/// <summary> Closes the underlying <see cref="Lucene.Net.Store.IndexInput" /> streams, including any ones associated with a
/// lazy implementation of a Field. This means that the Fields values will not be accessible.
///
/// </summary>
/// <throws> IOException </throws>
public void Dispose()
{
// Move to protected method if class becomes unsealed
if (!closed)
{
if (fieldsStream != null)
{
fieldsStream.Close();
}
if (isOriginal)
{
if (cloneableFieldsStream != null)
{
cloneableFieldsStream.Close();
}
if (cloneableIndexStream != null)
{
cloneableIndexStream.Close();
}
}
if (indexStream != null)
{
indexStream.Close();
}
fieldsStreamTL.Close();
closed = true;
}
}
/// <summary>
/// Read the segments.gen file to get the generation number.
/// </summary>
private void ReadDirectory()
{
IndexInput genInput = null;
try
{
genInput = directory.OpenInput(IndexFileNames.SEGMENTS_GEN);
if (genInput != null)
{
int version = genInput.ReadInt();
if (version == Lucene.Net.Index.SegmentInfos.FORMAT_LOCKLESS)
{
long gen0 = genInput.ReadLong();
long gen1 = genInput.ReadLong();
if (gen0 == gen1)
{
// The file is consistent, use the generation
this.generation = gen0;
}
}
}
}
finally
{
genInput.Close();
}
}
/// <summary>
/// Closes the file.
/// </summary>
public override void Close()
{
_indexInput.Close();
_indexInput = null;
_sqlServerDirectory = null;
GC.SuppressFinalize(this);
}
public override object DoBody(string segmentFileName)
{
IndexInput i = dir.OpenInput(segmentFileName);
var indexFormat = i.ReadInt();
i.Close();
return(indexFormat);
}
public override void Close()
{
_fileMutex.WaitOne();
try
{
var fileName = _name;
//make sure it's all written out
//we are only checking for null here in case Close is called multiple times
if (_cacheDirIndexOutput != null)
{
_cacheDirIndexOutput.Flush();
_cacheDirIndexOutput.Close();
IndexInput cacheInput = null;
try
{
cacheInput = CacheDirectory.OpenInput(fileName);
}
catch (IOException e)
{
//This would occur if the file doesn't exist! we previously threw when that happens so we'll keep
//doing that for now but this is quicker than first checking if it exists and then opening it.
throw;
}
if (cacheInput != null)
{
IndexOutput masterOutput = null;
try
{
masterOutput = MasterDirectory.CreateOutput(fileName);
cacheInput.CopyTo(masterOutput, fileName);
}
finally
{
masterOutput?.Close();
cacheInput?.Close();
}
}
#if FULLDEBUG
Trace.WriteLine($"CLOSED WRITESTREAM {_name}");
#endif
// clean up
_cacheDirIndexOutput = null;
}
GC.SuppressFinalize(this);
}
finally
{
_fileMutex.ReleaseMutex();
}
}
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 {
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)
{
}
}
}
}
/// <summary> Construct a FieldInfos object using the directory and the name of the file
/// IndexInput
/// </summary>
/// <param name="d">The directory to open the IndexInput from
/// </param>
/// <param name="name">The name of the file to open the IndexInput from in the Directory
/// </param>
/// <throws> IOException </throws>
public /*internal*/ FieldInfos(Directory d, String name)
{
IndexInput input = d.OpenInput(name);
try
{
try
{
Read(input, name);
}
catch (System.IO.IOException)
{
if (format == FORMAT_PRE)
{
// LUCENE-1623: FORMAT_PRE (before there was a
// format) may be 2.3.2 (pre-utf8) or 2.4.x (utf8)
// encoding; retry with input set to pre-utf8
input.Seek(0);
input.SetModifiedUTF8StringsMode();
byNumber.Clear();
byName.Clear();
bool rethrow = false;
try
{
Read(input, name);
}
catch (Exception)
{
// Ignore any new exception & set to throw original IOE
rethrow = true;
}
if (rethrow)
{
// Preserve stack trace
throw;
}
}
else
{
// The IOException cannot be caused by
// LUCENE-1623, so re-throw it
throw;
}
}
}
finally
{
input.Close();
}
}
///<summary>
/// Added safety check on last batch of additions for < 3.0 based Lucene implementations
/// <see cref="https://issues.apache.org/jira/browse/LUCENE-3255"/>
///</summary>
///<param name="dir">return true if OK false if a potentially corrupted segments file was removed</param>
public static bool CheckLastCommitOk(FSDirectory dir)
{
string latestSeg = SegmentInfos.GetCurrentSegmentFileName(dir);
IndexInput input = dir.OpenInput(latestSeg);
int format = input.ReadInt();
input.Close();
if (format != SegmentInfos.FORMAT_DIAGNOSTICS)
{
dir.DeleteFile(latestSeg);
return(false);
}
return(true);
}
/// <summary>Copy the contents of the file with specified extension into the
/// provided output stream. Use the provided buffer for moving data
/// to reduce memory allocation.
/// </summary>
private void CopyFile(FileEntry source, IndexOutput os, byte[] buffer)
{
IndexInput isRenamed = null;
try
{
long startPtr = os.FilePointer;
isRenamed = directory.OpenInput(source.file);
long length = isRenamed.Length();
long remainder = length;
int chunk = buffer.Length;
while (remainder > 0)
{
var len = (int)Math.Min(chunk, remainder);
isRenamed.ReadBytes(buffer, 0, len, false);
os.WriteBytes(buffer, len);
remainder -= len;
if (checkAbort != null)
{
// Roughly every 2 MB we will check if
// it's time to abort
checkAbort.Work(80);
}
}
// Verify that remainder is 0
if (remainder != 0)
{
throw new System.IO.IOException("Non-zero remainder length after copying: " + remainder + " (id: " + source.file + ", length: " + length + ", buffer size: " + chunk + ")");
}
// Verify that the output length diff is equal to original file
long endPtr = os.FilePointer;
long diff = endPtr - startPtr;
if (diff != length)
{
throw new System.IO.IOException("Difference in the output file offsets " + diff + " does not match the original file length " + length);
}
}
finally
{
if (isRenamed != null)
{
isRenamed.Close();
}
}
}
protected override void Dispose(bool disposing)
{
if (isDisposed)
{
return;
}
if (disposing)
{
if (base_Renamed != null)
{
base_Renamed.Close();
}
}
isDisposed = true;
}
public override void Close()
{
_fileMutex.WaitOne();
try
{
#if FULLDEBUG
Trace.WriteLine($"CLOSED READSTREAM local {_name}");
#endif
_cacheDirIndexInput.Close();
_cacheDirIndexInput = null;
_syncDirectory = null;
GC.SuppressFinalize(this);
}
finally
{
_fileMutex.ReleaseMutex();
}
}
public override void Close()
{
_fileMutex.WaitOne();
try
{
#if FULLDEBUG
Trace.WriteLine(String.Format("CLOSED READSTREAM local {0}", _name));
#endif
_indexInput.Close();
_indexInput = null;
_syncDirectory = null;
GC.SuppressFinalize(this);
}
finally
{
_fileMutex.ReleaseMutex();
}
}
private MemoryStream CompressStream(string fileName, long originalLength)
{
// unfortunately, deflate stream doesn't allow seek, and we need a seekable stream
// to pass to the blob storage stuff, so we compress into a memory stream
MemoryStream compressedStream = new MemoryStream();
IndexInput indexInput = null;
try
{
indexInput = CacheDirectory.OpenInput(fileName);
using (var compressor = new DeflateStream(compressedStream, CompressionMode.Compress, true))
{
// compress to compressedOutputStream
byte[] bytes = new byte[indexInput.Length()];
indexInput.ReadBytes(bytes, 0, (int)bytes.Length);
compressor.Write(bytes, 0, (int)bytes.Length);
}
// seek back to beginning of comrpessed stream
compressedStream.Seek(0, SeekOrigin.Begin);
Debug.WriteLine(string.Format("COMPRESSED {0} -> {1} {2}% to {3}",
originalLength,
compressedStream.Length,
((float)compressedStream.Length / (float)originalLength) * 100,
_name));
}
catch
{
// release the compressed stream resources if an error occurs
compressedStream.Dispose();
throw;
}
finally
{
if (indexInput != null)
{
indexInput.Close();
}
}
return(compressedStream);
}
/// <summary>
/// This will sync the requested file from master storage to the local fast cache storage
/// </summary>
/// <param name="fileName"></param>
private void SyncLocally(string fileName)
{
//get the master file stream
IndexInput masterInput = null;
try
{
masterInput = MasterDirectory.OpenInput(fileName);
}
catch (IOException ex)
{
//this will be a file not found (FileNotFoundException)
//TODO: It has been seen that OpenInput on the master can throw an exception due to a lucene file not found - which is very odd
// we need to check if the master is being written first before the sync dir. And if the file does not exist in the master,
// or the sync dir, then something has gone wrong, that shouldn't happen and we'll need to deal with that differently
// because the index will be in a state where it's just not readable.
//Hrmmm what to do? There's actually nothing that can be done :/ if we return false here then the instance of this item would be null
//which will then cause exceptions further on and take down the app pool anyways. I've looked through the Lucene source and there
//is no safety net to check against this situation, it just happily throws exceptions on a background thread.
_loggingService.Log(new LogEntry(LogLevel.Error, ex, $"File not found"));
throw ex;
}
if (masterInput != null)
{
IndexOutput cacheOutput = null;
try
{
cacheOutput = CacheDirectory.CreateOutput(fileName);
masterInput.CopyTo(cacheOutput, fileName);
}
finally
{
cacheOutput?.Close();
masterInput?.Close();
}
}
}
/// <summary>Constructs a bit vector from the file <c>name</c> in Directory
/// <c>d</c>, as written by the <see cref="Write" /> method.
/// </summary>
public BitVector(Directory d, System.String name)
{
IndexInput input = d.OpenInput(name);
try
{
size = input.ReadInt(); // read size
if (size == -1)
{
ReadDgaps(input);
}
else
{
ReadBits(input);
}
}
finally
{
input.Close();
}
}
public static void Main(System.String[] args)
{
System.String filename = null;
bool extract = false;
for (int i = 0; i < args.Length; ++i)
{
if (args[i].Equals("-extract"))
{
extract = true;
}
else if (filename == null)
{
filename = args[i];
}
}
if (filename == null)
{
System.Console.Out.WriteLine("Usage: Lucene.Net.Index.IndexReader [-extract] <cfsfile>");
return;
}
Directory dir = null;
CompoundFileReader cfr = null;
try
{
System.IO.FileInfo file = new System.IO.FileInfo(filename);
System.String dirname = new System.IO.FileInfo(file.FullName).DirectoryName;
filename = file.Name;
dir = FSDirectory.GetDirectory(dirname);
cfr = new CompoundFileReader(dir, filename);
System.String[] files = cfr.List();
System.Array.Sort(files); // sort the array of filename so that the output is more readable
for (int i = 0; i < files.Length; ++i)
{
long len = cfr.FileLength(files[i]);
if (extract)
{
System.Console.Out.WriteLine("extract " + files[i] + " with " + len + " bytes to local directory...");
IndexInput ii = cfr.OpenInput(files[i]);
System.IO.FileStream f = new System.IO.FileStream(files[i], System.IO.FileMode.Create);
// read and write with a small buffer, which is more effectiv than reading byte by byte
byte[] buffer = new byte[1024];
int chunk = buffer.Length;
while (len > 0)
{
int bufLen = (int)System.Math.Min(chunk, len);
ii.ReadBytes(buffer, 0, bufLen);
f.Write(buffer, 0, bufLen);
len -= bufLen;
}
f.Close();
ii.Close();
}
else
{
System.Console.Out.WriteLine(files[i] + ": " + len + " bytes");
}
}
}
catch (System.IO.IOException ioe)
{
System.Console.Error.WriteLine(ioe.StackTrace);
}
finally
{
try
{
if (dir != null)
{
dir.Close();
}
if (cfr != null)
{
cfr.Close();
}
}
catch (System.IO.IOException ioe)
{
System.Console.Error.WriteLine(ioe.StackTrace);
}
}
}
protected override void Dispose(bool disposing)
{
_fileMutex.WaitOne();
try
{
string fileName = _name;
// make sure it's all written out
_indexOutput.Flush();
long originalLength = _indexOutput.Length;
_indexOutput.Dispose();
Stream blobStream;
#if COMPRESSBLOBS
// optionally put a compressor around the blob stream
if (_azureDirectory.ShouldCompressFile(_name))
{
// unfortunately, deflate stream doesn't allow seek, and we need a seekable stream
// to pass to the blob storage stuff, so we compress into a memory stream
MemoryStream compressedStream = new MemoryStream();
try
{
IndexInput indexInput = CacheDirectory.OpenInput(fileName);
using (DeflateStream compressor = new DeflateStream(compressedStream, CompressionMode.Compress, true))
{
// compress to compressedOutputStream
byte[] bytes = new byte[indexInput.Length()];
indexInput.ReadBytes(bytes, 0, (int)bytes.Length);
compressor.Write(bytes, 0, (int)bytes.Length);
}
indexInput.Close();
// seek back to beginning of comrpessed stream
compressedStream.Seek(0, SeekOrigin.Begin);
Debug.WriteLine(string.Format("COMPRESSED {0} -> {1} {2}% to {3}",
originalLength,
compressedStream.Length,
((float)compressedStream.Length / (float)originalLength) * 100,
_name));
}
catch
{
// release the compressed stream resources if an error occurs
compressedStream.Dispose();
throw;
}
blobStream = compressedStream;
}
else
#endif
{
blobStream = new StreamInput(CacheDirectory.OpenInput(fileName));
}
try
{
// push the blobStream up to the cloud
_blob.UploadFromStream(blobStream);
// set the metadata with the original index file properties
_blob.Metadata["CachedLength"] = originalLength.ToString();
_blob.Metadata["CachedLastModified"] = CacheDirectory.FileModified(fileName).ToString();
_blob.SetMetadata();
Debug.WriteLine(string.Format("PUT {1} bytes to {0} in cloud", _name, blobStream.Length));
}
finally
{
blobStream.Dispose();
}
#if FULLDEBUG
Debug.WriteLine(string.Format("CLOSED WRITESTREAM {0}", _name));
#endif
// clean up
_indexOutput = null;
_blobContainer = null;
_blob = null;
GC.SuppressFinalize(this);
}
finally
{
_fileMutex.ReleaseMutex();
}
}
public override void Close()
{
main.Close();
}
public System.Object Run(IndexCommit commit)
{
if (commit != null)
{
if (directory != commit.Directory)
{
throw new System.IO.IOException("the specified commit does not match the specified Directory");
}
return(DoBody(commit.SegmentsFileName));
}
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 = Lucene.Net.Index.SegmentInfos.GetCurrentSegmentGeneration(files);
}
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 < Lucene.Net.Index.SegmentInfos.defaultGenFileRetryCount; i++)
{
IndexInput genInput = null;
try
{
genInput = directory.OpenInput(IndexFileNames.SEGMENTS_GEN);
}
catch (System.IO.FileNotFoundException e)
{
Lucene.Net.Index.SegmentInfos.Message("segments.gen open: FileNotFoundException " + e);
break;
}
catch (System.IO.IOException e)
{
Lucene.Net.Index.SegmentInfos.Message("segments.gen open: IOException " + e);
}
if (genInput != null)
{
try
{
int version = genInput.ReadInt();
if (version == Lucene.Net.Index.SegmentInfos.FORMAT_LOCKLESS)
{
long gen0 = genInput.ReadLong();
long gen1 = genInput.ReadLong();
Lucene.Net.Index.SegmentInfos.Message("fallback check: " + gen0 + "; " + gen1);
if (gen0 == gen1)
{
// The file is consistent.
genB = gen0;
break;
}
}
}
catch (System.IO.IOException)
{
// will retry
}
finally
{
genInput.Close();
}
}
System.Threading.Thread.Sleep(new TimeSpan((System.Int64) 10000 * Lucene.Net.Index.SegmentInfos.defaultGenFileRetryPauseMsec));
}
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)
{
throw new System.IO.FileNotFoundException("no segments* file found in " + directory + ": files:" + string.Join(" ", files));
}
}
// 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 < Lucene.Net.Index.SegmentInfos.defaultGenLookaheadCount)
{
gen++;
genLookaheadCount++;
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;
}
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);
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;
}
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)
{
Lucene.Net.Index.SegmentInfos.Message("fallback to prior segment file '" + prevSegmentFileName + "'");
try
{
System.Object v = DoBody(prevSegmentFileName);
if (exc != null)
{
Lucene.Net.Index.SegmentInfos.Message("success on fallback " + prevSegmentFileName);
}
return(v);
}
catch (System.IO.IOException err2)
{
Lucene.Net.Index.SegmentInfos.Message("secondary Exception on '" + prevSegmentFileName + "': " + err2 + "'; will retry");
}
}
}
}
}
}
/// <summary>Returns a <see cref="Status" /> instance detailing
/// the state of the index.
///
/// </summary>
/// <param name="onlySegments">list of specific segment names to check
///
/// <p/>As this method checks every byte in the specified
/// segments, on a large index it can take quite a long
/// time to run.
///
/// <p/><b>WARNING</b>: make sure
/// you only call this when the index is not opened by any
/// writer.
/// </param>
public virtual Status CheckIndex_Renamed_Method(List <string> onlySegments)
{
System.Globalization.NumberFormatInfo nf = System.Globalization.CultureInfo.CurrentCulture.NumberFormat;
SegmentInfos sis = new SegmentInfos();
Status result = new Status();
result.dir = dir;
try
{
sis.Read(dir);
}
catch (System.Exception t)
{
Msg("ERROR: could not read any segments file in directory");
result.missingSegments = true;
if (infoStream != null)
{
infoStream.WriteLine(t.StackTrace);
}
return(result);
}
int numSegments = sis.Count;
var segmentsFileName = sis.GetCurrentSegmentFileName();
IndexInput input = null;
try
{
input = dir.OpenInput(segmentsFileName);
}
catch (System.Exception t)
{
Msg("ERROR: could not open segments file in directory");
if (infoStream != null)
{
infoStream.WriteLine(t.StackTrace);
}
result.cantOpenSegments = true;
return(result);
}
int format = 0;
try
{
format = input.ReadInt();
}
catch (System.Exception t)
{
Msg("ERROR: could not read segment file version in directory");
if (infoStream != null)
{
infoStream.WriteLine(t.StackTrace);
}
result.missingSegmentVersion = true;
return(result);
}
finally
{
if (input != null)
{
input.Close();
}
}
System.String sFormat = "";
bool skip = false;
if (format == SegmentInfos.FORMAT)
{
sFormat = "FORMAT [Lucene Pre-2.1]";
}
if (format == SegmentInfos.FORMAT_LOCKLESS)
{
sFormat = "FORMAT_LOCKLESS [Lucene 2.1]";
}
else if (format == SegmentInfos.FORMAT_SINGLE_NORM_FILE)
{
sFormat = "FORMAT_SINGLE_NORM_FILE [Lucene 2.2]";
}
else if (format == SegmentInfos.FORMAT_SHARED_DOC_STORE)
{
sFormat = "FORMAT_SHARED_DOC_STORE [Lucene 2.3]";
}
else
{
if (format == SegmentInfos.FORMAT_CHECKSUM)
{
sFormat = "FORMAT_CHECKSUM [Lucene 2.4]";
}
else if (format == SegmentInfos.FORMAT_DEL_COUNT)
{
sFormat = "FORMAT_DEL_COUNT [Lucene 2.4]";
}
else if (format == SegmentInfos.FORMAT_HAS_PROX)
{
sFormat = "FORMAT_HAS_PROX [Lucene 2.4]";
}
else if (format == SegmentInfos.FORMAT_USER_DATA)
{
sFormat = "FORMAT_USER_DATA [Lucene 2.9]";
}
else if (format == SegmentInfos.FORMAT_DIAGNOSTICS)
{
sFormat = "FORMAT_DIAGNOSTICS [Lucene 2.9]";
}
else if (format < SegmentInfos.CURRENT_FORMAT)
{
sFormat = "int=" + format + " [newer version of Lucene than this tool]";
skip = true;
}
else
{
sFormat = format + " [Lucene 1.3 or prior]";
}
}
result.segmentsFileName = segmentsFileName;
result.numSegments = numSegments;
result.segmentFormat = sFormat;
result.userData = sis.UserData;
System.String userDataString;
if (sis.UserData.Count > 0)
{
userDataString = " userData=" + CollectionsHelper.CollectionToString(sis.UserData);
}
else
{
userDataString = "";
}
Msg("Segments file=" + segmentsFileName + " numSegments=" + numSegments + " version=" + sFormat + userDataString);
if (onlySegments != null)
{
result.partial = true;
if (infoStream != null)
{
infoStream.Write("\nChecking only these segments:");
}
foreach (string s in onlySegments)
{
if (infoStream != null)
{
infoStream.Write(" " + s);
}
}
result.segmentsChecked.AddRange(onlySegments);
Msg(":");
}
if (skip)
{
Msg("\nERROR: this index appears to be created by a newer version of Lucene than this tool was compiled on; please re-compile this tool on the matching version of Lucene; exiting");
result.toolOutOfDate = true;
return(result);
}
result.newSegments = (SegmentInfos)sis.Clone();
result.newSegments.Clear();
for (int i = 0; i < numSegments; i++)
{
SegmentInfo info = sis.Info(i);
if (onlySegments != null && !onlySegments.Contains(info.name))
{
continue;
}
var segInfoStat = new Status.SegmentInfoStatus();
result.segmentInfos.Add(segInfoStat);
Msg(" " + (1 + i) + " of " + numSegments + ": name=" + info.name + " docCount=" + info.docCount);
segInfoStat.name = info.name;
segInfoStat.docCount = info.docCount;
int toLoseDocCount = info.docCount;
SegmentReader reader = null;
try
{
Msg(" compound=" + info.GetUseCompoundFile());
segInfoStat.compound = info.GetUseCompoundFile();
Msg(" hasProx=" + info.HasProx);
segInfoStat.hasProx = info.HasProx;
Msg(" numFiles=" + info.Files().Count);
segInfoStat.numFiles = info.Files().Count;
Msg(System.String.Format(nf, " size (MB)={0:f}", new System.Object[] { (info.SizeInBytes() / (1024.0 * 1024.0)) }));
segInfoStat.sizeMB = info.SizeInBytes() / (1024.0 * 1024.0);
IDictionary <string, string> diagnostics = info.Diagnostics;
segInfoStat.diagnostics = diagnostics;
if (diagnostics.Count > 0)
{
Msg(" diagnostics = " + CollectionsHelper.CollectionToString(diagnostics));
}
int docStoreOffset = info.DocStoreOffset;
if (docStoreOffset != -1)
{
Msg(" docStoreOffset=" + docStoreOffset);
segInfoStat.docStoreOffset = docStoreOffset;
Msg(" docStoreSegment=" + info.DocStoreSegment);
segInfoStat.docStoreSegment = info.DocStoreSegment;
Msg(" docStoreIsCompoundFile=" + info.DocStoreIsCompoundFile);
segInfoStat.docStoreCompoundFile = info.DocStoreIsCompoundFile;
}
System.String delFileName = info.GetDelFileName();
if (delFileName == null)
{
Msg(" no deletions");
segInfoStat.hasDeletions = false;
}
else
{
Msg(" has deletions [delFileName=" + delFileName + "]");
segInfoStat.hasDeletions = true;
segInfoStat.deletionsFileName = delFileName;
}
if (infoStream != null)
{
infoStream.Write(" test: open reader.........");
}
reader = SegmentReader.Get(true, info, IndexReader.DEFAULT_TERMS_INDEX_DIVISOR);
segInfoStat.openReaderPassed = true;
int numDocs = reader.NumDocs();
toLoseDocCount = numDocs;
if (reader.HasDeletions)
{
if (reader.deletedDocs.Count() != info.GetDelCount())
{
throw new System.SystemException("delete count mismatch: info=" + info.GetDelCount() + " vs deletedDocs.count()=" + reader.deletedDocs.Count());
}
if (reader.deletedDocs.Count() > reader.MaxDoc)
{
throw new System.SystemException("too many deleted docs: MaxDoc=" + reader.MaxDoc + " vs deletedDocs.count()=" + reader.deletedDocs.Count());
}
if (info.docCount - numDocs != info.GetDelCount())
{
throw new System.SystemException("delete count mismatch: info=" + info.GetDelCount() + " vs reader=" + (info.docCount - numDocs));
}
segInfoStat.numDeleted = info.docCount - numDocs;
Msg("OK [" + (segInfoStat.numDeleted) + " deleted docs]");
}
else
{
if (info.GetDelCount() != 0)
{
throw new System.SystemException("delete count mismatch: info=" + info.GetDelCount() + " vs reader=" + (info.docCount - numDocs));
}
Msg("OK");
}
if (reader.MaxDoc != info.docCount)
{
throw new System.SystemException("SegmentReader.MaxDoc " + reader.MaxDoc + " != SegmentInfos.docCount " + info.docCount);
}
// Test getFieldNames()
if (infoStream != null)
{
infoStream.Write(" test: fields..............");
}
ICollection <string> fieldNames = reader.GetFieldNames(IndexReader.FieldOption.ALL);
Msg("OK [" + fieldNames.Count + " fields]");
segInfoStat.numFields = fieldNames.Count;
// Test Field Norms
segInfoStat.fieldNormStatus = TestFieldNorms(fieldNames, reader);
// Test the Term Index
segInfoStat.termIndexStatus = TestTermIndex(info, reader);
// Test Stored Fields
segInfoStat.storedFieldStatus = TestStoredFields(info, reader, nf);
// Test Term Vectors
segInfoStat.termVectorStatus = TestTermVectors(info, reader, nf);
// Rethrow the first exception we encountered
// This will cause stats for failed segments to be incremented properly
if (segInfoStat.fieldNormStatus.error != null)
{
throw new SystemException("Field Norm test failed");
}
else if (segInfoStat.termIndexStatus.error != null)
{
throw new SystemException("Term Index test failed");
}
else if (segInfoStat.storedFieldStatus.error != null)
{
throw new SystemException("Stored Field test failed");
}
else if (segInfoStat.termVectorStatus.error != null)
{
throw new System.SystemException("Term Vector test failed");
}
Msg("");
}
catch (System.Exception t)
{
Msg("FAILED");
const string comment = "fixIndex() would remove reference to this segment";
Msg(" WARNING: " + comment + "; full exception:");
if (infoStream != null)
{
infoStream.WriteLine(t.StackTrace);
}
Msg("");
result.totLoseDocCount += toLoseDocCount;
result.numBadSegments++;
continue;
}
finally
{
if (reader != null)
{
reader.Close();
}
}
// Keeper
result.newSegments.Add((SegmentInfo)info.Clone());
}
if (0 == result.numBadSegments)
{
result.clean = true;
Msg("No problems were detected with this index.\n");
}
else
{
Msg("WARNING: " + result.numBadSegments + " broken segments (containing " + result.totLoseDocCount + " documents) detected");
}
return(result);
}
protected virtual void Dispose(bool disposing)
{
if (isDisposed)
{
return;
}
if (disposing)
{
// make all effort to close up. Keep the first exception
// and throw it as a new one.
System.IO.IOException keep = null;
if (tvx != null)
{
try
{
tvx.Close();
}
catch (System.IO.IOException e)
{
if (keep == null)
{
keep = e;
}
}
}
if (tvd != null)
{
try
{
tvd.Close();
}
catch (System.IO.IOException e)
{
if (keep == null)
{
keep = e;
}
}
}
if (tvf != null)
{
try
{
tvf.Close();
}
catch (System.IO.IOException e)
{
if (keep == null)
{
keep = e;
}
}
}
if (keep != null)
{
throw new System.IO.IOException(keep.StackTrace);
}
}
isDisposed = true;
}
public override void Close()
{
input.Close();
}
