PackIndexV2.java

/*
 * Copyright (C) 2008, Shawn O. Pearce <spearce@spearce.org>
 * and other copyright owners as documented in the project's IP log.
 *
 * This program and the accompanying materials are made available
 * under the terms of the Eclipse Distribution License v1.0 which
 * accompanies this distribution, is reproduced below, and is
 * available at http://www.eclipse.org/org/documents/edl-v10.php
 *
 * All rights reserved.
 *
 * Redistribution and use in source and binary forms, with or
 * without modification, are permitted provided that the following
 * conditions are met:
 *
 * - Redistributions of source code must retain the above copyright
 *   notice, this list of conditions and the following disclaimer.
 *
 * - Redistributions in binary form must reproduce the above
 *   copyright notice, this list of conditions and the following
 *   disclaimer in the documentation and/or other materials provided
 *   with the distribution.
 *
 * - Neither the name of the Eclipse Foundation, Inc. nor the
 *   names of its contributors may be used to endorse or promote
 *   products derived from this software without specific prior
 *   written permission.
 *
 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND
 * CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES,
 * INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
 * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
 * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR
 * CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
 * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
 * NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
 * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
 * CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT,
 * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF
 * ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
 */

package org.eclipse.jgit.internal.storage.file;

import java.io.IOException;
import java.io.InputStream;
import java.text.MessageFormat;
import java.util.Arrays;
import java.util.Iterator;
import java.util.NoSuchElementException;
import java.util.Set;

import org.eclipse.jgit.errors.MissingObjectException;
import org.eclipse.jgit.internal.JGitText;
import org.eclipse.jgit.lib.AbbreviatedObjectId;
import org.eclipse.jgit.lib.AnyObjectId;
import org.eclipse.jgit.lib.Constants;
import org.eclipse.jgit.lib.ObjectId;
import org.eclipse.jgit.util.IO;
import org.eclipse.jgit.util.NB;

/** Support for the pack index v2 format. */
class PackIndexV2 extends PackIndex {
    private static final long IS_O64 = 1L << 31;

    private static final int FANOUT = 256;

    private static final int[] NO_INTS = {};

    private static final byte[] NO_BYTES = {};

    private long objectCnt;

    private final long[] fanoutTable;

    /** 256 arrays of contiguous object names. */
    int[][] names;

    /** 256 arrays of the 32 bit offset data, matching {@link #names}. */
    byte[][] offset32;

    /** 256 arrays of the CRC-32 of objects, matching {@link #names}. */
    private byte[][] crc32;

    /** 64 bit offset table. */
    byte[] offset64;

    PackIndexV2(final InputStream fd) throws IOException {
        final byte[] fanoutRaw = new byte[4 * FANOUT];
        IO.readFully(fd, fanoutRaw, 0, fanoutRaw.length);
        fanoutTable = new long[FANOUT];
        for (int k = 0; k < FANOUT; k++)
            fanoutTable[k] = NB.decodeUInt32(fanoutRaw, k * 4);
        objectCnt = fanoutTable[FANOUT - 1];

        names = new int[FANOUT][];
        offset32 = new byte[FANOUT][];
        crc32 = new byte[FANOUT][];

        // Object name table. The size we can permit per fan-out bucket
        // is limited to Java's 2 GB per byte array limitation. That is
        // no more than 107,374,182 objects per fan-out.
        //
        for (int k = 0; k < FANOUT; k++) {
            final long bucketCnt;
            if (k == 0)
                bucketCnt = fanoutTable[k];
            else
                bucketCnt = fanoutTable[k] - fanoutTable[k - 1];

            if (bucketCnt == 0) {
                names[k] = NO_INTS;
                offset32[k] = NO_BYTES;
                crc32[k] = NO_BYTES;
                continue;
            } else if (bucketCnt < 0)
                throw new IOException(MessageFormat.format(
                        JGitText.get().indexFileCorruptedNegativeBucketCount,
                        Long.valueOf(bucketCnt)));

            final long nameLen = bucketCnt * Constants.OBJECT_ID_LENGTH;
            if (nameLen > Integer.MAX_VALUE - 8) // see http://stackoverflow.com/a/8381338
                throw new IOException(JGitText.get().indexFileIsTooLargeForJgit);

            final int intNameLen = (int) nameLen;
            final byte[] raw = new byte[intNameLen];
            final int[] bin = new int[intNameLen >>> 2];
            IO.readFully(fd, raw, 0, raw.length);
            for (int i = 0; i < bin.length; i++)
                bin[i] = NB.decodeInt32(raw, i << 2);

            names[k] = bin;
            offset32[k] = new byte[(int) (bucketCnt * 4)];
            crc32[k] = new byte[(int) (bucketCnt * 4)];
        }

        // CRC32 table.
        for (int k = 0; k < FANOUT; k++)
            IO.readFully(fd, crc32[k], 0, crc32[k].length);

        // 32 bit offset table. Any entries with the most significant bit
        // set require a 64 bit offset entry in another table.
        //
        int o64cnt = 0;
        for (int k = 0; k < FANOUT; k++) {
            final byte[] ofs = offset32[k];
            IO.readFully(fd, ofs, 0, ofs.length);
            for (int p = 0; p < ofs.length; p += 4)
                if (ofs[p] < 0)
                    o64cnt++;
        }

        // 64 bit offset table. Most objects should not require an entry.
        //
        if (o64cnt > 0) {
            offset64 = new byte[o64cnt * 8];
            IO.readFully(fd, offset64, 0, offset64.length);
        } else {
            offset64 = NO_BYTES;
        }

        packChecksum = new byte[20];
        IO.readFully(fd, packChecksum, 0, packChecksum.length);
    }

    /** {@inheritDoc} */
    @Override
    public long getObjectCount() {
        return objectCnt;
    }

    /** {@inheritDoc} */
    @Override
    public long getOffset64Count() {
        return offset64.length / 8;
    }

    private int findLevelOne(long nthPosition) {
        int levelOne = Arrays.binarySearch(fanoutTable, nthPosition + 1);
        if (levelOne >= 0) {
            // If we hit the bucket exactly the item is in the bucket, or
            // any bucket before it which has the same object count.
            //
            long base = fanoutTable[levelOne];
            while (levelOne > 0 && base == fanoutTable[levelOne - 1])
                levelOne--;
        } else {
            // The item is in the bucket we would insert it into.
            //
            levelOne = -(levelOne + 1);
        }
        return levelOne;
    }

    private int getLevelTwo(long nthPosition, int levelOne) {
        final long base = levelOne > 0 ? fanoutTable[levelOne - 1] : 0;
        return (int) (nthPosition - base);
    }

    /** {@inheritDoc} */
    @Override
    public ObjectId getObjectId(long nthPosition) {
        final int levelOne = findLevelOne(nthPosition);
        final int p = getLevelTwo(nthPosition, levelOne);
        final int p4 = p << 2;
        return ObjectId.fromRaw(names[levelOne], p4 + p); // p * 5
    }

    /** {@inheritDoc} */
    @Override
    public long getOffset(long nthPosition) {
        final int levelOne = findLevelOne(nthPosition);
        final int levelTwo = getLevelTwo(nthPosition, levelOne);
        return getOffset(levelOne, levelTwo);
    }

    /** {@inheritDoc} */
    @Override
    public long findOffset(AnyObjectId objId) {
        final int levelOne = objId.getFirstByte();
        final int levelTwo = binarySearchLevelTwo(objId, levelOne);
        if (levelTwo == -1)
            return -1;
        return getOffset(levelOne, levelTwo);
    }

    private long getOffset(int levelOne, int levelTwo) {
        final long p = NB.decodeUInt32(offset32[levelOne], levelTwo << 2);
        if ((p & IS_O64) != 0)
            return NB.decodeUInt64(offset64, (8 * (int) (p & ~IS_O64)));
        return p;
    }

    /** {@inheritDoc} */
    @Override
    public long findCRC32(AnyObjectId objId) throws MissingObjectException {
        final int levelOne = objId.getFirstByte();
        final int levelTwo = binarySearchLevelTwo(objId, levelOne);
        if (levelTwo == -1)
            throw new MissingObjectException(objId.copy(), "unknown"); //$NON-NLS-1$
        return NB.decodeUInt32(crc32[levelOne], levelTwo << 2);
    }

    /** {@inheritDoc} */
    @Override
    public boolean hasCRC32Support() {
        return true;
    }

    /** {@inheritDoc} */
    @Override
    public Iterator<MutableEntry> iterator() {
        return new EntriesIteratorV2();
    }

    /** {@inheritDoc} */
    @Override
    public void resolve(Set<ObjectId> matches, AbbreviatedObjectId id,
            int matchLimit) throws IOException {
        int[] data = names[id.getFirstByte()];
        int max = offset32[id.getFirstByte()].length >>> 2;
        int high = max;
        if (high == 0)
            return;
        int low = 0;
        do {
            int p = (low + high) >>> 1;
            final int cmp = id.prefixCompare(data, idOffset(p));
            if (cmp < 0)
                high = p;
            else if (cmp == 0) {
                // We may have landed in the middle of the matches.  Move
                // backwards to the start of matches, then walk forwards.
                //
                while (0 < p && id.prefixCompare(data, idOffset(p - 1)) == 0)
                    p--;
                for (; p < max && id.prefixCompare(data, idOffset(p)) == 0; p++) {
                    matches.add(ObjectId.fromRaw(data, idOffset(p)));
                    if (matches.size() > matchLimit)
                        break;
                }
                return;
            } else
                low = p + 1;
        } while (low < high);
    }

    private static int idOffset(int p) {
        return (p << 2) + p; // p * 5
    }

    private int binarySearchLevelTwo(AnyObjectId objId, int levelOne) {
        final int[] data = names[levelOne];
        int high = offset32[levelOne].length >>> 2;
        if (high == 0)
            return -1;
        int low = 0;
        do {
            final int mid = (low + high) >>> 1;
            final int mid4 = mid << 2;
            final int cmp;

            cmp = objId.compareTo(data, mid4 + mid); // mid * 5
            if (cmp < 0)
                high = mid;
            else if (cmp == 0) {
                return mid;
            } else
                low = mid + 1;
        } while (low < high);
        return -1;
    }

    private class EntriesIteratorV2 extends EntriesIterator {
        int levelOne;

        int levelTwo;

        @Override
        protected MutableEntry initEntry() {
            return new MutableEntry() {
                @Override
                protected void ensureId() {
                    idBuffer.fromRaw(names[levelOne], levelTwo
                            - Constants.OBJECT_ID_LENGTH / 4);
                }
            };
        }

        @Override
        public MutableEntry next() {
            for (; levelOne < names.length; levelOne++) {
                if (levelTwo < names[levelOne].length) {
                    int idx = levelTwo / (Constants.OBJECT_ID_LENGTH / 4) * 4;
                    long offset = NB.decodeUInt32(offset32[levelOne], idx);
                    if ((offset & IS_O64) != 0) {
                        idx = (8 * (int) (offset & ~IS_O64));
                        offset = NB.decodeUInt64(offset64, idx);
                    }
                    entry.offset = offset;

                    levelTwo += Constants.OBJECT_ID_LENGTH / 4;
                    returnedNumber++;
                    return entry;
                }
                levelTwo = 0;
            }
            throw new NoSuchElementException();
        }
    }

}