Example usage for org.apache.lucene.search DocIdSetIterator cost

Introduction

In this page you can find the example usage for org.apache.lucene.search DocIdSetIterator cost.

Prototype

public abstract long cost();

Source Link

Document

Returns the estimated cost of this DocIdSetIterator .

Usage

From source file:lucene.security.search.DocumentVisibilityFilter.java

License:Apache License

public static DocIdSet getLogicalOr(final List<DocIdSet> list) throws IOException {
    if (list.size() == 0) {
        return DocIdSet.EMPTY_DOCIDSET;
    }/*from   ww  w. j av  a2  s.  co m*/
    if (list.size() == 1) {
        DocIdSet docIdSet = list.get(0);
        Bits bits = docIdSet.bits();
        if (bits == null) {
            throw new IOException("Bits are not allowed to be null for DocIdSet [" + docIdSet + "].");
        }
        return docIdSet;
    }
    int index = 0;
    final Bits[] bitsArray = new Bits[list.size()];
    int length = -1;
    for (DocIdSet docIdSet : list) {
        Bits bits = docIdSet.bits();
        if (bits == null) {
            throw new IOException("Bits are not allowed to be null for DocIdSet [" + docIdSet + "].");
        }
        bitsArray[index] = bits;
        index++;
        if (length < 0) {
            length = bits.length();
        } else if (length != bits.length()) {
            throw new IOException(
                    "Bits length need to be the same [" + length + "] and [" + bits.length() + "]");
        }
    }
    final int len = length;
    return new DocIdSet() {

        @Override
        public Bits bits() throws IOException {
            return new Bits() {

                @Override
                public boolean get(int index) {
                    for (int i = 0; i < bitsArray.length; i++) {
                        if (bitsArray[i].get(index)) {
                            return true;
                        }
                    }
                    return false;
                }

                @Override
                public int length() {
                    return len;
                }

            };
        }

        @Override
        public boolean isCacheable() {
            return true;
        }

        @Override
        public DocIdSetIterator iterator() throws IOException {
            final DocIdSetIterator[] docIdSetIteratorArray = new DocIdSetIterator[list.size()];
            long c = 0;
            int index = 0;
            for (DocIdSet docIdSet : list) {
                DocIdSetIterator iterator = docIdSet.iterator();
                iterator.nextDoc();
                docIdSetIteratorArray[index] = iterator;
                c += iterator.cost();
                index++;
            }
            final long cost = c;
            return new DocIdSetIterator() {

                private int _docId = -1;

                @Override
                public int advance(int target) throws IOException {
                    callAdvanceOnAllThatAreBehind(target);
                    Arrays.sort(docIdSetIteratorArray, COMPARATOR);
                    DocIdSetIterator iterator = docIdSetIteratorArray[0];
                    return _docId = iterator.docID();
                }

                private void callAdvanceOnAllThatAreBehind(int target) throws IOException {
                    for (int i = 0; i < docIdSetIteratorArray.length; i++) {
                        DocIdSetIterator iterator = docIdSetIteratorArray[i];
                        if (iterator.docID() < target) {
                            iterator.advance(target);
                        }
                    }
                }

                @Override
                public int nextDoc() throws IOException {
                    return advance(_docId + 1);
                }

                @Override
                public int docID() {
                    return _docId;
                }

                @Override
                public long cost() {
                    return cost;
                }

            };
        }
    };
}

---

From source file:org.apache.solr.search.DocSetBuilder.java

License:Apache License

public void add(DocIdSetIterator iter, int base) throws IOException {
    grow((int) Math.min(Integer.MAX_VALUE, iter.cost()));

    if (bitSet != null) {
        add(bitSet, iter, base);/* w w w .  j  a  v  a 2  s.co m*/
    } else {
        while (true) {
            for (int i = pos; i < buffer.length; ++i) {
                final int doc = iter.nextDoc();
                if (doc == DocIdSetIterator.NO_MORE_DOCS) {
                    pos = i; // update pos
                    return;
                }
                buffer[i] = doc + base; // using the loop counter may help with removal of bounds checking
            }

            pos = buffer.length; // update pos
            if (pos + 1 >= threshold) {
                break;
            }

            growBuffer(pos + 1);
        }

        upgradeToBitSet();
        add(bitSet, iter, base);
    }
}

---

From source file:org.codelibs.elasticsearch.search.profile.query.ProfileScorer.java

License:Apache License

@Override
public DocIdSetIterator iterator() {
    final DocIdSetIterator in = scorer.iterator();
    return new DocIdSetIterator() {

        @Override//  w w w  .  ja v a 2  s . co  m
        public int advance(int target) throws IOException {
            profile.startTime(QueryTimingType.ADVANCE);
            try {
                return in.advance(target);
            } finally {
                profile.stopAndRecordTime();
            }
        }

        @Override
        public int nextDoc() throws IOException {
            profile.startTime(QueryTimingType.NEXT_DOC);
            try {
                return in.nextDoc();
            } finally {
                profile.stopAndRecordTime();
            }
        }

        @Override
        public int docID() {
            return in.docID();
        }

        @Override
        public long cost() {
            return in.cost();
        }
    };
}

---

From source file:org.codelibs.elasticsearch.search.profile.query.ProfileScorer.java

License:Apache License

@Override
public TwoPhaseIterator twoPhaseIterator() {
    final TwoPhaseIterator in = scorer.twoPhaseIterator();
    if (in == null) {
        return null;
    }//  w  w w .jav  a2 s .c om
    final DocIdSetIterator inApproximation = in.approximation();
    final DocIdSetIterator approximation = new DocIdSetIterator() {

        @Override
        public int advance(int target) throws IOException {
            profile.startTime(QueryTimingType.ADVANCE);
            try {
                return inApproximation.advance(target);
            } finally {
                profile.stopAndRecordTime();
            }
        }

        @Override
        public int nextDoc() throws IOException {
            profile.startTime(QueryTimingType.NEXT_DOC);
            try {
                return inApproximation.nextDoc();
            } finally {
                profile.stopAndRecordTime();
            }
        }

        @Override
        public int docID() {
            return inApproximation.docID();
        }

        @Override
        public long cost() {
            return inApproximation.cost();
        }
    };
    return new TwoPhaseIterator(approximation) {
        @Override
        public boolean matches() throws IOException {
            profile.startTime(QueryTimingType.MATCH);
            try {
                return in.matches();
            } finally {
                profile.stopAndRecordTime();
            }
        }

        @Override
        public float matchCost() {
            return in.matchCost();
        }
    };
}

---

From source file:org.elasticsearch.search.aggregations.bucket.composite.SortedDocsProducer.java

License:Apache License

/**
 * Visits all non-deleted documents in <code>iterator</code> and fills the provided <code>queue</code>
 * with the top composite buckets extracted from the collection.
 * Documents that contain a top composite bucket are added in the provided <code>builder</code> if it is not null.
 *
 * Returns true if the queue is full and the current <code>leadSourceBucket</code> did not produce any competitive
 * composite buckets./*from w  w  w  .j  a va2  s. c om*/
 */
protected boolean processBucket(CompositeValuesCollectorQueue queue, LeafReaderContext context,
        DocIdSetIterator iterator, Comparable<?> leadSourceBucket, @Nullable DocIdSetBuilder builder)
        throws IOException {
    final int[] topCompositeCollected = new int[1];
    final boolean[] hasCollected = new boolean[1];
    final LeafBucketCollector queueCollector = new LeafBucketCollector() {
        int lastDoc = -1;

        // we need to add the matching document in the builder
        // so we build a bulk adder from the approximate cost of the iterator
        // and rebuild the adder during the collection if needed
        int remainingBits = (int) Math.min(iterator.cost(), Integer.MAX_VALUE);
        DocIdSetBuilder.BulkAdder adder = builder == null ? null : builder.grow(remainingBits);

        @Override
        public void collect(int doc, long bucket) throws IOException {
            hasCollected[0] = true;
            int slot = queue.addIfCompetitive();
            if (slot != -1) {
                topCompositeCollected[0]++;
                if (adder != null && doc != lastDoc) {
                    if (remainingBits == 0) {
                        // the cost approximation was lower than the real size, we need to grow the adder
                        // by some numbers (128) to ensure that we can add the extra documents
                        adder = builder.grow(128);
                        remainingBits = 128;
                    }
                    adder.add(doc);
                    remainingBits--;
                    lastDoc = doc;
                }
            }
        }
    };
    final Bits liveDocs = context.reader().getLiveDocs();
    final LeafBucketCollector collector = queue.getLeafCollector(leadSourceBucket, context, queueCollector);
    while (iterator.nextDoc() != DocIdSetIterator.NO_MORE_DOCS) {
        if (liveDocs == null || liveDocs.get(iterator.docID())) {
            collector.collect(iterator.docID());
        }
    }
    if (queue.isFull() && hasCollected[0] && topCompositeCollected[0] == 0) {
        return true;
    }
    return false;
}

---

From source file:org.elasticsearch.search.profile.ProfileScorer.java

License:Apache License

@Override
public DocIdSetIterator iterator() {
    final DocIdSetIterator in = scorer.iterator();
    return new DocIdSetIterator() {

        @Override// w w w  .  ja  v  a 2  s .  c  o  m
        public int advance(int target) throws IOException {
            profile.startTime(ProfileBreakdown.TimingType.ADVANCE);
            try {
                return in.advance(target);
            } finally {
                profile.stopAndRecordTime();
            }
        }

        @Override
        public int nextDoc() throws IOException {
            profile.startTime(ProfileBreakdown.TimingType.NEXT_DOC);
            try {
                return in.nextDoc();
            } finally {
                profile.stopAndRecordTime();
            }
        }

        @Override
        public int docID() {
            return in.docID();
        }

        @Override
        public long cost() {
            return in.cost();
        }
    };
}

---

From source file:org.elasticsearch.search.profile.ProfileScorer.java

License:Apache License

@Override
public TwoPhaseIterator twoPhaseIterator() {
    final TwoPhaseIterator in = scorer.twoPhaseIterator();
    if (in == null) {
        return null;
    }//w ww .j  av  a 2s  .  co m
    final DocIdSetIterator inApproximation = in.approximation();
    final DocIdSetIterator approximation = new DocIdSetIterator() {

        @Override
        public int advance(int target) throws IOException {
            profile.startTime(ProfileBreakdown.TimingType.ADVANCE);
            try {
                return inApproximation.advance(target);
            } finally {
                profile.stopAndRecordTime();
            }
        }

        @Override
        public int nextDoc() throws IOException {
            profile.startTime(ProfileBreakdown.TimingType.NEXT_DOC);
            try {
                return inApproximation.nextDoc();
            } finally {
                profile.stopAndRecordTime();
            }
        }

        @Override
        public int docID() {
            return inApproximation.docID();
        }

        @Override
        public long cost() {
            return inApproximation.cost();
        }
    };
    return new TwoPhaseIterator(approximation) {
        @Override
        public boolean matches() throws IOException {
            profile.startTime(ProfileBreakdown.TimingType.MATCH);
            try {
                return in.matches();
            } finally {
                profile.stopAndRecordTime();
            }
        }

        @Override
        public float matchCost() {
            return in.matchCost();
        }
    };
}

---

From source file:org.opengrok.indexer.search.context.OGKUnifiedHighlighter.java

License:Apache License

/**
 * Produces original text by reading from OpenGrok source content relative
 * to {@link RuntimeEnvironment#getSourceRootPath()} and returns the content
 * for each document if the timestamp matches -- or else just {@code null}
 * for a missing file or a timestamp mismatch (as "the returned Strings must
 * be identical to what was indexed.")/* w  ww.j a  v  a 2 s  . c o m*/
 * <p>
 * "This method must load fields for at least one document from the given
 * {@link DocIdSetIterator} but need not return all of them; by default the
 * character lengths are summed and this method will return early when
 * {@code cacheCharsThreshold} is exceeded. Specifically if that number is
 * 0, then only one document is fetched no matter what. Values in the array
 * of {@link CharSequence} will be {@code null} if no value was found."
 * @return a defined instance
 * @throws IOException if an I/O error occurs
 */
@Override
protected List<CharSequence[]> loadFieldValues(String[] fields, DocIdSetIterator docIter,
        int cacheCharsThreshold) throws IOException {

    List<CharSequence[]> docListOfFields = new ArrayList<>(
            cacheCharsThreshold == 0 ? 1 : (int) Math.min(64, docIter.cost()));

    int sumChars = 0;
    do {
        int docId = docIter.nextDoc();
        if (docId == DocIdSetIterator.NO_MORE_DOCS) {
            break;
        }
        Document doc = searcher.doc(docId);

        String path = doc.get(QueryBuilder.PATH);
        String storedU = doc.get(QueryBuilder.U);
        String content = getRepoFileContent(path, storedU);

        CharSequence[] seqs = new CharSequence[fields.length];
        Arrays.fill(seqs, content);
        docListOfFields.add(seqs);

        if (content != null) {
            sumChars += content.length();
        }
    } while (sumChars <= cacheCharsThreshold && cacheCharsThreshold != 0);

    return docListOfFields;
}

---