public string HighlightContent(string text)
{
QueryScorer scorer = new QueryScorer(GetQuery());
Formatter formatter = new SimpleHTMLFormatter("<span style='color:maroon; font-weight:bold;'>", "</span>");
Highlighter highlighter = new Highlighter(formatter, scorer);
highlighter.SetTextFragmenter(new SimpleFragmenter(120));
TokenStream stream = new StandardAnalyzer().TokenStream("Content", new StringReader(text));
var fragments = highlighter.GetBestFragments(stream, text, 3);
if (fragments == null || fragments.Length == 0) return text.Length > 120 ? text.Substring(0, 120) + "..." : text;
string highlighted = "";
foreach (var fragment in fragments)
{
if (text.StartsWith(fragment))
highlighted += "<p>" + fragment + " ... </p>";
else if (text.EndsWith(fragment))
highlighted += "<p> ... " + fragment + "</p>";
else
highlighted += "<p> ... " + fragment + " ... </p>";
}
return highlighted;
}
public IEnumerable<Hit> Search(string query, int maxResults)
{
var analyzer = new StandardAnalyzer(Lucene.Net.Util.Version.LUCENE_29);
QueryParser qp = new QueryParser(
Lucene.Net.Util.Version.LUCENE_29,
"contents",
analyzer
);
Query q = qp.Parse(query);
TopDocs top = searcher.Search(q, maxResults);
List<Hit> result = new List<Hit>(top.totalHits);
for (int index = 0; index < top.totalHits; index++)
{
var doc = searcher.Doc(top.scoreDocs[index].doc);
string contents = doc.Get("contents");
var scorer = new QueryScorer(q, searcher.GetIndexReader(), "contents");
var highlighter = new Highlighter(scorer);
result.Add(new Hit()
{
Relevance = top.scoreDocs[index].score,
Title = doc.Get("title"),
Url = doc.Get("path"),
Excerpt = highlighter.GetBestFragment(analyzer, "contents", contents)
});
}
return result;
}
public string GetHighlight(string value, IndexSearcher searcher, string highlightField, Query luceneQuery)
{
var scorer = new QueryScorer(luceneQuery.Rewrite(searcher.GetIndexReader()));
var highlighter = new Highlighter(HighlightFormatter, scorer);
var tokenStream = HighlightAnalyzer.TokenStream(highlightField, new StringReader(value));
return highlighter.GetBestFragments(tokenStream, value, MaxNumHighlights, Separator);
}
public string GetHighlight(string value, string highlightField, Searcher searcher, string luceneRawQuery)
{
var query = GetQueryParser(highlightField).Parse(luceneRawQuery);
var scorer = new QueryScorer(searcher.Rewrite(query));
var highlighter = new Highlighter(HighlightFormatter, scorer);
var tokenStream = HighlightAnalyzer.TokenStream(highlightField, new StringReader(value));
string bestFragments = highlighter.GetBestFragments(tokenStream, value, MaxNumHighlights, Separator);
return bestFragments;
}
public string HighlightTitle(string text)
{
QueryScorer scorer = new QueryScorer(GetQuery());
Formatter formatter = new SimpleHTMLFormatter("<span style='color:maroon; font-weight:bold;'>", "</span>");
Highlighter highlighter = new Highlighter(formatter, scorer);
highlighter.SetTextFragmenter(new NullFragmenter());
TokenStream stream = new StandardAnalyzer().TokenStream("Title", new StringReader(text));
var title = highlighter.GetBestFragment(stream, text);
return title ?? text;
}
public IEnumerable<SearchResult> Search(string query)
{
Analyzer analyzer = new SnowballAnalyzer("English");
QueryParser parser = new QueryParser("text", analyzer);
Query luceneQuery = parser.Parse(query);
Directory directory = FSDirectory.GetDirectory(indexPath);
IndexSearcher searcher = new IndexSearcher(directory);
QueryScorer queryScorer = new QueryScorer(luceneQuery);
Highlighter highlighter = new Highlighter(queryScorer);
TopDocs topDocs = searcher.Search(luceneQuery, 100);
var searchResults = new List<SearchResult>();
foreach (ScoreDoc scoreDoc in topDocs.scoreDocs)
{
Document doc = searcher.Doc(scoreDoc.doc);
searchResults.Add(new SearchResult { Path = doc.Get("path"), Score = scoreDoc.score, Title = doc.Get("title"), Preview = highlighter.GetBestFragment(analyzer, "text", doc.Get("text")) });
}
return searchResults;
}
public IEnumerable<SearchResult> Search(string luceneQuery, int maxResults = 500, string highlightOpenTag = null, string highlightCloseTag = null, params string[] fieldsToHighlight)
{
var results = new List<SearchResult>();
if (String.IsNullOrWhiteSpace(luceneQuery)) return results;
var parser = new MultiFieldQueryParser(LuceneVersion, searchFields, new StandardAnalyzer(LuceneVersion));
Query query = parser.Parse(luceneQuery);
TopDocs topDocs = searcher.Search(query, maxResults);
foreach (ScoreDoc scoreDoc in topDocs.ScoreDocs)
{
Document document = reader.Document(scoreDoc.doc);
var result = new SearchResult(document, scoreDoc.score);
results.Add(result);
}
if(!String.IsNullOrEmpty(highlightOpenTag) && !String.IsNullOrEmpty(highlightCloseTag) && fieldsToHighlight.Length > 0)
{
var scorer = new QueryScorer(query);
var formatter = new SimpleHTMLFormatter(highlightOpenTag, highlightCloseTag);
var highlighter = new Highlighter(formatter, scorer);
highlighter.SetTextFragmenter(new SimpleFragmenter());
foreach (SearchResult result in results)
{
foreach (string highlightField in fieldsToHighlight)
{
if(!result.Fields.ContainsKey(highlightField)) continue;
string fieldValue = result[highlightField];
TokenStream stream = new StandardAnalyzer(LuceneVersion).TokenStream(highlightField, new StringReader(fieldValue));
string highlightedFieldValue = highlighter.GetBestFragments(stream, fieldValue, 500, "...");
if (!String.IsNullOrWhiteSpace(highlightedFieldValue))
{
result.Fields[highlightField] = highlightedFieldValue;
}
}
}
}
return results;
}
private string GeneratePreviewText(Query q, string text)
{
var scorer = new QueryScorer(q);
var formatter = new SimpleHTMLFormatter("<em>", "</em>");
var highlighter = new Highlighter(formatter, scorer);
highlighter.SetTextFragmenter(new SimpleFragmenter(250));
var stream = new StandardAnalyzer(Version.LUCENE_29).TokenStream("bodyText", new StringReader(text));
return highlighter.GetBestFragments(stream, text, 3, "...");
}
private SearchResults Query(Query query, string cachePath,
int offset, int resultAmount, bool includeContentSnippets, int limit)
{
SearchResults results = new SearchResults();
Lucene.Net.Store.FSDirectory directory = Lucene.Net.Store.FSDirectory.Open(new System.IO.DirectoryInfo(_indexPath));
IndexReader reader = IndexReader.Open(directory, true);
IndexSearcher searcher = new IndexSearcher(reader);
// Request all results up to the page we actually need (this is quick)
TopDocs topDocs = searcher.Search(query, limit);
ScoreDoc[] hits = topDocs.scoreDocs;
// Save num results
results.NumResults = hits.Length;
// Only loop through the hits that should be on the page
for (int i = offset; i < hits.Length && i < offset + resultAmount; i++)
{
int docId = hits[i].doc;
Document doc = searcher.Doc(docId);
if (includeContentSnippets)
{
// Read the whole file from the cache to find the content snippet.
string filepath = CacheManager.GetRelativeCacheFileName(doc.Get("uri"), "GET");
string documentContent = Utils.ReadFileAsString(cachePath + filepath);
// Remove unusable stuff.
documentContent = HtmlUtils.RemoveHead(documentContent);
documentContent = HtmlUtils.ExtractText(documentContent);
// Find (and highlight) content snippets
QueryScorer scorer = new QueryScorer(query);
SimpleHTMLFormatter formatter = new SimpleHTMLFormatter("<b>", "</b>");
Highlighter highlighter = new Highlighter(formatter, scorer);
highlighter.SetTextFragmenter(new SentenceFragmenter());
TokenStream stream = _analyzer.TokenStream("content", new StringReader(documentContent));
// Get 1 fragment
string contentSnippet = "";
try
{
string[] fragments = highlighter.GetBestFragments(stream, documentContent, 1);
if (fragments.Length > 0)
{
contentSnippet = HtmlUtils.StripTagsCharArray(fragments[0], false);
// If the content snippet does end in mid of a sentence, let's append "..."
if (!new char[] { '.', '!', '?' }.Contains(contentSnippet[contentSnippet.Length - 1]))
{
contentSnippet += "...";
}
}
}
catch (Exception)
{
}
results.AddLuceneDocument(doc, contentSnippet);
}
else
{
results.AddLuceneDocument(doc);
}
}
searcher.Close();
return results;
}
