/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ using System; using System.Collections.Generic; using System.Linq; using Lucene.Net.Support; using Document = Lucene.Net.Documents.Document; using FieldSelector = Lucene.Net.Documents.FieldSelector; using Directory = Lucene.Net.Store.Directory; using Lock = Lucene.Net.Store.Lock; using LockObtainFailedException = Lucene.Net.Store.LockObtainFailedException; using DefaultSimilarity = Lucene.Net.Search.DefaultSimilarity; namespace Lucene.Net.Index { /// An IndexReader which reads indexes with multiple segments. public class DirectoryReader:IndexReader { /*new*/ private class AnonymousClassFindSegmentsFile:SegmentInfos.FindSegmentsFile { private void InitBlock(bool readOnly, IndexDeletionPolicy deletionPolicy, int termInfosIndexDivisor) { this.readOnly = readOnly; this.deletionPolicy = deletionPolicy; this.termInfosIndexDivisor = termInfosIndexDivisor; } private bool readOnly; private IndexDeletionPolicy deletionPolicy; private int termInfosIndexDivisor; internal AnonymousClassFindSegmentsFile(bool readOnly, Lucene.Net.Index.IndexDeletionPolicy deletionPolicy, int termInfosIndexDivisor, Lucene.Net.Store.Directory Param1):base(Param1) { InitBlock(readOnly, deletionPolicy, termInfosIndexDivisor); } public /*protected internal*/ override System.Object DoBody(System.String segmentFileName) { var infos = new SegmentInfos(); infos.Read(directory, segmentFileName); if (readOnly) return new ReadOnlyDirectoryReader(directory, infos, deletionPolicy, termInfosIndexDivisor); else return new DirectoryReader(directory, infos, deletionPolicy, false, termInfosIndexDivisor); } } private class AnonymousClassFindSegmentsFile1:SegmentInfos.FindSegmentsFile { private void InitBlock(bool openReadOnly, DirectoryReader enclosingInstance) { this.openReadOnly = openReadOnly; this.enclosingInstance = enclosingInstance; } private bool openReadOnly; private DirectoryReader enclosingInstance; public DirectoryReader Enclosing_Instance { get { return enclosingInstance; } } internal AnonymousClassFindSegmentsFile1(bool openReadOnly, DirectoryReader enclosingInstance, Lucene.Net.Store.Directory Param1):base(Param1) { InitBlock(openReadOnly, enclosingInstance); } public /*protected internal*/ override System.Object DoBody(System.String segmentFileName) { var infos = new SegmentInfos(); infos.Read(directory, segmentFileName); return Enclosing_Instance.DoReopen(infos, false, openReadOnly); } } protected internal Directory internalDirectory; protected internal bool readOnly; internal IndexWriter writer; private IndexDeletionPolicy deletionPolicy; private readonly HashSet synced = new HashSet(); private Lock writeLock; private readonly SegmentInfos segmentInfos; private readonly SegmentInfos segmentInfosStart; private bool stale; private readonly int termInfosIndexDivisor; private bool rollbackHasChanges; private SegmentReader[] subReaders; private int[] starts; // 1st docno for each segment private System.Collections.Generic.IDictionary normsCache = new HashMap(); private int maxDoc = 0; private int numDocs = - 1; private bool hasDeletions = false; // Max version in index as of when we opened; this can be // > our current segmentInfos version in case we were // opened on a past IndexCommit: private long maxIndexVersion; internal static IndexReader Open(Directory directory, IndexDeletionPolicy deletionPolicy, IndexCommit commit, bool readOnly, int termInfosIndexDivisor) { return (IndexReader) new AnonymousClassFindSegmentsFile(readOnly, deletionPolicy, termInfosIndexDivisor, directory).Run(commit); } /// Construct reading the named set of readers. internal DirectoryReader(Directory directory, SegmentInfos sis, IndexDeletionPolicy deletionPolicy, bool readOnly, int termInfosIndexDivisor) { internalDirectory = directory; this.readOnly = readOnly; this.segmentInfos = sis; this.deletionPolicy = deletionPolicy; this.termInfosIndexDivisor = termInfosIndexDivisor; if (!readOnly) { // We assume that this segments_N was previously // properly sync'd: synced.UnionWith(sis.Files(directory, true)); } // To reduce the chance of hitting FileNotFound // (and having to retry), we open segments in // reverse because IndexWriter merges & deletes // the newest segments first. var readers = new SegmentReader[sis.Count]; for (int i = sis.Count - 1; i >= 0; i--) { bool success = false; try { readers[i] = SegmentReader.Get(readOnly, sis.Info(i), termInfosIndexDivisor); success = true; } finally { if (!success) { // Close all readers we had opened: for (i++; i < sis.Count; i++) { try { readers[i].Close(); } catch (System.Exception) { // keep going - we want to clean up as much as possible } } } } } Initialize(readers); } // Used by near real-time search internal DirectoryReader(IndexWriter writer, SegmentInfos infos, int termInfosIndexDivisor) { this.internalDirectory = writer.Directory; this.readOnly = true; segmentInfos = infos; segmentInfosStart = (SegmentInfos) infos.Clone(); this.termInfosIndexDivisor = termInfosIndexDivisor; if (!readOnly) { // We assume that this segments_N was previously // properly sync'd: synced.UnionWith(infos.Files(internalDirectory, true)); } // IndexWriter synchronizes externally before calling // us, which ensures infos will not change; so there's // no need to process segments in reverse order int numSegments = infos.Count; var readers = new SegmentReader[numSegments]; Directory dir = writer.Directory; int upto = 0; for (int i = 0; i < numSegments; i++) { bool success = false; try { SegmentInfo info = infos.Info(i); if (info.dir == dir) { readers[upto++] = writer.readerPool.GetReadOnlyClone(info, true, termInfosIndexDivisor); } success = true; } finally { if (!success) { // Close all readers we had opened: for (upto--; upto >= 0; upto--) { try { readers[upto].Close(); } catch (System.Exception) { // keep going - we want to clean up as much as possible } } } } } this.writer = writer; if (upto < readers.Length) { // This means some segments were in a foreign Directory var newReaders = new SegmentReader[upto]; Array.Copy(readers, 0, newReaders, 0, upto); readers = newReaders; } Initialize(readers); } /// This constructor is only used for internal DirectoryReader(Directory directory, SegmentInfos infos, SegmentReader[] oldReaders, int[] oldStarts, IEnumerable> oldNormsCache, bool readOnly, bool doClone, int termInfosIndexDivisor) { this.internalDirectory = directory; this.readOnly = readOnly; this.segmentInfos = infos; this.termInfosIndexDivisor = termInfosIndexDivisor; if (!readOnly) { // We assume that this segments_N was previously // properly sync'd: synced.UnionWith(infos.Files(directory, true)); } // we put the old SegmentReaders in a map, that allows us // to lookup a reader using its segment name IDictionary segmentReaders = new HashMap(); if (oldReaders != null) { // create a Map SegmentName->SegmentReader for (int i = 0; i < oldReaders.Length; i++) { segmentReaders[oldReaders[i].SegmentName] = i; } } var newReaders = new SegmentReader[infos.Count]; // remember which readers are shared between the old and the re-opened // DirectoryReader - we have to incRef those readers var readerShared = new bool[infos.Count]; for (int i = infos.Count - 1; i >= 0; i--) { // find SegmentReader for this segment if (!segmentReaders.ContainsKey(infos.Info(i).name)) { // this is a new segment, no old SegmentReader can be reused newReaders[i] = null; } else { // there is an old reader for this segment - we'll try to reopen it newReaders[i] = oldReaders[segmentReaders[infos.Info(i).name]]; } bool success = false; try { SegmentReader newReader; if (newReaders[i] == null || infos.Info(i).GetUseCompoundFile() != newReaders[i].SegmentInfo.GetUseCompoundFile()) { // We should never see a totally new segment during cloning System.Diagnostics.Debug.Assert(!doClone); // this is a new reader; in case we hit an exception we can close it safely newReader = SegmentReader.Get(readOnly, infos.Info(i), termInfosIndexDivisor); } else { newReader = newReaders[i].ReopenSegment(infos.Info(i), doClone, readOnly); } if (newReader == newReaders[i]) { // this reader will be shared between the old and the new one, // so we must incRef it readerShared[i] = true; newReader.IncRef(); } else { readerShared[i] = false; newReaders[i] = newReader; } success = true; } finally { if (!success) { for (i++; i < infos.Count; i++) { if (newReaders[i] != null) { try { if (!readerShared[i]) { // this is a new subReader that is not used by the old one, // we can close it newReaders[i].Close(); } else { // this subReader is also used by the old reader, so instead // closing we must decRef it newReaders[i].DecRef(); } } catch (System.IO.IOException) { // keep going - we want to clean up as much as possible } } } } } } // initialize the readers to calculate maxDoc before we try to reuse the old normsCache Initialize(newReaders); // try to copy unchanged norms from the old normsCache to the new one if (oldNormsCache != null) { foreach(var entry in oldNormsCache) { String field = entry.Key; if (!HasNorms(field)) { continue; } byte[] oldBytes = entry.Value; var bytes = new byte[MaxDoc]; for (int i = 0; i < subReaders.Length; i++) { int oldReaderIndex = segmentReaders[subReaders[i].SegmentName]; // this SegmentReader was not re-opened, we can copy all of its norms if (segmentReaders.ContainsKey(subReaders[i].SegmentName) && (oldReaders[oldReaderIndex] == subReaders[i] || oldReaders[oldReaderIndex].norms[field] == subReaders[i].norms[field])) { // we don't have to synchronize here: either this constructor is called from a SegmentReader, // in which case no old norms cache is present, or it is called from MultiReader.reopen(), // which is synchronized Array.Copy(oldBytes, oldStarts[oldReaderIndex], bytes, starts[i], starts[i + 1] - starts[i]); } else { subReaders[i].Norms(field, bytes, starts[i]); } } normsCache[field] = bytes; // update cache } } } private void Initialize(SegmentReader[] subReaders) { this.subReaders = subReaders; starts = new int[subReaders.Length + 1]; // build starts array for (int i = 0; i < subReaders.Length; i++) { starts[i] = maxDoc; maxDoc += subReaders[i].MaxDoc; // compute maxDocs if (subReaders[i].HasDeletions) hasDeletions = true; } starts[subReaders.Length] = maxDoc; if (!readOnly) { maxIndexVersion = SegmentInfos.ReadCurrentVersion(internalDirectory); } } public override Object Clone() { lock (this) { try { return Clone(readOnly); // Preserve current readOnly } catch (Exception ex) { throw new SystemException(ex.Message, ex); // TODO: why rethrow this way? } } } public override IndexReader Clone(bool openReadOnly) { lock (this) { DirectoryReader newReader = DoReopen((SegmentInfos) segmentInfos.Clone(), true, openReadOnly); if (this != newReader) { newReader.deletionPolicy = deletionPolicy; } newReader.writer = writer; // If we're cloning a non-readOnly reader, move the // writeLock (if there is one) to the new reader: if (!openReadOnly && writeLock != null) { // In near real-time search, reader is always readonly System.Diagnostics.Debug.Assert(writer == null); newReader.writeLock = writeLock; newReader.hasChanges = hasChanges; newReader.hasDeletions = hasDeletions; writeLock = null; hasChanges = false; } return newReader; } } public override IndexReader Reopen() { // Preserve current readOnly return DoReopen(readOnly, null); } public override IndexReader Reopen(bool openReadOnly) { return DoReopen(openReadOnly, null); } public override IndexReader Reopen(IndexCommit commit) { return DoReopen(true, commit); } private IndexReader DoReopenFromWriter(bool openReadOnly, IndexCommit commit) { System.Diagnostics.Debug.Assert(readOnly); if (!openReadOnly) { throw new System.ArgumentException("a reader obtained from IndexWriter.getReader() can only be reopened with openReadOnly=true (got false)"); } if (commit != null) { throw new System.ArgumentException("a reader obtained from IndexWriter.getReader() cannot currently accept a commit"); } // TODO: right now we *always* make a new reader; in // the future we could have write make some effort to // detect that no changes have occurred return writer.GetReader(); } internal virtual IndexReader DoReopen(bool openReadOnly, IndexCommit commit) { EnsureOpen(); System.Diagnostics.Debug.Assert(commit == null || openReadOnly); // If we were obtained by writer.getReader(), re-ask the // writer to get a new reader. if (writer != null) { return DoReopenFromWriter(openReadOnly, commit); } else { return DoReopenNoWriter(openReadOnly, commit); } } private IndexReader DoReopenNoWriter(bool openReadOnly, IndexCommit commit) { lock (this) { if (commit == null) { if (hasChanges) { // We have changes, which means we are not readOnly: System.Diagnostics.Debug.Assert(readOnly == false); // and we hold the write lock: System.Diagnostics.Debug.Assert(writeLock != null); // so no other writer holds the write lock, which // means no changes could have been done to the index: System.Diagnostics.Debug.Assert(IsCurrent()); if (openReadOnly) { return Clone(openReadOnly); } else { return this; } } else if (IsCurrent()) { if (openReadOnly != readOnly) { // Just fallback to clone return Clone(openReadOnly); } else { return this; } } } else { if (internalDirectory != commit.Directory) throw new System.IO.IOException("the specified commit does not match the specified Directory"); if (segmentInfos != null && commit.SegmentsFileName.Equals(segmentInfos.GetCurrentSegmentFileName())) { if (readOnly != openReadOnly) { // Just fallback to clone return Clone(openReadOnly); } else { return this; } } } return (IndexReader)new AnonymousFindSegmentsFile(internalDirectory, openReadOnly, this).Run(commit); } } class AnonymousFindSegmentsFile : SegmentInfos.FindSegmentsFile { readonly DirectoryReader enclosingInstance; readonly bool openReadOnly; readonly Directory dir; public AnonymousFindSegmentsFile(Directory directory, bool openReadOnly, DirectoryReader dirReader) : base(directory) { this.dir = directory; this.openReadOnly = openReadOnly; enclosingInstance = dirReader; } public override object DoBody(string segmentFileName) { var infos = new SegmentInfos(); infos.Read(dir, segmentFileName); return enclosingInstance.DoReopen(infos, false, openReadOnly); } } private DirectoryReader DoReopen(SegmentInfos infos, bool doClone, bool openReadOnly) { lock (this) { DirectoryReader reader; if (openReadOnly) { reader = new ReadOnlyDirectoryReader(internalDirectory, infos, subReaders, starts, normsCache, doClone, termInfosIndexDivisor); } else { reader = new DirectoryReader(internalDirectory, infos, subReaders, starts, normsCache, false, doClone, termInfosIndexDivisor); } return reader; } } /// Version number when this IndexReader was opened. public override long Version { get { EnsureOpen(); return segmentInfos.Version; } } public override ITermFreqVector[] GetTermFreqVectors(int n) { EnsureOpen(); int i = ReaderIndex(n); // find segment num return subReaders[i].GetTermFreqVectors(n - starts[i]); // dispatch to segment } public override ITermFreqVector GetTermFreqVector(int n, System.String field) { EnsureOpen(); int i = ReaderIndex(n); // find segment num return subReaders[i].GetTermFreqVector(n - starts[i], field); } public override void GetTermFreqVector(int docNumber, System.String field, TermVectorMapper mapper) { EnsureOpen(); int i = ReaderIndex(docNumber); // find segment num subReaders[i].GetTermFreqVector(docNumber - starts[i], field, mapper); } public override void GetTermFreqVector(int docNumber, TermVectorMapper mapper) { EnsureOpen(); int i = ReaderIndex(docNumber); // find segment num subReaders[i].GetTermFreqVector(docNumber - starts[i], mapper); } /// Checks is the index is optimized (if it has a single segment and no deletions) /// &lt;c&gt;true&lt;/c&gt; if the index is optimized; &lt;c&gt;false&lt;/c&gt; otherwise public override bool IsOptimized() { EnsureOpen(); return segmentInfos.Count == 1 && !HasDeletions; } public override int NumDocs() { // Don't call ensureOpen() here (it could affect performance) // NOTE: multiple threads may wind up init'ing // numDocs... but that's harmless if (numDocs == - 1) { // check cache int n = subReaders.Sum(t => t.NumDocs()); // cache miss--recompute numDocs = n; } return numDocs; } public override int MaxDoc { get { // Don't call ensureOpen() here (it could affect performance) return maxDoc; } } // inherit javadoc public override Document Document(int n, FieldSelector fieldSelector) { EnsureOpen(); int i = ReaderIndex(n); // find segment num return subReaders[i].Document(n - starts[i], fieldSelector); // dispatch to segment reader } public override bool IsDeleted(int n) { // Don't call ensureOpen() here (it could affect performance) int i = ReaderIndex(n); // find segment num return subReaders[i].IsDeleted(n - starts[i]); // dispatch to segment reader } public override bool HasDeletions { get { // Don't call ensureOpen() here (it could affect performance) return hasDeletions; } } protected internal override void DoDelete(int n) { numDocs = - 1; // invalidate cache int i = ReaderIndex(n); // find segment num subReaders[i].DeleteDocument(n - starts[i]); // dispatch to segment reader hasDeletions = true; } protected internal override void DoUndeleteAll() { foreach (SegmentReader t in subReaders) t.UndeleteAll(); hasDeletions = false; numDocs = - 1; // invalidate cache } private int ReaderIndex(int n) { // find reader for doc n: return ReaderIndex(n, this.starts, this.subReaders.Length); } internal static int ReaderIndex(int n, int[] starts, int numSubReaders) { // find reader for doc n: int lo = 0; // search starts array int hi = numSubReaders - 1; // for first element less while (hi >= lo) { int mid = Number.URShift((lo + hi), 1); int midValue = starts[mid]; if (n < midValue) hi = mid - 1; else if (n > midValue) lo = mid + 1; else { // found a match while (mid + 1 < numSubReaders && starts[mid + 1] == midValue) { mid++; // scan to last match } return mid; } } return hi; } public override bool HasNorms(System.String field) { EnsureOpen(); return subReaders.Any(t => t.HasNorms(field)); } public override byte[] Norms(System.String field) { lock (this) { EnsureOpen(); byte[] bytes = normsCache[field]; if (bytes != null) return bytes; // cache hit if (!HasNorms(field)) return null; bytes = new byte[MaxDoc]; for (int i = 0; i < subReaders.Length; i++) subReaders[i].Norms(field, bytes, starts[i]); normsCache[field] = bytes; // update cache return bytes; } } public override void Norms(System.String field, byte[] result, int offset) { lock (this) { EnsureOpen(); byte[] bytes = normsCache[field]; if (bytes == null && !HasNorms(field)) { byte val = DefaultSimilarity.EncodeNorm(1.0f); for (int index = offset; index < result.Length; index++) result.SetValue(val, index); } else if (bytes != null) { // cache hit Array.Copy(bytes, 0, result, offset, MaxDoc); } else { for (int i = 0; i < subReaders.Length; i++) { // read from segments subReaders[i].Norms(field, result, offset + starts[i]); } } } } protected internal override void DoSetNorm(int n, System.String field, byte value_Renamed) { lock (normsCache) { normsCache.Remove(field); // clear cache } int i = ReaderIndex(n); // find segment num subReaders[i].SetNorm(n - starts[i], field, value_Renamed); // dispatch } public override TermEnum Terms() { EnsureOpen(); return new MultiTermEnum(this, subReaders, starts, null); } public override TermEnum Terms(Term term) { EnsureOpen(); return new MultiTermEnum(this, subReaders, starts, term); } public override int DocFreq(Term t) { EnsureOpen(); int total = 0; // sum freqs in segments for (int i = 0; i < subReaders.Length; i++) total += subReaders[i].DocFreq(t); return total; } public override TermDocs TermDocs() { EnsureOpen(); return new MultiTermDocs(this, subReaders, starts); } public override TermPositions TermPositions() { EnsureOpen(); return new MultiTermPositions(this, subReaders, starts); } /// Tries to acquire the WriteLock on this directory. this method is only valid if this IndexReader is directory /// owner. /// /// /// StaleReaderException if the index has changed since this reader was opened /// CorruptIndexException if the index is corrupt /// Lucene.Net.Store.LockObtainFailedException /// if another writer has this index open (write.lock could not be /// obtained) /// /// IOException if there is a low-level IO error protected internal override void AcquireWriteLock() { if (readOnly) { // NOTE: we should not reach this code w/ the core // IndexReader classes; however, an external subclass // of IndexReader could reach this. ReadOnlySegmentReader.NoWrite(); } if (segmentInfos != null) { EnsureOpen(); if (stale) throw new StaleReaderException("IndexReader out of date and no longer valid for delete, undelete, or setNorm operations"); if (this.writeLock == null) { Lock writeLock = internalDirectory.MakeLock(IndexWriter.WRITE_LOCK_NAME); if (!writeLock.Obtain(IndexWriter.WRITE_LOCK_TIMEOUT)) // obtain write lock { throw new LockObtainFailedException("Index locked for write: " + writeLock); } this.writeLock = writeLock; // we have to check whether index has changed since this reader was opened. // if so, this reader is no longer valid for // deletion if (SegmentInfos.ReadCurrentVersion(internalDirectory) > maxIndexVersion) { stale = true; this.writeLock.Release(); this.writeLock = null; throw new StaleReaderException("IndexReader out of date and no longer valid for delete, undelete, or setNorm operations"); } } } } /// Commit changes resulting from delete, undeleteAll, or setNorm operations ///

/// If an exception is hit, then either no changes or all changes will have been committed to the index (transactional /// semantics). /// ///

/// IOException if there is a low-level IO error protected internal override void DoCommit(IDictionary commitUserData) { if (hasChanges) { segmentInfos.UserData = commitUserData; // Default deleter (for backwards compatibility) is // KeepOnlyLastCommitDeleter: var deleter = new IndexFileDeleter(internalDirectory, deletionPolicy ?? new KeepOnlyLastCommitDeletionPolicy(), segmentInfos, null, null, synced); segmentInfos.UpdateGeneration(deleter.LastSegmentInfos); // Checkpoint the state we are about to change, in // case we have to roll back: StartCommit(); bool success = false; try { foreach (SegmentReader t in subReaders) t.Commit(); // Sync all files we just wrote foreach(string fileName in segmentInfos.Files(internalDirectory, false)) { if(!synced.Contains(fileName)) { System.Diagnostics.Debug.Assert(internalDirectory.FileExists(fileName)); internalDirectory.Sync(fileName); synced.Add(fileName); } } segmentInfos.Commit(internalDirectory); success = true; } finally { if (!success) { // Rollback changes that were made to // SegmentInfos but failed to get [fully] // committed. This way this reader instance // remains consistent (matched to what's // actually in the index): RollbackCommit(); // Recompute deletable files & remove them (so // partially written .del files, etc, are // removed): deleter.Refresh(); } } // Have the deleter remove any now unreferenced // files due to this commit: deleter.Checkpoint(segmentInfos, true); deleter.Dispose(); maxIndexVersion = segmentInfos.Version; if (writeLock != null) { writeLock.Release(); // release write lock writeLock = null; } } hasChanges = false; } internal virtual void StartCommit() { rollbackHasChanges = hasChanges; foreach (SegmentReader t in subReaders) { t.StartCommit(); } } internal virtual void RollbackCommit() { hasChanges = rollbackHasChanges; foreach (SegmentReader t in subReaders) { t.RollbackCommit(); } } public override IDictionary CommitUserData { get { EnsureOpen(); return segmentInfos.UserData; } } public override bool IsCurrent() { EnsureOpen(); if (writer == null || writer.IsClosed()) { // we loaded SegmentInfos from the directory return SegmentInfos.ReadCurrentVersion(internalDirectory) == segmentInfos.Version; } else { return writer.NrtIsCurrent(segmentInfosStart); } } protected internal override void DoClose() { lock (this) { System.IO.IOException ioe = null; normsCache = null; foreach (SegmentReader t in subReaders) { // try to close each reader, even if an exception is thrown try { t.DecRef(); } catch (System.IO.IOException e) { if (ioe == null) ioe = e; } } // NOTE: only needed in case someone had asked for // FieldCache for top-level reader (which is generally // not a good idea): Search.FieldCache_Fields.DEFAULT.Purge(this); // throw the first exception if (ioe != null) throw ioe; } } public override ICollection GetFieldNames(IndexReader.FieldOption fieldNames) { EnsureOpen(); return GetFieldNames(fieldNames, this.subReaders); } internal static ICollection GetFieldNames(IndexReader.FieldOption fieldNames, IndexReader[] subReaders) { // maintain a unique set of field names ISet fieldSet = Support.Compatibility.SetFactory.CreateHashSet(); foreach (IndexReader reader in subReaders) { fieldSet.UnionWith(reader.GetFieldNames(fieldNames)); } return fieldSet; } public override IndexReader[] GetSequentialSubReaders() { return subReaders; } /// Returns the directory this index resides in. public override Directory Directory() { // Don't ensureOpen here -- in certain cases, when a // cloned/reopened reader needs to commit, it may call // this method on the closed original reader return internalDirectory; } public override int TermInfosIndexDivisor { get { return termInfosIndexDivisor; } } /// Expert: return the IndexCommit that this reader has opened. ///

///

WARNING: this API is new and experimental and may suddenly change.

///

public override IndexCommit IndexCommit { get { return new ReaderCommit(segmentInfos, internalDirectory); } } /// /// public static new ICollection ListCommits(Directory dir) { String[] files = dir.ListAll(); ICollection commits = new List(); var latest = new SegmentInfos(); latest.Read(dir); long currentGen = latest.Generation; commits.Add(new ReaderCommit(latest, dir)); foreach (string fileName in files) { if (fileName.StartsWith(IndexFileNames.SEGMENTS) && !fileName.Equals(IndexFileNames.SEGMENTS_GEN) && SegmentInfos.GenerationFromSegmentsFileName(fileName) < currentGen) { var sis = new SegmentInfos(); try { // IOException allowed to throw there, in case // segments_N is corrupt sis.Read(dir, fileName); } catch (System.IO.FileNotFoundException) { // LUCENE-948: on NFS (and maybe others), if // you have writers switching back and forth // between machines, it's very likely that the // dir listing will be stale and will claim a // file segments_X exists when in fact it // doesn't. So, we catch this and handle it // as if the file does not exist sis = null; } if (sis != null) commits.Add(new ReaderCommit(sis, dir)); } } return commits; } private sealed class ReaderCommit:IndexCommit { private readonly String segmentsFileName; private readonly ICollection files; private readonly Directory dir; private readonly long generation; private readonly long version; private readonly bool isOptimized; private readonly IDictionary userData; internal ReaderCommit(SegmentInfos infos, Directory dir) { segmentsFileName = infos.GetCurrentSegmentFileName(); this.dir = dir; userData = infos.UserData; files = infos.Files(dir, true); version = infos.Version; generation = infos.Generation; isOptimized = infos.Count == 1 && !infos.Info(0).HasDeletions(); } public override string ToString() { return "DirectoryReader.ReaderCommit(" + segmentsFileName + ")"; } public override bool IsOptimized { get { return isOptimized; } } public override string SegmentsFileName { get { return segmentsFileName; } } public override ICollection FileNames { get { return files; } } public override Directory Directory { get { return dir; } } public override long Version { get { return version; } } public override long Generation { get { return generation; } } public override bool IsDeleted { get { return false; } } public override IDictionary UserData { get { return userData; } } public override void Delete() { throw new System.NotSupportedException("This IndexCommit does not support deletions"); } } internal class MultiTermEnum:TermEnum { internal IndexReader topReader; // used for matching TermEnum to TermDocs private readonly SegmentMergeQueue queue; private Term term; private int docFreq; internal SegmentMergeInfo[] matchingSegments; // null terminated array of matching segments public MultiTermEnum(IndexReader topReader, IndexReader[] readers, int[] starts, Term t) { this.topReader = topReader; queue = new SegmentMergeQueue(readers.Length); matchingSegments = new SegmentMergeInfo[readers.Length + 1]; for (int i = 0; i < readers.Length; i++) { IndexReader reader = readers[i]; TermEnum termEnum = t != null ? reader.Terms(t) : reader.Terms(); var smi = new SegmentMergeInfo(starts[i], termEnum, reader) {ord = i}; if (t == null?smi.Next():termEnum.Term != null) queue.Add(smi); // initialize queue else smi.Dispose(); } if (t != null && queue.Size() > 0) { Next(); } } public override bool Next() { foreach (SegmentMergeInfo smi in matchingSegments) { if (smi == null) break; if (smi.Next()) queue.Add(smi); else smi.Dispose(); // done with segment } int numMatchingSegments = 0; matchingSegments[0] = null; SegmentMergeInfo top = queue.Top(); if (top == null) { term = null; return false; } term = top.term; docFreq = 0; while (top != null && term.CompareTo(top.term) == 0) { matchingSegments[numMatchingSegments++] = top; queue.Pop(); docFreq += top.termEnum.DocFreq(); // increment freq top = queue.Top(); } matchingSegments[numMatchingSegments] = null; return true; } public override Term Term { get { return term; } } public override int DocFreq() { return docFreq; } protected override void Dispose(bool disposing) { if (disposing) { queue.Dispose(); } } } internal class MultiTermDocs : TermDocs { internal IndexReader topReader; // used for matching TermEnum to TermDocs protected internal IndexReader[] readers; protected internal int[] starts; protected internal Term term; protected internal int base_Renamed = 0; protected internal int pointer = 0; private readonly TermDocs[] readerTermDocs; protected internal TermDocs current; // == readerTermDocs[pointer] private MultiTermEnum tenum; // the term enum used for seeking... can be null internal int matchingSegmentPos; // position into the matching segments from tenum internal SegmentMergeInfo smi; // current segment mere info... can be null public MultiTermDocs(IndexReader topReader, IndexReader[] r, int[] s) { this.topReader = topReader; readers = r; starts = s; readerTermDocs = new TermDocs[r.Length]; } public virtual int Doc { get { return base_Renamed + current.Doc; } } public virtual int Freq { get { return current.Freq; } } public virtual void Seek(Term term) { this.term = term; this.base_Renamed = 0; this.pointer = 0; this.current = null; this.tenum = null; this.smi = null; this.matchingSegmentPos = 0; } public virtual void Seek(TermEnum termEnum) { Seek(termEnum.Term); var multiTermEnum = termEnum as MultiTermEnum; if (multiTermEnum != null) { tenum = multiTermEnum; if (topReader != tenum.topReader) tenum = null; } } public virtual bool Next() { for (; ; ) { if (current != null && current.Next()) { return true; } else if (pointer < readers.Length) { if (tenum != null) { smi = tenum.matchingSegments[matchingSegmentPos++]; if (smi == null) { pointer = readers.Length; return false; } pointer = smi.ord; } base_Renamed = starts[pointer]; current = TermDocs(pointer++); } else { return false; } } } /// Optimized implementation. public virtual int Read(int[] docs, int[] freqs) { while (true) { while (current == null) { if (pointer < readers.Length) { // try next segment if (tenum != null) { smi = tenum.matchingSegments[matchingSegmentPos++]; if (smi == null) { pointer = readers.Length; return 0; } pointer = smi.ord; } base_Renamed = starts[pointer]; current = TermDocs(pointer++); } else { return 0; } } int end = current.Read(docs, freqs); if (end == 0) { // none left in segment current = null; } else { // got some int b = base_Renamed; // adjust doc numbers for (int i = 0; i < end; i++) docs[i] += b; return end; } } } /* A Possible future optimization could skip entire segments */ public virtual bool SkipTo(int target) { for (; ; ) { if (current != null && current.SkipTo(target - base_Renamed)) { return true; } else if (pointer < readers.Length) { if (tenum != null) { SegmentMergeInfo smi = tenum.matchingSegments[matchingSegmentPos++]; if (smi == null) { pointer = readers.Length; return false; } pointer = smi.ord; } base_Renamed = starts[pointer]; current = TermDocs(pointer++); } else return false; } } private TermDocs TermDocs(int i) { TermDocs result = readerTermDocs[i] ?? (readerTermDocs[i] = TermDocs(readers[i])); if (smi != null) { System.Diagnostics.Debug.Assert((smi.ord == i)); System.Diagnostics.Debug.Assert((smi.termEnum.Term.Equals(term))); result.Seek(smi.termEnum); } else { result.Seek(term); } return result; } protected internal virtual TermDocs TermDocs(IndexReader reader) { return term == null ? reader.TermDocs(null):reader.TermDocs(); } public virtual void Close() { Dispose(); } public virtual void Dispose() { Dispose(true); } protected virtual void Dispose(bool disposing) { if (disposing) { foreach (TermDocs t in readerTermDocs) { if (t != null) t.Close(); } } } } internal class MultiTermPositions:MultiTermDocs, TermPositions { public MultiTermPositions(IndexReader topReader, IndexReader[] r, int[] s):base(topReader, r, s) { } protected internal override TermDocs TermDocs(IndexReader reader) { return reader.TermPositions(); } public virtual int NextPosition() { return ((TermPositions) current).NextPosition(); } public virtual int PayloadLength { get { return ((TermPositions) current).PayloadLength; } } public virtual byte[] GetPayload(byte[] data, int offset) { return ((TermPositions) current).GetPayload(data, offset); } // TODO: Remove warning after API has been finalized public virtual bool IsPayloadAvailable { get { return ((TermPositions) current).IsPayloadAvailable; } } } } }