/// <summary>Retrieve the length (in bytes) of the tvd and tvf /// entries for the next numDocs starting with /// startDocID. This is used for bulk copying when /// merging segments, if the field numbers are /// congruent. Once this returns, the tvf & tvd streams /// are seeked to the startDocID. /// </summary> internal void RawDocs(int[] tvdLengths, int[] tvfLengths, int startDocID, int numDocs) { if (tvx == null) { for (int i = 0; i < tvdLengths.Length; i++) { tvdLengths[i] = 0; } for (int i = 0; i < tvfLengths.Length; i++) { tvfLengths[i] = 0; } return; } // SegmentMerger calls canReadRawDocs() first and should // not call us if that returns false. if (format < FORMAT_VERSION2) { throw new System.SystemException("cannot read raw docs with older term vector formats"); } SeekTvx(startDocID); long tvdPosition = tvx.ReadLong(); tvd.Seek(tvdPosition); long tvfPosition = tvx.ReadLong(); tvf.Seek(tvfPosition); long lastTvdPosition = tvdPosition; long lastTvfPosition = tvfPosition; int count = 0; while (count < numDocs) { int docID = docStoreOffset + startDocID + count + 1; System.Diagnostics.Debug.Assert(docID <= numTotalDocs); if (docID < numTotalDocs) { tvdPosition = tvx.ReadLong(); tvfPosition = tvx.ReadLong(); } else { tvdPosition = tvd.Length(); tvfPosition = tvf.Length(); System.Diagnostics.Debug.Assert(count == numDocs - 1); } tvdLengths[count] = (int)(tvdPosition - lastTvdPosition); tvfLengths[count] = (int)(tvfPosition - lastTvfPosition); count++; lastTvdPosition = tvdPosition; lastTvfPosition = tvfPosition; } }
public CompoundFileReader(Directory dir, System.String name, int readBufferSize) { directory = dir; fileName = name; this.readBufferSize = readBufferSize; bool success = false; try { stream = dir.OpenInput(name, readBufferSize); // read the directory and init files int count = stream.ReadVInt(); FileEntry entry = null; for (int i = 0; i < count; i++) { long offset = stream.ReadLong(); System.String id = stream.ReadString(); if (entry != null) { // set length of the previous entry entry.length = offset - entry.offset; } entry = new FileEntry(); entry.offset = offset; entries[id] = entry; } // set the length of the final entry if (entry != null) { entry.length = stream.Length() - entry.offset; } success = true; } finally { if (!success && (stream != null)) { try { stream.Close(); } catch (System.IO.IOException e) { } } } }
internal SegmentTermEnum(IndexInput i, FieldInfos fis, bool isi) { input = i; fieldInfos = fis; isIndex = isi; maxSkipLevels = 1; // use single-level skip lists for formats > -3 int firstInt = input.ReadInt(); if (firstInt >= 0) { // original-format file, without explicit format version number format = 0; size = firstInt; // back-compatible settings indexInterval = 128; skipInterval = System.Int32.MaxValue; // switch off skipTo optimization } else { // we have a format version number format = firstInt; // check that it is a format we can understand if (format < TermInfosWriter.FORMAT_CURRENT) { throw new CorruptIndexException("Unknown format version:" + format + " expected " + TermInfosWriter.FORMAT_CURRENT + " or higher"); } size = input.ReadLong(); // read the size if (format == -1) { if (!isIndex) { indexInterval = input.ReadInt(); formatM1SkipInterval = input.ReadInt(); } // switch off skipTo optimization for file format prior to 1.4rc2 in order to avoid a bug in // skipTo implementation of these versions skipInterval = System.Int32.MaxValue; } else { indexInterval = input.ReadInt(); skipInterval = input.ReadInt(); if (format <= TermInfosWriter.FORMAT) { // this new format introduces multi-level skipping maxSkipLevels = input.ReadInt(); } } System.Diagnostics.Debug.Assert(indexInterval > 0, "indexInterval=" + indexInterval + " is negative; must be > 0"); System.Diagnostics.Debug.Assert(skipInterval > 0, "skipInterval=" + skipInterval + " is negative; must be > 0"); } if (format > TermInfosWriter.FORMAT_VERSION_UTF8_LENGTH_IN_BYTES) { termBuffer.SetPreUTF8Strings(); scanBuffer.SetPreUTF8Strings(); prevBuffer.SetPreUTF8Strings(); } }
/// <summary> Construct a new SegmentInfo instance by reading a /// previously saved SegmentInfo from input. /// /// </summary> /// <param name="dir">directory to load from /// </param> /// <param name="format">format of the segments info file /// </param> /// <param name="input">input handle to read segment info from /// </param> internal SegmentInfo(Directory dir, int format, IndexInput input) { this.dir = dir; name = input.ReadString(); docCount = input.ReadInt(); if (format <= SegmentInfos.FORMAT_LOCKLESS) { delGen = input.ReadLong(); if (format <= SegmentInfos.FORMAT_SHARED_DOC_STORE) { docStoreOffset = input.ReadInt(); if (docStoreOffset != - 1) { docStoreSegment = input.ReadString(); docStoreIsCompoundFile = (1 == input.ReadByte()); } else { docStoreSegment = name; docStoreIsCompoundFile = false; } } else { docStoreOffset = - 1; docStoreSegment = name; docStoreIsCompoundFile = false; } if (format <= SegmentInfos.FORMAT_SINGLE_NORM_FILE) { hasSingleNormFile = (1 == input.ReadByte()); } else { hasSingleNormFile = false; } int numNormGen = input.ReadInt(); if (numNormGen == NO) { normGen = null; } else { normGen = new long[numNormGen]; for (int j = 0; j < numNormGen; j++) { normGen[j] = input.ReadLong(); } } isCompoundFile = (sbyte) input.ReadByte(); preLockless = (isCompoundFile == CHECK_DIR); if (format <= SegmentInfos.FORMAT_DEL_COUNT) { delCount = input.ReadInt(); System.Diagnostics.Debug.Assert(delCount <= docCount); } else delCount = - 1; if (format <= SegmentInfos.FORMAT_HAS_PROX) hasProx = input.ReadByte() == 1; else hasProx = true; if (format <= SegmentInfos.FORMAT_DIAGNOSTICS) { diagnostics = input.ReadStringStringMap(); } else { diagnostics = new System.Collections.Generic.Dictionary<string,string>(); } } else { delGen = CHECK_DIR; normGen = null; isCompoundFile = (sbyte) (CHECK_DIR); preLockless = true; hasSingleNormFile = false; docStoreOffset = - 1; docStoreIsCompoundFile = false; docStoreSegment = null; delCount = - 1; hasProx = true; diagnostics = new System.Collections.Generic.Dictionary<string,string>(); } }
/// <summary> Construct a new SegmentInfo instance by reading a /// previously saved SegmentInfo from input. /// /// </summary> /// <param name="dir">directory to load from /// </param> /// <param name="format">format of the segments info file /// </param> /// <param name="input">input handle to read segment info from /// </param> internal SegmentInfo(Directory dir, int format, IndexInput input) { this.dir = dir; name = input.ReadString(); docCount = input.ReadInt(); if (format <= SegmentInfos.FORMAT_LOCKLESS) { delGen = input.ReadLong(); if (format <= SegmentInfos.FORMAT_SHARED_DOC_STORE) { docStoreOffset = input.ReadInt(); if (docStoreOffset != -1) { docStoreSegment = input.ReadString(); docStoreIsCompoundFile = (1 == input.ReadByte()); } else { docStoreSegment = name; docStoreIsCompoundFile = false; } } else { docStoreOffset = -1; docStoreSegment = name; docStoreIsCompoundFile = false; } if (format <= SegmentInfos.FORMAT_SINGLE_NORM_FILE) { hasSingleNormFile = (1 == input.ReadByte()); } else { hasSingleNormFile = false; } int numNormGen = input.ReadInt(); if (numNormGen == NO) { normGen = null; } else { normGen = new long[numNormGen]; for (int j = 0; j < numNormGen; j++) { normGen[j] = input.ReadLong(); } } isCompoundFile = (sbyte)input.ReadByte(); preLockless = (isCompoundFile == CHECK_DIR); if (format <= SegmentInfos.FORMAT_DEL_COUNT) { delCount = input.ReadInt(); System.Diagnostics.Debug.Assert(delCount <= docCount); } else { delCount = -1; } if (format <= SegmentInfos.FORMAT_HAS_PROX) { hasProx = input.ReadByte() == 1; } else { hasProx = true; } if (format <= SegmentInfos.FORMAT_DIAGNOSTICS) { diagnostics = input.ReadStringStringMap(); } else { diagnostics = new System.Collections.Generic.Dictionary <string, string>(); } } else { delGen = CHECK_DIR; normGen = null; isCompoundFile = (sbyte)(CHECK_DIR); preLockless = true; hasSingleNormFile = false; docStoreOffset = -1; docStoreIsCompoundFile = false; docStoreSegment = null; delCount = -1; hasProx = true; diagnostics = new System.Collections.Generic.Dictionary <string, string>(); } }
public System.Object Run(IndexCommit commit) { if (commit != null) { if (directory != commit.GetDirectory()) { throw new System.IO.IOException("the specified commit does not match the specified Directory"); } return(DoBody(commit.GetSegmentsFileName())); } System.String segmentFileName = null; long lastGen = -1; long gen = 0; int genLookaheadCount = 0; System.IO.IOException exc = null; bool retry = false; int method = 0; // Loop until we succeed in calling doBody() without // hitting an IOException. An IOException most likely // means a commit was in process and has finished, in // the time it took us to load the now-old infos files // (and segments files). It's also possible it's a // true error (corrupt index). To distinguish these, // on each retry we must see "forward progress" on // which generation we are trying to load. If we // don't, then the original error is real and we throw // it. // We have three methods for determining the current // generation. We try the first two in parallel, and // fall back to the third when necessary. while (true) { if (0 == method) { // Method 1: list the directory and use the highest // segments_N file. This method works well as long // as there is no stale caching on the directory // contents (NOTE: NFS clients often have such stale // caching): System.String[] files = null; long genA = -1; files = directory.ListAll(); if (files != null) { genA = Mono.Lucene.Net.Index.SegmentInfos.GetCurrentSegmentGeneration(files); } Mono.Lucene.Net.Index.SegmentInfos.Message("directory listing genA=" + genA); // Method 2: open segments.gen and read its // contents. Then we take the larger of the two // gens. This way, if either approach is hitting // a stale cache (NFS) we have a better chance of // getting the right generation. long genB = -1; for (int i = 0; i < Mono.Lucene.Net.Index.SegmentInfos.defaultGenFileRetryCount; i++) { IndexInput genInput = null; try { genInput = directory.OpenInput(IndexFileNames.SEGMENTS_GEN); } catch (System.IO.FileNotFoundException e) { Mono.Lucene.Net.Index.SegmentInfos.Message("segments.gen open: FileNotFoundException " + e); break; } catch (System.IO.IOException e) { Mono.Lucene.Net.Index.SegmentInfos.Message("segments.gen open: IOException " + e); } if (genInput != null) { try { int version = genInput.ReadInt(); if (version == Mono.Lucene.Net.Index.SegmentInfos.FORMAT_LOCKLESS) { long gen0 = genInput.ReadLong(); long gen1 = genInput.ReadLong(); Mono.Lucene.Net.Index.SegmentInfos.Message("fallback check: " + gen0 + "; " + gen1); if (gen0 == gen1) { // The file is consistent. genB = gen0; break; } } } catch (System.IO.IOException err2) { // will retry } finally { genInput.Close(); } } try { System.Threading.Thread.Sleep(new System.TimeSpan((System.Int64) 10000 * Mono.Lucene.Net.Index.SegmentInfos.defaultGenFileRetryPauseMsec)); } catch (System.Threading.ThreadInterruptedException ie) { // In 3.0 we will change this to throw // InterruptedException instead SupportClass.ThreadClass.Current().Interrupt(); throw new System.SystemException(ie.Message, ie); } } Mono.Lucene.Net.Index.SegmentInfos.Message(IndexFileNames.SEGMENTS_GEN + " check: genB=" + genB); // Pick the larger of the two gen's: if (genA > genB) { gen = genA; } else { gen = genB; } if (gen == -1) { // Neither approach found a generation System.String s; if (files != null) { s = ""; for (int i = 0; i < files.Length; i++) { s += (" " + files[i]); } } else { s = " null"; } throw new System.IO.FileNotFoundException("no segments* file found in " + directory + ": files:" + s); } } // Third method (fallback if first & second methods // are not reliable): since both directory cache and // file contents cache seem to be stale, just // advance the generation. if (1 == method || (0 == method && lastGen == gen && retry)) { method = 1; if (genLookaheadCount < Mono.Lucene.Net.Index.SegmentInfos.defaultGenLookaheadCount) { gen++; genLookaheadCount++; Mono.Lucene.Net.Index.SegmentInfos.Message("look ahead increment gen to " + gen); } } if (lastGen == gen) { // This means we're about to try the same // segments_N last tried. This is allowed, // exactly once, because writer could have been in // the process of writing segments_N last time. if (retry) { // OK, we've tried the same segments_N file // twice in a row, so this must be a real // error. We throw the original exception we // got. throw exc; } else { retry = true; } } else if (0 == method) { // Segment file has advanced since our last loop, so // reset retry: retry = false; } lastGen = gen; segmentFileName = IndexFileNames.FileNameFromGeneration(IndexFileNames.SEGMENTS, "", gen); try { System.Object v = DoBody(segmentFileName); Mono.Lucene.Net.Index.SegmentInfos.Message("success on " + segmentFileName); return(v); } catch (System.IO.IOException err) { // Save the original root cause: if (exc == null) { exc = err; } Mono.Lucene.Net.Index.SegmentInfos.Message("primary Exception on '" + segmentFileName + "': " + err + "'; will retry: retry=" + retry + "; gen = " + gen); if (!retry && gen > 1) { // This is our first time trying this segments // file (because retry is false), and, there is // possibly a segments_(N-1) (because gen > 1). // So, check if the segments_(N-1) exists and // try it if so: System.String prevSegmentFileName = IndexFileNames.FileNameFromGeneration(IndexFileNames.SEGMENTS, "", gen - 1); bool prevExists; prevExists = directory.FileExists(prevSegmentFileName); if (prevExists) { Mono.Lucene.Net.Index.SegmentInfos.Message("fallback to prior segment file '" + prevSegmentFileName + "'"); try { System.Object v = DoBody(prevSegmentFileName); if (exc != null) { Mono.Lucene.Net.Index.SegmentInfos.Message("success on fallback " + prevSegmentFileName); } return(v); } catch (System.IO.IOException err2) { Mono.Lucene.Net.Index.SegmentInfos.Message("secondary Exception on '" + prevSegmentFileName + "': " + err2 + "'; will retry"); } } } } } }