Java Code Examples for org.apache.lucene.codecs.CodecUtil#checkHeader()

/**
 * Reads blob with specified name without resolving the blobName using using {@link #blobName} method.
 *
 * @param blobContainer blob container
 * @param blobName blob name
 */
public T readBlob(BlobContainer blobContainer, String blobName) throws IOException {
    try (InputStream inputStream = blobContainer.readBlob(blobName)) {
        byte[] bytes = ByteStreams.toByteArray(inputStream);
        final String resourceDesc = "ChecksumBlobStoreFormat.readBlob(blob=\"" + blobName + "\")";
        try (ByteArrayIndexInput indexInput = new ByteArrayIndexInput(resourceDesc, bytes)) {
            CodecUtil.checksumEntireFile(indexInput);
            CodecUtil.checkHeader(indexInput, codec, VERSION, VERSION);
            long filePointer = indexInput.getFilePointer();
            long contentSize = indexInput.length() - CodecUtil.footerLength() - filePointer;
            BytesReference bytesReference = new BytesArray(bytes, (int) filePointer, (int) contentSize);
            return read(bytesReference);
        } catch (CorruptIndexException | IndexFormatTooOldException | IndexFormatTooNewException ex) {
            // we trick this into a dedicated exception with the original stacktrace
            throw new CorruptStateException(ex);
        }
    }
}
 

/**
 * Reads the state from a given file and compares the expected version against the actual version of
 * the state.
 */
public final T read(Path file) throws IOException {
    try (Directory dir = newDirectory(file.getParent())) {
        try (final IndexInput indexInput = dir.openInput(file.getFileName().toString(), IOContext.DEFAULT)) {
             // We checksum the entire file before we even go and parse it. If it's corrupted we barf right here.
            CodecUtil.checksumEntireFile(indexInput);
            CodecUtil.checkHeader(indexInput, STATE_FILE_CODEC, STATE_FILE_VERSION, STATE_FILE_VERSION);
            final XContentType xContentType = XContentType.values()[indexInput.readInt()];
            indexInput.readLong(); // version currently unused
            long filePointer = indexInput.getFilePointer();
            long contentSize = indexInput.length() - CodecUtil.footerLength() - filePointer;
            try (IndexInput slice = indexInput.slice("state_xcontent", filePointer, contentSize)) {
                try (XContentParser parser = XContentFactory.xContent(xContentType).createParser(new InputStreamIndexInput(slice, contentSize))) {
                    return fromXContent(parser);
                }
            }
        } catch(CorruptIndexException | IndexFormatTooOldException | IndexFormatTooNewException ex) {
            // we trick this into a dedicated exception with the original stacktrace
            throw new CorruptStateException(ex);
        }
    }
}
 

@Override
public boolean load(DataInput input) throws IOException {
  CodecUtil.checkHeader(input, CODEC_NAME, VERSION_START, VERSION_START);
  count = input.readVLong();
  byte separatorOrig = input.readByte();
  if (separatorOrig != separator) {
    throw new IllegalStateException("separator=" + separator + " is incorrect: original model was built with separator=" + separatorOrig);
  }
  int gramsOrig = input.readVInt();
  if (gramsOrig != grams) {
    throw new IllegalStateException("grams=" + grams + " is incorrect: original model was built with grams=" + gramsOrig);
  }
  totTokens = input.readVLong();

  fst = new FST<>(input, input, PositiveIntOutputs.getSingleton());

  return true;
}
 

@Override
public FieldsProducer fieldsProducer(SegmentReadState readState)
  throws IOException {

  // Load our ID:
  final String idFileName = IndexFileNames.segmentFileName(readState.segmentInfo.name, readState.segmentSuffix, ID_EXTENSION);
  IndexInput in = readState.directory.openInput(idFileName, readState.context);
  boolean success = false;
  final int id;
  try {
    CodecUtil.checkHeader(in, RAM_ONLY_NAME, VERSION_START, VERSION_LATEST);
    id = in.readVInt();
    success = true;
  } finally {
    if (!success) {
      IOUtils.closeWhileHandlingException(in);
    } else {
      IOUtils.close(in);
    }
  }
  
  synchronized(state) {
    return state.get(id);
  }
}
 

/**
 * Reads blob with specified name without resolving the blobName using using {@link #blobName} method.
 *
 * @param blobContainer blob container
 * @param blobName blob name
 */
public T readBlob(BlobContainer blobContainer, String blobName) throws IOException {
    final BytesReference bytes = Streams.readFully(blobContainer.readBlob(blobName));
    final String resourceDesc = "ChecksumBlobStoreFormat.readBlob(blob=\"" + blobName + "\")";
    try (ByteArrayIndexInput indexInput =
             new ByteArrayIndexInput(resourceDesc, BytesReference.toBytes(bytes))) {
        CodecUtil.checksumEntireFile(indexInput);
        CodecUtil.checkHeader(indexInput, codec, VERSION, VERSION);
        long filePointer = indexInput.getFilePointer();
        long contentSize = indexInput.length() - CodecUtil.footerLength() - filePointer;
        try (XContentParser parser = XContentHelper.createParser(namedXContentRegistry, LoggingDeprecationHandler.INSTANCE,
                                                                 bytes.slice((int) filePointer, (int) contentSize), XContentType.SMILE)) {
            return reader.apply(parser);
        }
    } catch (CorruptIndexException | IndexFormatTooOldException | IndexFormatTooNewException ex) {
        // we trick this into a dedicated exception with the original stacktrace
        throw new CorruptStateException(ex);
    }
}
 

public CompletionFieldsProducer(SegmentReadState state) throws IOException {
    String suggestFSTFile = IndexFileNames.segmentFileName(state.segmentInfo.name, state.segmentSuffix, EXTENSION);
    IndexInput input = state.directory.openInput(suggestFSTFile, state.context);
    version = CodecUtil.checkHeader(input, CODEC_NAME, SUGGEST_CODEC_VERSION, SUGGEST_VERSION_CURRENT);
    FieldsProducer delegateProducer = null;
    boolean success = false;
    try {
        PostingsFormat delegatePostingsFormat = PostingsFormat.forName(input.readString());
        String providerName = input.readString();
        CompletionLookupProvider completionLookupProvider = providers.get(providerName);
        if (completionLookupProvider == null) {
            throw new IllegalStateException("no provider with name [" + providerName + "] registered");
        }
        // TODO: we could clone the ReadState and make it always forward IOContext.MERGE to prevent unecessary heap usage?
        delegateProducer = delegatePostingsFormat.fieldsProducer(state);
        /*
         * If we are merging we don't load the FSTs at all such that we
         * don't consume so much memory during merge
         */
        if (state.context.context != Context.MERGE) {
            // TODO: maybe we can do this in a fully lazy fashion based on some configuration
            // eventually we should have some kind of curciut breaker that prevents us from going OOM here
            // with some configuration
            this.lookupFactory = completionLookupProvider.load(input);
        } else {
            this.lookupFactory = null;
        }
        this.delegateProducer = delegateProducer;
        success = true;
    } finally {
        if (!success) {
            IOUtils.closeWhileHandlingException(delegateProducer, input);
        } else {
            IOUtils.close(input);
        }
    }
}
 

@Override
public SegmentInfo read(Directory dir, String segment, IOContext context) throws IOException {
  final String fileName = IndexFileNames.segmentFileName(segment, "", Blur022SegmentInfoFormat.SI_EXTENSION);
  final IndexInput input = dir.openInput(fileName, context);
  boolean success = false;
  try {
    CodecUtil.checkHeader(input, Blur022SegmentInfoFormat.CODEC_NAME, Blur022SegmentInfoFormat.VERSION_START,
        Blur022SegmentInfoFormat.VERSION_CURRENT);
    final String version = input.readString();
    final int docCount = input.readInt();
    if (docCount < 0) {
      throw new CorruptIndexException("invalid docCount: " + docCount + " (resource=" + input + ")");
    }
    final boolean isCompoundFile = input.readByte() == SegmentInfo.YES;
    final Map<String, String> diagnostics = input.readStringStringMap();
    final Map<String, String> attributes = input.readStringStringMap();
    final Set<String> files = input.readStringSet();

    if (input.getFilePointer() != input.length()) {
      throw new CorruptIndexException("did not read all bytes from file \"" + fileName + "\": read "
          + input.getFilePointer() + " vs size " + input.length() + " (resource: " + input + ")");
    }

    final SegmentInfo si = new SegmentInfo(dir, version, segment, docCount, isCompoundFile, null, diagnostics,
        Collections.unmodifiableMap(attributes));
    si.setFiles(files);

    success = true;

    return si;

  } finally {
    if (!success) {
      IOUtils.closeWhileHandlingException(input);
    } else {
      input.close();
    }
  }
}
 

DiskDocValuesProducer(SegmentReadState state, String dataCodec, String dataExtension, String metaCodec,
    String metaExtension) throws IOException {
  String metaName = IndexFileNames.segmentFileName(state.segmentInfo.name, state.segmentSuffix, metaExtension);
  // read in the entries from the metadata file.
  IndexInput in = state.directory.openInput(metaName, state.context);
  boolean success = false;
  try {
    CodecUtil.checkHeader(in, metaCodec, DiskDocValuesFormat.VERSION_START, DiskDocValuesFormat.VERSION_START);
    numerics = new ConcurrentHashMap<Integer, NumericEntry>();
    ords = new ConcurrentHashMap<Integer, NumericEntry>();
    ordIndexes = new ConcurrentHashMap<Integer, NumericEntry>();
    binaries = new ConcurrentHashMap<Integer, BinaryEntry>();
    _binaryDocValuesCache = new ConcurrentHashMap<Integer, BinaryDocValues>();
    _numericDocValuesCache = new ConcurrentHashMap<Integer, NumericDocValues>();
    _sortedDocValuesCache = new ConcurrentHashMap<Integer, SortedDocValues>();
    _sortedSetDocValuesCache = new ConcurrentHashMap<Integer, SortedSetDocValues>();
    readFields(in, state.fieldInfos);
    success = true;
  } finally {
    if (success) {
      IOUtils.close(in);
    } else {
      IOUtils.closeWhileHandlingException(in);
    }
  }

  String dataName = IndexFileNames.segmentFileName(state.segmentInfo.name, state.segmentSuffix, dataExtension);
  data = state.directory.openInput(dataName, state.context);
  CodecUtil.checkHeader(data, dataCodec, DiskDocValuesFormat.VERSION_START, DiskDocValuesFormat.VERSION_START);
}
 

/**
 * Reads the state from a given file and compares the expected version against the actual version of
 * the state.
 */
public final T read(NamedXContentRegistry namedXContentRegistry, Path file) throws IOException {
    try (Directory dir = newDirectory(file.getParent())) {
        try (IndexInput indexInput = dir.openInput(file.getFileName().toString(), IOContext.DEFAULT)) {
            // We checksum the entire file before we even go and parse it. If it's corrupted we barf right here.
            CodecUtil.checksumEntireFile(indexInput);
            CodecUtil.checkHeader(indexInput, STATE_FILE_CODEC, MIN_COMPATIBLE_STATE_FILE_VERSION, STATE_FILE_VERSION);
            final XContentType xContentType = XContentType.values()[indexInput.readInt()];
            if (xContentType != FORMAT) {
                throw new IllegalStateException("expected state in " + file + " to be " + FORMAT + " format but was " + xContentType);
            }
            long filePointer = indexInput.getFilePointer();
            long contentSize = indexInput.length() - CodecUtil.footerLength() - filePointer;
            try (IndexInput slice = indexInput.slice("state_xcontent", filePointer, contentSize)) {
                try (InputStreamIndexInput in = new InputStreamIndexInput(slice, contentSize)) {
                    try (XContentParser parser = XContentFactory.xContent(FORMAT)
                            .createParser(namedXContentRegistry, LoggingDeprecationHandler.INSTANCE,
                                in)) {
                        return fromXContent(parser);
                    }
                }
            }
        } catch (CorruptIndexException | IndexFormatTooOldException | IndexFormatTooNewException ex) {
            // we trick this into a dedicated exception with the original stacktrace
            throw new CorruptStateException(ex);
        }
    }
}
 

void writeMetaData() throws IOException {
  if (totalChunks == 0) {
    return;
  }
  
  long startDMW = data.getFilePointer();
  meta.writeLong(startDMW);
  
  meta.writeVInt(totalChunks);
  meta.writeVInt(Lucene80DocValuesFormat.BINARY_BLOCK_SHIFT);
  meta.writeVInt(maxUncompressedBlockLength);
  meta.writeVInt(DIRECT_MONOTONIC_BLOCK_SHIFT);
  

  CodecUtil.writeFooter(tempBinaryOffsets);
  IOUtils.close(tempBinaryOffsets);             
  //write the compressed block offsets info to the meta file by reading from temp file
  try (ChecksumIndexInput filePointersIn = state.directory.openChecksumInput(tempBinaryOffsets.getName(), IOContext.READONCE)) {
    CodecUtil.checkHeader(filePointersIn, Lucene80DocValuesFormat.META_CODEC + "FilePointers", Lucene80DocValuesFormat.VERSION_CURRENT,
      Lucene80DocValuesFormat.VERSION_CURRENT);
    Throwable priorE = null;
    try {
      final DirectMonotonicWriter filePointers = DirectMonotonicWriter.getInstance(meta, data, totalChunks, DIRECT_MONOTONIC_BLOCK_SHIFT);
      long fp = blockAddressesStart;
      for (int i = 0; i < totalChunks; ++i) {
        filePointers.add(fp);
        fp += filePointersIn.readVLong();
      }
      if (maxPointer < fp) {
        throw new CorruptIndexException("File pointers don't add up ("+fp+" vs expected "+maxPointer+")", filePointersIn);
      }
      filePointers.finish();
    } catch (Throwable e) {
      priorE = e;
    } finally {
      CodecUtil.checkFooter(filePointersIn, priorE);
    }
  }
  // Write the length of the DMW block in the data 
  meta.writeLong(data.getFilePointer() - startDMW);
}