Package org.apache.lucene.index

Source Code of org.apache.lucene.index.MultiDocValues$MultiSortedDocValues

package org.apache.lucene.index;

/*
* 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.
*/

import java.io.IOException;
import java.util.List;

import org.apache.lucene.index.MultiTermsEnum.TermsEnumIndex;
import org.apache.lucene.index.MultiTermsEnum.TermsEnumWithSlice;
import org.apache.lucene.util.Bits;
import org.apache.lucene.util.BytesRef;
import org.apache.lucene.util.packed.AppendingPackedLongBuffer;
import org.apache.lucene.util.packed.MonotonicAppendingLongBuffer;
import org.apache.lucene.util.packed.PackedInts;

/**
* A wrapper for CompositeIndexReader providing access to DocValues.
*
* <p><b>NOTE</b>: for multi readers, you'll get better
* performance by gathering the sub readers using
* {@link IndexReader#getContext()} to get the
* atomic leaves and then operate per-AtomicReader,
* instead of using this class.
*
* <p><b>NOTE</b>: This is very costly.
*
* @lucene.experimental
* @lucene.internal
*/
public class MultiDocValues {
 
  /** No instantiation */
  private MultiDocValues() {}
 
  /** Returns a NumericDocValues for a reader's norms (potentially merging on-the-fly).
   * <p>
   * This is a slow way to access normalization values. Instead, access them per-segment
   * with {@link AtomicReader#getNormValues(String)}
   * </p>
   */
  public static NumericDocValues getNormValues(final IndexReader r, final String field) throws IOException {
    final List<AtomicReaderContext> leaves = r.leaves();
    final int size = leaves.size();
    if (size == 0) {
      return null;
    } else if (size == 1) {
      return leaves.get(0).reader().getNormValues(field);
    }
    FieldInfo fi = MultiFields.getMergedFieldInfos(r).fieldInfo(field);
    if (fi == null || fi.hasNorms() == false) {
      return null;
    }

    boolean anyReal = false;
    final NumericDocValues[] values = new NumericDocValues[size];
    final int[] starts = new int[size+1];
    for (int i = 0; i < size; i++) {
      AtomicReaderContext context = leaves.get(i);
      NumericDocValues v = context.reader().getNormValues(field);
      if (v == null) {
        v = NumericDocValues.EMPTY;
      } else {
        anyReal = true;
      }
      values[i] = v;
      starts[i] = context.docBase;
    }
    starts[size] = r.maxDoc();
   
    assert anyReal;

    return new NumericDocValues() {
      @Override
      public long get(int docID) {
        int subIndex = ReaderUtil.subIndex(docID, starts);
        return values[subIndex].get(docID - starts[subIndex]);
      }
    };
  }

  /** Returns a NumericDocValues for a reader's docvalues (potentially merging on-the-fly)
   * <p>
   * This is a slow way to access numeric values. Instead, access them per-segment
   * with {@link AtomicReader#getNumericDocValues(String)}
   * </p>
   * */
  public static NumericDocValues getNumericValues(final IndexReader r, final String field) throws IOException {
    final List<AtomicReaderContext> leaves = r.leaves();
    final int size = leaves.size();
    if (size == 0) {
      return null;
    } else if (size == 1) {
      return leaves.get(0).reader().getNumericDocValues(field);
    }

    boolean anyReal = false;
    final NumericDocValues[] values = new NumericDocValues[size];
    final int[] starts = new int[size+1];
    for (int i = 0; i < size; i++) {
      AtomicReaderContext context = leaves.get(i);
      NumericDocValues v = context.reader().getNumericDocValues(field);
      if (v == null) {
        v = NumericDocValues.EMPTY;
      } else {
        anyReal = true;
      }
      values[i] = v;
      starts[i] = context.docBase;
    }
    starts[size] = r.maxDoc();

    if (!anyReal) {
      return null;
    } else {
      return new NumericDocValues() {
        @Override
        public long get(int docID) {
          int subIndex = ReaderUtil.subIndex(docID, starts);
          return values[subIndex].get(docID - starts[subIndex]);
        }
      };
    }
  }
 
  /** Returns a Bits for a reader's docsWithField (potentially merging on-the-fly)
   * <p>
   * This is a slow way to access this bitset. Instead, access them per-segment
   * with {@link AtomicReader#getDocsWithField(String)}
   * </p>
   * */
  public static Bits getDocsWithField(final IndexReader r, final String field) throws IOException {
    final List<AtomicReaderContext> leaves = r.leaves();
    final int size = leaves.size();
    if (size == 0) {
      return null;
    } else if (size == 1) {
      return leaves.get(0).reader().getDocsWithField(field);
    }

    boolean anyReal = false;
    boolean anyMissing = false;
    final Bits[] values = new Bits[size];
    final int[] starts = new int[size+1];
    for (int i = 0; i < size; i++) {
      AtomicReaderContext context = leaves.get(i);
      Bits v = context.reader().getDocsWithField(field);
      if (v == null) {
        v = new Bits.MatchNoBits(context.reader().maxDoc());
        anyMissing = true;
      } else {
        anyReal = true;
        if (v instanceof Bits.MatchAllBits == false) {
          anyMissing = true;
        }
      }
      values[i] = v;
      starts[i] = context.docBase;
    }
    starts[size] = r.maxDoc();

    if (!anyReal) {
      return null;
    } else if (!anyMissing) {
      return new Bits.MatchAllBits(r.maxDoc());
    } else {
      return new MultiBits(values, starts, false);
    }
  }

  /** Returns a BinaryDocValues for a reader's docvalues (potentially merging on-the-fly)
   * <p>
   * This is a slow way to access binary values. Instead, access them per-segment
   * with {@link AtomicReader#getBinaryDocValues(String)}
   * </p> 
   */
  public static BinaryDocValues getBinaryValues(final IndexReader r, final String field) throws IOException {
    final List<AtomicReaderContext> leaves = r.leaves();
    final int size = leaves.size();
   
    if (size == 0) {
      return null;
    } else if (size == 1) {
      return leaves.get(0).reader().getBinaryDocValues(field);
    }
   
    boolean anyReal = false;
    final BinaryDocValues[] values = new BinaryDocValues[size];
    final int[] starts = new int[size+1];
    for (int i = 0; i < size; i++) {
      AtomicReaderContext context = leaves.get(i);
      BinaryDocValues v = context.reader().getBinaryDocValues(field);
      if (v == null) {
        v = BinaryDocValues.EMPTY;
      } else {
        anyReal = true;
      }
      values[i] = v;
      starts[i] = context.docBase;
    }
    starts[size] = r.maxDoc();
   
    if (!anyReal) {
      return null;
    } else {
      return new BinaryDocValues() {
        @Override
        public void get(int docID, BytesRef result) {
          int subIndex = ReaderUtil.subIndex(docID, starts);
          values[subIndex].get(docID - starts[subIndex], result);
        }
      };
    }
  }
 
  /** Returns a SortedDocValues for a reader's docvalues (potentially doing extremely slow things).
   * <p>
   * This is an extremely slow way to access sorted values. Instead, access them per-segment
   * with {@link AtomicReader#getSortedDocValues(String)}
   * </p> 
   */
  public static SortedDocValues getSortedValues(final IndexReader r, final String field) throws IOException {
    final List<AtomicReaderContext> leaves = r.leaves();
    final int size = leaves.size();
   
    if (size == 0) {
      return null;
    } else if (size == 1) {
      return leaves.get(0).reader().getSortedDocValues(field);
    }
   
    boolean anyReal = false;
    final SortedDocValues[] values = new SortedDocValues[size];
    final int[] starts = new int[size+1];
    for (int i = 0; i < size; i++) {
      AtomicReaderContext context = leaves.get(i);
      SortedDocValues v = context.reader().getSortedDocValues(field);
      if (v == null) {
        v = SortedDocValues.EMPTY;
      } else {
        anyReal = true;
      }
      values[i] = v;
      starts[i] = context.docBase;
    }
    starts[size] = r.maxDoc();
   
    if (!anyReal) {
      return null;
    } else {
      TermsEnum enums[] = new TermsEnum[values.length];
      for (int i = 0; i < values.length; i++) {
        enums[i] = values[i].termsEnum();
      }
      OrdinalMap mapping = new OrdinalMap(r.getCoreCacheKey(), enums);
      return new MultiSortedDocValues(values, starts, mapping);
    }
  }
 
  /** Returns a SortedSetDocValues for a reader's docvalues (potentially doing extremely slow things).
   * <p>
   * This is an extremely slow way to access sorted values. Instead, access them per-segment
   * with {@link AtomicReader#getSortedSetDocValues(String)}
   * </p> 
   */
  public static SortedSetDocValues getSortedSetValues(final IndexReader r, final String field) throws IOException {
    final List<AtomicReaderContext> leaves = r.leaves();
    final int size = leaves.size();
   
    if (size == 0) {
      return null;
    } else if (size == 1) {
      return leaves.get(0).reader().getSortedSetDocValues(field);
    }
   
    boolean anyReal = false;
    final SortedSetDocValues[] values = new SortedSetDocValues[size];
    final int[] starts = new int[size+1];
    for (int i = 0; i < size; i++) {
      AtomicReaderContext context = leaves.get(i);
      SortedSetDocValues v = context.reader().getSortedSetDocValues(field);
      if (v == null) {
        v = SortedSetDocValues.EMPTY;
      } else {
        anyReal = true;
      }
      values[i] = v;
      starts[i] = context.docBase;
    }
    starts[size] = r.maxDoc();
   
    if (!anyReal) {
      return null;
    } else {
      TermsEnum enums[] = new TermsEnum[values.length];
      for (int i = 0; i < values.length; i++) {
        enums[i] = values[i].termsEnum();
      }
      OrdinalMap mapping = new OrdinalMap(r.getCoreCacheKey(), enums);
      return new MultiSortedSetDocValues(values, starts, mapping);
    }
  }
 
  /** maps per-segment ordinals to/from global ordinal space */
  // TODO: use more efficient packed ints structures?
  // TODO: pull this out? its pretty generic (maps between N ord()-enabled TermsEnums)
  public static class OrdinalMap {
    // cache key of whoever asked for this awful thing
    final Object owner;
    // globalOrd -> (globalOrd - segmentOrd) where segmentOrd is the the ordinal in the first segment that contains this term
    final MonotonicAppendingLongBuffer globalOrdDeltas;
    // globalOrd -> first segment container
    final AppendingPackedLongBuffer firstSegments;
    // for every segment, segmentOrd -> (globalOrd - segmentOrd)
    final MonotonicAppendingLongBuffer ordDeltas[];
   
    /**
     * Creates an ordinal map that allows mapping ords to/from a merged
     * space from <code>subs</code>.
     * @param owner a cache key
     * @param subs TermsEnums that support {@link TermsEnum#ord()}. They need
     *             not be dense (e.g. can be FilteredTermsEnums}.
     * @throws IOException if an I/O error occurred.
     */
    public OrdinalMap(Object owner, TermsEnum subs[]) throws IOException {
      // create the ordinal mappings by pulling a termsenum over each sub's
      // unique terms, and walking a multitermsenum over those
      this.owner = owner;
      globalOrdDeltas = new MonotonicAppendingLongBuffer(PackedInts.COMPACT);
      firstSegments = new AppendingPackedLongBuffer(PackedInts.COMPACT);
      ordDeltas = new MonotonicAppendingLongBuffer[subs.length];
      for (int i = 0; i < ordDeltas.length; i++) {
        ordDeltas[i] = new MonotonicAppendingLongBuffer();
      }
      long segmentOrds[] = new long[subs.length];
      ReaderSlice slices[] = new ReaderSlice[subs.length];
      TermsEnumIndex indexes[] = new TermsEnumIndex[slices.length];
      for (int i = 0; i < slices.length; i++) {
        slices[i] = new ReaderSlice(0, 0, i);
        indexes[i] = new TermsEnumIndex(subs[i], i);
      }
      MultiTermsEnum mte = new MultiTermsEnum(slices);
      mte.reset(indexes);
      long globalOrd = 0;
      while (mte.next() != null) {       
        TermsEnumWithSlice matches[] = mte.getMatchArray();
        for (int i = 0; i < mte.getMatchCount(); i++) {
          int segmentIndex = matches[i].index;
          long segmentOrd = matches[i].terms.ord();
          long delta = globalOrd - segmentOrd;
          // for each unique term, just mark the first segment index/delta where it occurs
          if (i == 0) {
            firstSegments.add(segmentIndex);
            globalOrdDeltas.add(delta);
          }
          // for each per-segment ord, map it back to the global term.
          while (segmentOrds[segmentIndex] <= segmentOrd) {
            ordDeltas[segmentIndex].add(delta);
            segmentOrds[segmentIndex]++;
          }
        }
        globalOrd++;
      }
      firstSegments.freeze();
      globalOrdDeltas.freeze();
      for (int i = 0; i < ordDeltas.length; ++i) {
        ordDeltas[i].freeze();
      }
    }
   
    /**
     * Given a segment number and segment ordinal, returns
     * the corresponding global ordinal.
     */
    public long getGlobalOrd(int segmentIndex, long segmentOrd) {
      return segmentOrd + ordDeltas[segmentIndex].get(segmentOrd);
    }

    /**
     * Given global ordinal, returns the ordinal of the first segment which contains
     * this ordinal (the corresponding to the segment return {@link #getFirstSegmentNumber}).
     */
    public long getFirstSegmentOrd(long globalOrd) {
      return globalOrd - globalOrdDeltas.get(globalOrd);
    }
   
    /**
     * Given a global ordinal, returns the index of the first
     * segment that contains this term.
     */
    public int getFirstSegmentNumber(long globalOrd) {
      return (int) firstSegments.get(globalOrd);
    }
   
    /**
     * Returns the total number of unique terms in global ord space.
     */
    public long getValueCount() {
      return globalOrdDeltas.size();
    }
   
    /**
     * Returns total byte size used by this ordinal map.
     */
    public long ramBytesUsed() {
      long size = globalOrdDeltas.ramBytesUsed() + firstSegments.ramBytesUsed();
      for (int i = 0; i < ordDeltas.length; i++) {
        size += ordDeltas[i].ramBytesUsed();
      }
      return size;
    }
  }
 
  /**
   * Implements SortedDocValues over n subs, using an OrdinalMap
   * @lucene.internal
   */
  public static class MultiSortedDocValues extends SortedDocValues {
    /** docbase for each leaf: parallel with {@link #values} */
    public final int docStarts[];
    /** leaf values */
    public final SortedDocValues values[];
    /** ordinal map mapping ords from <code>values</code> to global ord space */
    public final OrdinalMap mapping;
 
    /** Creates a new MultiSortedDocValues over <code>values</code> */
    MultiSortedDocValues(SortedDocValues values[], int docStarts[], OrdinalMap mapping) throws IOException {
      assert values.length == mapping.ordDeltas.length;
      assert docStarts.length == values.length + 1;
      this.values = values;
      this.docStarts = docStarts;
      this.mapping = mapping;
    }
      
    @Override
    public int getOrd(int docID) {
      int subIndex = ReaderUtil.subIndex(docID, docStarts);
      int segmentOrd = values[subIndex].getOrd(docID - docStarts[subIndex]);
      return segmentOrd == -1 ? segmentOrd : (int) mapping.getGlobalOrd(subIndex, segmentOrd);
    }
    @Override
    public void lookupOrd(int ord, BytesRef result) {
      int subIndex = mapping.getFirstSegmentNumber(ord);
      int segmentOrd = (int) mapping.getFirstSegmentOrd(ord);
      values[subIndex].lookupOrd(segmentOrd, result);
    }
    @Override
    public int getValueCount() {
      return (int) mapping.getValueCount();
    }
  }
 
  /**
   * Implements MultiSortedSetDocValues over n subs, using an OrdinalMap
   * @lucene.internal
   */
  public static class MultiSortedSetDocValues extends SortedSetDocValues {
    /** docbase for each leaf: parallel with {@link #values} */
    public final int docStarts[];
    /** leaf values */
    public final SortedSetDocValues values[];
    /** ordinal map mapping ords from <code>values</code> to global ord space */
    public final OrdinalMap mapping;
    int currentSubIndex;
   
    /** Creates a new MultiSortedSetDocValues over <code>values</code> */
    MultiSortedSetDocValues(SortedSetDocValues values[], int docStarts[], OrdinalMap mapping) throws IOException {
      assert values.length == mapping.ordDeltas.length;
      assert docStarts.length == values.length + 1;
      this.values = values;
      this.docStarts = docStarts;
      this.mapping = mapping;
    }
   
    @Override
    public long nextOrd() {
      long segmentOrd = values[currentSubIndex].nextOrd();
      if (segmentOrd == NO_MORE_ORDS) {
        return segmentOrd;
      } else {
        return mapping.getGlobalOrd(currentSubIndex, segmentOrd);
      }
    }

    @Override
    public void setDocument(int docID) {
      currentSubIndex = ReaderUtil.subIndex(docID, docStarts);
      values[currentSubIndex].setDocument(docID - docStarts[currentSubIndex]);
    }
    @Override
    public void lookupOrd(long ord, BytesRef result) {
      int subIndex = mapping.getFirstSegmentNumber(ord);
      long segmentOrd = mapping.getFirstSegmentOrd(ord);
      values[subIndex].lookupOrd(segmentOrd, result);
    }
    @Override
    public long getValueCount() {
      return mapping.getValueCount();
    }
  }
}
TOP

Related Classes of org.apache.lucene.index.MultiDocValues$MultiSortedDocValues

TOP
Copyright © 2018 www.massapi.com. All rights reserved.
All source code are property of their respective owners. Java is a trademark of Sun Microsystems, Inc and owned by ORACLE Inc. Contact coftware#gmail.com.