUCENE_CURRENT);
//Lucene.Net.Analysis.Analyzer analyzer = new Lucene.Net.Analysis.Standard.StandardAnalyzer(Lucene.Net.Util.Version.LUCENE_CURRENT);
using (IndexSearcher searcher = new IndexSearcher(luceneIndexDirectory))
{
QueryParser parser = new QueryParser(Lucene.Net.Util.Version.LUCENE_CURRENT, IndexModel.LineText, analyzer);
Query query = parser.Parse(query_str);
TopDocs hits = searcher.Search(query, max_search_hits);
// code highlighting
var formatter = new Lucene.Net.Search.Highlight.SimpleHTMLFormatter("<span style=\"background:yellow;\">", "</span>");
var fragmenter = new Lucene.Net.Search.Highlight.SimpleFragmenter(200);
Lucene.Net.Search.Highlight.QueryScorer scorer = new Lucene.Net.Search.Highlight.QueryScorer(query);
Lucene.Net.Search.Highlight.Highlighter highlighter = new Lucene.Net.Search.Highlight.Highlighter(formatter, scorer);
highlighter.TextFragmenter = fragmenter;
foreach (ScoreDoc hit in hits.ScoreDocs)
{
Document doc = searcher.Doc(hit.Doc);
float score = hit.Score;
Field line_number = doc.GetField(IndexModel.LineNumber);
Field line_text = doc.GetField(IndexModel.LineText);
Lucene.Net.Analysis.TokenStream stream = analyzer.TokenStream("", new System.IO.StringReader(line_text.StringValue));
string highlightedText = highlighter.GetBestFragments(stream, doc.Get(IndexModel.LineText), 1, "...");
result_hits.Add(new SampleHit {
line_number = line_number.StringValue, sample_text = line_text.StringValue, html_highlighting = highlightedText
});
}
}
}
return(result_hits);
}
/// <summary>
/// Determines a list of search query matches.
/// </summary>
/// <param name="searchType">Type of search.</param>
/// <param name="query">Query, whose matches should be returned.</param>
/// <param name="filesAnalyzer">Files analyzer to be used.</param>
/// <param name="contentsAnalyzer">Contents analyzer to be used.</param>
/// <param name="tagsAnalyzer">Tags analyzer to be used.</param>
/// <param name="text">Original text</param>
/// <param name="createFragment">Determines, whether fragments should be created or not.</param>
/// <returns>Collection of search query match occurrences.</returns>
public static ReadOnlyCollection <IOccurrence> DetermineOccurrences(string searchType, Query query, string text, Analyzer filesAnalyzer, Analyzer contentsAnalyzer, Analyzer tagsAnalyzer, bool createFragment)
{
// Create and configure a Lucene highlighter in order to obtain a list of search
// query matches in the document.
var scorer = new Lucene.Net.Search.Highlight.QueryScorer(query);
var occurrenceCollector = new OccurrenceCollectorFormatter();
var highlighter = new Lucene.Net.Search.Highlight.Highlighter(occurrenceCollector, scorer);
// We must extend the analyzer limit to the whole text.
highlighter.MaxDocCharsToAnalyze = text.Length;
// We want to analyze the whole text, thus the NullFragmenter.
highlighter.TextFragmenter = new Lucene.Net.Search.Highlight.NullFragmenter();
switch (searchType)
{
case SearchType.Files:
highlighter.GetBestFragments(filesAnalyzer.ReusableTokenStream(LuceneIndexBuilder.FIELD_PATH, new System.IO.StringReader(text)), text, 1);
break;
case SearchType.Contents:
highlighter.GetBestFragments(contentsAnalyzer.ReusableTokenStream(LuceneIndexBuilder.FIELD_CONTENTS, new System.IO.StringReader(text)), text, 1);
break;
case SearchType.Tags:
highlighter.GetBestFragments(tagsAnalyzer.ReusableTokenStream(LuceneIndexBuilder.FIELD_TAGS, new System.IO.StringReader(text)), text, 1);
break;
}
// Obtain line offsets from original file. This is needed in order to initialize the
// Line and Column properts of IOccurrence.
int[] lineOffsets = TextUtilities.GetLineOffsets(text).ToArray();
// Process the list of matches obtained through the Lucene highlighter
// and create a list of IOccurrenc instances.
List <IOccurrence> occurrences = new List <IOccurrence>();
foreach (var occurrence in occurrenceCollector.Occurrences)
{
// Obtain line number from offset by doing binary search.
int line = TextUtilities.OffsetToLineIndex(occurrence.Value.Item3, lineOffsets);
// Get start offset of line, this is needed to obtain the right column number.
int lineOffset = lineOffsets[line];
List <KeyValuePair <string, int> > fragment = null;
// Create occurrence fragment. Currently the fragment consists of 6 lines around the match.
if (createFragment)
{
int fragmentStartLine = Math.Max(line - 2, 0);
int fragmentEndLine = Math.Min(fragmentStartLine + 3, lineOffsets.Length - 2);
fragment = new List <KeyValuePair <string, int> >();
for (int i = fragmentStartLine; i < fragmentEndLine; ++i)
{
fragment.Add(new KeyValuePair <string, int>(text.Substring(lineOffsets[i], lineOffsets[i + 1] - lineOffsets[i]), i));
}
}
// Add occurrence
occurrences.Add(new Occurrence(occurrence.Key.Substring(occurrence.Value.Item3 - occurrence.Value.Item1, occurrence.Value.Item4 - occurrence.Value.Item3), line, occurrence.Value.Item3 - lineOffset, fragment != null ? fragment.ToArray() : null));
}
return(new ReadOnlyCollection <IOccurrence>(occurrences));
}