/// <summary>
/// Runs the indexing process.
/// </summary>
/// <param name="users">The users.</param>
/// <param name="created">if set to <c>true</c> [created].</param>
public void CreateOrUpdate(IList<User> users, bool created)
{
var modifier = new IndexModifier(
Setting.UserSearchIndexPath.Value,
new StandardAnalyzer(),
(!created) ? true : false
);
foreach (var user in users)
{
if (created)
{
modifier.DeleteDocuments(new Term("id", user.Id.ToString()));
}
var document = new Document();
UserToDocument(user, document);
modifier.AddDocument(document);
}
modifier.Optimize();
modifier.Close();
}
/// <summary>
/// Deletes the index.
/// </summary>
/// <param name="indexPath">The index path.</param>
/// <param name="docList">The doc list.</param>
/// <returns></returns>
public bool DeleteIndex(string indexPath, List <string> docList)
{
bool result = false;
try
{
using (Lucene.Net.Store.Directory directory = FSDirectory.Open(new DirectoryInfo(indexPath)))
{
Analyzer analyzer = new StandardAnalyzer(Lucene.Net.Util.Version.LUCENE_29);
var parser = new QueryParser(Lucene.Net.Util.Version.LUCENE_29, RegularExpression, analyzer);
IndexModifier modifier = new Lucene.Net.Index.IndexModifier(indexPath, analyzer, false);
foreach (string doc in docList)
{
modifier.DeleteDocuments(new Lucene.Net.Index.Term(FilePath, doc));
}
modifier.Flush();
modifier.Close();
}
}
catch (Exception ex)
{
// Swallow Exceptions here
}
return(result);
}
public void threadproc_update(object obj)
{
lock (locker) // If a thread is updating the index, no other thread should be doing anything with it.
{
try
{
if (searcher != null)
{
try
{
searcher.Close();
}
catch (Exception e)
{
}
searcher = null;
}
Lucene.Net.Index.IndexModifier modifier = new Lucene.Net.Index.IndexModifier(DBNLConfigurationManager.LuceneElement.IndexingFolder, analyzer, false);
// same as build, but uses "modifier" instead of write.
// uses additional "where" clause for bugid
int id = (int)obj;
modifier.DeleteDocuments(new Lucene.Net.Index.Term("id", Convert.ToString(id)));
var item = new ContentService().GetContentById(id);
modifier.AddDocument(create_doc(
item.ContentId, item.Content1));
modifier.Flush();
modifier.Close();
}
catch (Exception e)
{
}
}
}
public virtual void TestIndex()
{
Directory ramDir = new RAMDirectory();
IndexModifier i = new IndexModifier(ramDir, new StandardAnalyzer(), true);
i.AddDocument(GetDoc());
Assert.AreEqual(1, i.DocCount());
i.Flush();
i.AddDocument(GetDoc(), new SimpleAnalyzer());
Assert.AreEqual(2, i.DocCount());
i.Optimize();
Assert.AreEqual(2, i.DocCount());
i.Flush();
i.DeleteDocument(0);
Assert.AreEqual(1, i.DocCount());
i.Flush();
Assert.AreEqual(1, i.DocCount());
i.AddDocument(GetDoc());
i.AddDocument(GetDoc());
i.Flush();
// depend on merge policy - Assert.AreEqual(3, i.docCount());
i.DeleteDocuments(allDocTerm);
Assert.AreEqual(0, i.DocCount());
i.Optimize();
Assert.AreEqual(0, i.DocCount());
// Lucene defaults:
Assert.IsNull(i.GetInfoStream());
Assert.IsTrue(i.GetUseCompoundFile());
Assert.AreEqual(IndexWriter.DISABLE_AUTO_FLUSH, i.GetMaxBufferedDocs());
Assert.AreEqual(10000, i.GetMaxFieldLength());
Assert.AreEqual(10, i.GetMergeFactor());
// test setting properties:
i.SetMaxBufferedDocs(100);
i.SetMergeFactor(25);
i.SetMaxFieldLength(250000);
i.AddDocument(GetDoc());
i.SetUseCompoundFile(false);
i.Flush();
Assert.AreEqual(100, i.GetMaxBufferedDocs());
Assert.AreEqual(25, i.GetMergeFactor());
Assert.AreEqual(250000, i.GetMaxFieldLength());
Assert.IsFalse(i.GetUseCompoundFile());
// test setting properties when internally the reader is opened:
i.DeleteDocuments(allDocTerm);
i.SetMaxBufferedDocs(100);
i.SetMergeFactor(25);
i.SetMaxFieldLength(250000);
i.AddDocument(GetDoc());
i.SetUseCompoundFile(false);
i.Optimize();
Assert.AreEqual(100, i.GetMaxBufferedDocs());
Assert.AreEqual(25, i.GetMergeFactor());
Assert.AreEqual(250000, i.GetMaxFieldLength());
Assert.IsFalse(i.GetUseCompoundFile());
i.Close();
try
{
i.DocCount();
Assert.Fail();
}
catch (System.SystemException e)
{
// expected exception
}
}
public virtual void TestIndex()
{
Directory ramDir = new RAMDirectory();
IndexModifier i = new IndexModifier(ramDir, new StandardAnalyzer(), true);
i.AddDocument(GetDoc());
Assert.AreEqual(1, i.DocCount());
i.Flush();
i.AddDocument(GetDoc(), new SimpleAnalyzer());
Assert.AreEqual(2, i.DocCount());
i.Optimize();
Assert.AreEqual(2, i.DocCount());
i.Flush();
i.DeleteDocument(0);
Assert.AreEqual(1, i.DocCount());
i.Flush();
Assert.AreEqual(1, i.DocCount());
i.AddDocument(GetDoc());
i.AddDocument(GetDoc());
i.Flush();
// depend on merge policy - Assert.AreEqual(3, i.docCount());
i.DeleteDocuments(allDocTerm);
Assert.AreEqual(0, i.DocCount());
i.Optimize();
Assert.AreEqual(0, i.DocCount());
// Lucene defaults:
Assert.IsNull(i.GetInfoStream());
Assert.IsTrue(i.GetUseCompoundFile());
Assert.AreEqual(IndexWriter.DISABLE_AUTO_FLUSH, i.GetMaxBufferedDocs());
Assert.AreEqual(10000, i.GetMaxFieldLength());
Assert.AreEqual(10, i.GetMergeFactor());
// test setting properties:
i.SetMaxBufferedDocs(100);
i.SetMergeFactor(25);
i.SetMaxFieldLength(250000);
i.AddDocument(GetDoc());
i.SetUseCompoundFile(false);
i.Flush();
Assert.AreEqual(100, i.GetMaxBufferedDocs());
Assert.AreEqual(25, i.GetMergeFactor());
Assert.AreEqual(250000, i.GetMaxFieldLength());
Assert.IsFalse(i.GetUseCompoundFile());
// test setting properties when internally the reader is opened:
i.DeleteDocuments(allDocTerm);
i.SetMaxBufferedDocs(100);
i.SetMergeFactor(25);
i.SetMaxFieldLength(250000);
i.AddDocument(GetDoc());
i.SetUseCompoundFile(false);
i.Optimize();
Assert.AreEqual(100, i.GetMaxBufferedDocs());
Assert.AreEqual(25, i.GetMergeFactor());
Assert.AreEqual(250000, i.GetMaxFieldLength());
Assert.IsFalse(i.GetUseCompoundFile());
i.Close();
try
{
i.DocCount();
Assert.Fail();
}
catch (System.SystemException e)
{
// expected exception
}
}
override public void Run()
{
long endTime = (DateTime.Now.Ticks / TimeSpan.TicksPerMillisecond) + 1000 * TEST_SECONDS;
try
{
while ((DateTime.Now.Ticks / TimeSpan.TicksPerMillisecond) < endTime)
{
int rand = random.Next(101);
if (rand < 5)
{
index.Optimize();
}
else if (rand < 60)
{
Document doc = GetDocument();
index.AddDocument(doc);
idStack.Add(doc.Get("id"));
added++;
}
else
{
// we just delete the last document added and remove it
// from the id stack so that it won't be removed twice:
System.String delId = null;
try
{
delId = idStack[idStack.Count - 1] as System.String;
idStack.RemoveAt(idStack.Count - 1);
}
catch (System.ArgumentOutOfRangeException e)
{
continue;
}
Term delTerm = new Term("id", System.Int32.Parse(delId).ToString());
int delCount = index.DeleteDocuments(delTerm);
if (delCount != 1)
{
throw new System.SystemException("Internal error: " + threadNumber + " deleted " + delCount + " documents, term=" + delTerm);
}
deleted++;
}
if (maxWait > 0)
{
rand = random.Next(maxWait);
//System.out.println("waiting " + rand + "ms");
try
{
System.Threading.Thread.Sleep(new System.TimeSpan((System.Int64) 10000 * rand));
}
catch (System.Threading.ThreadInterruptedException ie)
{
SupportClass.ThreadClass.Current().Interrupt();
throw new System.SystemException("", ie);
}
}
}
}
catch (System.IO.IOException e)
{
throw new System.SystemException("", e);
}
}
/// <summary>
/// Loops thro a list of stories and adds them to the index. If the crawl is an incremental
/// update then first the story is removed then added again.
/// </summary>
/// <param name="modifier">IndexModifer used to update the index</param>
/// <param name="isIncrementalCrawl">bool indicating if the stories should
/// be removed from the existing index before being added again.</param>
/// <param name="stories">StoryCollection containing the stories to add/update
/// in the index</param>
private void AddStoriesToIndex(IndexModifier modifier, bool isIncrementalCrawl, StoryCollection stories)
{
if (isIncrementalCrawl)
{
//remove the stories from the index that have been updated
Log.DebugFormat("Updating index, removing {0} stories", stories.Count);
foreach (Story s in stories)
{
Term existingItem = new Term("id", s.StoryID.ToString());
int j = modifier.DeleteDocuments(existingItem);
}
}
//add the new documents
Log.DebugFormat("Adding batch of {0} stories to the index", stories.Count);
foreach (Story story in stories)
{
//spam stories shouldnt be added to the index
if (story.IsSpam)
continue;
Document doc = new Document();
doc.Add(new Field("url", story.Url, Field.Store.NO, Field.Index.TOKENIZED));
doc.Add(new Field("title", story.Title, Field.Store.NO, Field.Index.TOKENIZED, Field.TermVector.YES));
doc.Add(new Field("description", story.Description, Field.Store.NO, Field.Index.TOKENIZED));
doc.Add(new Field("users", GetUserWhoKickedSearchString(story), Field.Store.NO, Field.Index.TOKENIZED));
doc.Add(new Field("category", story.Category.Name, Field.Store.NO, Field.Index.TOKENIZED));
doc.Add(new Field("tags", GetStoryTags(story), Field.Store.NO, Field.Index.TOKENIZED, Field.TermVector.YES));
doc.Add(new Field("id", story.StoryID.ToString(), Field.Store.YES, Field.Index.UN_TOKENIZED));
doc.Add(new Field("kickCount", story.KickCount.ToString(), Field.Store.NO, Field.Index.UN_TOKENIZED));
doc.Add(new Field("dateAdded", DateField.DateToString(story.CreatedOn), Field.Store.NO, Field.Index.UN_TOKENIZED));
modifier.AddDocument(doc);
Log.DebugFormat("StoryId {0} added to index", story.StoryID);
}
}
