Java Code Examples for org.apache.lucene.index.FieldInfo#hasVectors()

static DocumentField of(FieldInfo finfo, IndexableField field, IndexReader reader, int docId)
    throws IOException {

  Objects.requireNonNull(finfo);
  Objects.requireNonNull(reader);

  DocumentField dfield = new DocumentField();

  dfield.name = finfo.name;
  dfield.idxOptions = finfo.getIndexOptions();
  dfield.hasTermVectors = finfo.hasVectors();
  dfield.hasPayloads = finfo.hasPayloads();
  dfield.hasNorms = finfo.hasNorms();

  if (finfo.hasNorms()) {
    NumericDocValues norms = MultiDocValues.getNormValues(reader, finfo.name);
    if (norms.advanceExact(docId)) {
      dfield.norm = norms.longValue();
    }
  }

  dfield.dvType = finfo.getDocValuesType();

  dfield.pointDimensionCount = finfo.getPointDimensionCount();
  dfield.pointNumBytes = finfo.getPointNumBytes();

  if (field != null) {
    dfield.isStored = field.fieldType().stored();
    dfield.stringValue = field.stringValue();
    if (field.binaryValue() != null) {
      dfield.binaryValue = BytesRef.deepCopyOf(field.binaryValue());
    }
    dfield.numericValue = field.numericValue();
  }

  return dfield;
}
 

@Override
public void setField(FieldInfo fieldInfo) {
  super.setField(fieldInfo);
  if (fieldInfo.getIndexOptions() != IndexOptions.DOCS_AND_FREQS_AND_POSITIONS) {
    throw new IllegalArgumentException("field must be index using IndexOptions.DOCS_AND_FREQS_AND_POSITIONS");
  }
  // LUCENE-5693: because CheckIndex cross-checks term vectors with postings even for deleted docs, and because our PF only indexes the
  // non-deleted documents on flush, CheckIndex will see this as corruption:
  if (fieldInfo.hasVectors()) {
    throw new IllegalArgumentException("field cannot index term vectors: CheckIndex will report this as index corruption");
  }
  lastState = emptyState;
}
 

@Override
public void write(Directory directory, SegmentInfo segmentInfo, String segmentSuffix, FieldInfos infos, IOContext context) throws IOException {
  final String fileName = IndexFileNames.segmentFileName(segmentInfo.name, segmentSuffix, EXTENSION);
  try (IndexOutput output = directory.createOutput(fileName, context)) {
    CodecUtil.writeIndexHeader(output, Lucene50FieldInfosFormat.CODEC_NAME, Lucene50FieldInfosFormat.FORMAT_CURRENT, segmentInfo.getId(), segmentSuffix);
    output.writeVInt(infos.size());
    for (FieldInfo fi : infos) {
      fi.checkConsistency();

      output.writeString(fi.name);
      output.writeVInt(fi.number);

      byte bits = 0x0;
      if (fi.hasVectors()) bits |= STORE_TERMVECTOR;
      if (fi.omitsNorms()) bits |= OMIT_NORMS;
      if (fi.hasPayloads()) bits |= STORE_PAYLOADS;
      output.writeByte(bits);

      output.writeByte(indexOptionsByte(fi.getIndexOptions()));

      // pack the DV type and hasNorms in one byte
      output.writeByte(docValuesByte(fi.getDocValuesType()));
      output.writeLong(fi.getDocValuesGen());
      output.writeMapOfStrings(fi.attributes());
    }
    CodecUtil.writeFooter(output);
  }
}
 

FilterFieldInfos(FieldInfos src, Collection<String> filterFields) {
  // Copy all the input FieldInfo objects since the field numbering must be kept consistent
  super(toArray(src));

  boolean hasVectors = false;
  boolean hasProx = false;
  boolean hasPayloads = false;
  boolean hasOffsets = false;
  boolean hasFreq = false;
  boolean hasNorms = false;
  boolean hasDocValues = false;
  boolean hasPointValues = false;

  this.filteredNames = new HashSet<>(filterFields);
  this.filtered = new ArrayList<>(filterFields.size());
  for (FieldInfo fi : src) {
    if (this.filteredNames.contains(fi.name)) {
      this.filtered.add(fi);
      hasVectors |= fi.hasVectors();
      hasProx |= fi.getIndexOptions().compareTo(IndexOptions.DOCS_AND_FREQS_AND_POSITIONS) >= 0;
      hasFreq |= fi.getIndexOptions() != IndexOptions.DOCS;
      hasOffsets |= fi.getIndexOptions().compareTo(IndexOptions.DOCS_AND_FREQS_AND_POSITIONS_AND_OFFSETS) >= 0;
      hasNorms |= fi.hasNorms();
      hasDocValues |= fi.getDocValuesType() != DocValuesType.NONE;
      hasPayloads |= fi.hasPayloads();
      hasPointValues |= (fi.getPointDimensionCount() != 0);
    }
  }

  this.filteredHasVectors = hasVectors;
  this.filteredHasProx = hasProx;
  this.filteredHasPayloads = hasPayloads;
  this.filteredHasOffsets = hasOffsets;
  this.filteredHasFreq = hasFreq;
  this.filteredHasNorms = hasNorms;
  this.filteredHasDocValues = hasDocValues;
  this.filteredHasPointValues = hasPointValues;
}
 

@Override
public void write(Directory directory, SegmentInfo segmentInfo, String segmentSuffix, FieldInfos infos, IOContext context) throws IOException {
  final String fileName = IndexFileNames.segmentFileName(segmentInfo.name, segmentSuffix, EXTENSION);
  try (IndexOutput output = directory.createOutput(fileName, context)) {
    CodecUtil.writeIndexHeader(output, Lucene60FieldInfosFormat.CODEC_NAME, Lucene60FieldInfosFormat.FORMAT_CURRENT, segmentInfo.getId(), segmentSuffix);
    output.writeVInt(infos.size());
    for (FieldInfo fi : infos) {
      fi.checkConsistency();

      output.writeString(fi.name);
      output.writeVInt(fi.number);

      byte bits = 0x0;
      if (fi.hasVectors()) bits |= STORE_TERMVECTOR;
      if (fi.omitsNorms()) bits |= OMIT_NORMS;
      if (fi.hasPayloads()) bits |= STORE_PAYLOADS;
      if (fi.isSoftDeletesField()) bits |= SOFT_DELETES_FIELD;
      output.writeByte(bits);

      output.writeByte(indexOptionsByte(fi.getIndexOptions()));

      // pack the DV type and hasNorms in one byte
      output.writeByte(docValuesByte(fi.getDocValuesType()));
      output.writeLong(fi.getDocValuesGen());
      output.writeMapOfStrings(fi.attributes());
      output.writeVInt(fi.getPointDimensionCount());
      if (fi.getPointDimensionCount() != 0) {
        output.writeVInt(fi.getPointIndexDimensionCount());
        output.writeVInt(fi.getPointNumBytes());
      }
    }
    CodecUtil.writeFooter(output);
  }
}
 

ReaderWrapper(LeafReader leafReader, String field) {
  super(leafReader);

  // TODO can we just do "field" and not bother with the other fields?
  List<FieldInfo> newInfos = new ArrayList<>(in.getFieldInfos().size());
  for (FieldInfo fieldInfo : in.getFieldInfos()) {
    if (fieldInfo.name.equals(field)) {
      FieldInfo f = new FieldInfo(fieldInfo.name,
          fieldInfo.number,
          fieldInfo.hasVectors(),
          fieldInfo.hasNorms(),
          fieldInfo.hasPayloads(),
          fieldInfo.getIndexOptions(),
          DocValuesType.NONE,
          fieldInfo.getDocValuesGen(),
          fieldInfo.attributes(),
          fieldInfo.getPointDimensionCount(),
          fieldInfo.getPointIndexDimensionCount(),
          fieldInfo.getPointNumBytes(),
          fieldInfo.isSoftDeletesField());
      newInfos.add(f);
    } else {
      newInfos.add(fieldInfo);
    }
  }
  FieldInfos infos = new FieldInfos(newInfos.toArray(new FieldInfo[newInfos.size()]));
  this.fieldInfos = infos;
}
 

/**
 * Determine the offset source for the specified field.  The default algorithm is as follows:
 * <ol>
 * <li>This calls {@link #getFieldInfo(String)}. Note this returns null if there is no searcher or if the
 * field isn't found there.</li>
 * <li> If there's a field info it has
 * {@link IndexOptions#DOCS_AND_FREQS_AND_POSITIONS_AND_OFFSETS} then {@link OffsetSource#POSTINGS} is
 * returned.</li>
 * <li>If there's a field info and {@link FieldInfo#hasVectors()} then {@link OffsetSource#TERM_VECTORS} is
 * returned (note we can't check here if the TV has offsets; if there isn't then an exception will get thrown
 * down the line).</li>
 * <li>Fall-back: {@link OffsetSource#ANALYSIS} is returned.</li>
 * </ol>
 * <p>
 * Note that the highlighter sometimes switches to something else based on the query, such as if you have
 * {@link OffsetSource#POSTINGS_WITH_TERM_VECTORS} but in fact don't need term vectors.
 */
protected OffsetSource getOffsetSource(String field) {
  FieldInfo fieldInfo = getFieldInfo(field);
  if (fieldInfo != null) {
    if (fieldInfo.getIndexOptions() == IndexOptions.DOCS_AND_FREQS_AND_POSITIONS_AND_OFFSETS) {
      return fieldInfo.hasVectors() ? OffsetSource.POSTINGS_WITH_TERM_VECTORS : OffsetSource.POSTINGS;
    }
    if (fieldInfo.hasVectors()) { // unfortunately we can't also check if the TV has offsets
      return OffsetSource.TERM_VECTORS;
    }
  }
  return OffsetSource.ANALYSIS;
}