org.neo4j.kernel.api.index.IndexAccessor Maven / Gradle / Ivy
/*
* Copyright (c) 2018-2020 "Graph Foundation,"
* Graph Foundation, Inc. [https://graphfoundation.org]
*
* This file is part of ONgDB.
*
* ONgDB is free software: you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation, either version 3 of the License, or
* (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program. If not, see .
*/
/*
* Copyright (c) 2002-2020 "Neo4j,"
* Neo4j Sweden AB [http://neo4j.com]
*
* This file is part of Neo4j.
*
* Neo4j is free software: you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation, either version 3 of the License, or
* (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program. If not, see .
*/
package org.neo4j.kernel.api.index;
import java.io.Closeable;
import java.io.File;
import java.io.IOException;
import java.util.Iterator;
import org.neo4j.graphdb.ResourceIterator;
import org.neo4j.helpers.collection.BoundedIterable;
import org.neo4j.io.pagecache.IOLimiter;
import org.neo4j.kernel.api.exceptions.index.IndexEntryConflictException;
import org.neo4j.kernel.impl.api.index.IndexUpdateMode;
import org.neo4j.kernel.impl.api.index.updater.SwallowingIndexUpdater;
import org.neo4j.kernel.impl.util.Validator;
import org.neo4j.storageengine.api.schema.IndexReader;
import org.neo4j.values.storable.Value;
import static java.util.Collections.emptyIterator;
import static org.neo4j.helpers.collection.Iterators.emptyResourceIterator;
/**
* Used for online operation of an index.
*/
public interface IndexAccessor extends Closeable
{
IndexAccessor EMPTY = new Adapter();
/**
* Deletes this index as well as closes all used external resources.
* There will not be any interactions after this call.
*
* @throws IOException if unable to drop index.
*/
void drop() throws IOException;
/**
* Return an updater for applying a set of changes to this index.
* Updates must be visible in {@link #newReader() readers} created after this update.
*
* This is called with IndexUpdateMode.RECOVERY when starting up after
* a crash or similar. Updates given then may have already been applied to this index, so
* additional checks must be in place so that data doesn't get duplicated, but is idempotent.
*/
IndexUpdater newUpdater( IndexUpdateMode mode );
/**
* Forces this index to disk. Called at certain points from within ONgDB for example when
* rotating the logical log. After completion of this call there cannot be any essential state that
* hasn't been forced to disk.
*
* @param ioLimiter The {@link IOLimiter} to use for implementations living on top of {@link org.neo4j.io.pagecache.PageCache}.
* @throws IOException if there was a problem forcing the state to persistent storage.
*/
void force( IOLimiter ioLimiter ) throws IOException;
/**
* Refreshes this index, so that {@link #newReader() readers} created after completion of this call
* will see the latest updates. This happens automatically on closing {@link #newUpdater(IndexUpdateMode)}
* w/ {@link IndexUpdateMode#ONLINE}, but not guaranteed for {@link IndexUpdateMode#RECOVERY}.
* Therefore this call is complementary for updates that has taken place with {@link IndexUpdateMode#RECOVERY}.
*
* @throws IOException if there was a problem refreshing the index.
*/
void refresh() throws IOException;
/**
* Closes this index accessor. There will not be any interactions after this call.
* After completion of this call there cannot be any essential state that hasn't been forced to disk.
*
* @throws IOException if unable to close index.
*/
@Override
void close() throws IOException;
/**
* @return a new {@link IndexReader} responsible for looking up results in the index. The returned
* reader must honor repeatable reads.
*/
IndexReader newReader();
BoundedIterable newAllEntriesReader();
/**
* Should return a full listing of all files needed by this index accessor to work with the index. The files
* need to remain available until the resource iterator returned here is closed. This is used to duplicate created
* indexes across clusters, among other things.
*/
ResourceIterator snapshotFiles() throws IOException;
/**
* Verifies that each value in this index is unique.
* Index is guaranteed to not change while this call executes.
*
* @param propertyAccessor {@link PropertyAccessor} for accessing properties from database storage
* in the event of conflicting values.
* @throws IndexEntryConflictException for first detected uniqueness conflict, if any.
* @throws IOException on error reading from source files.
*/
void verifyDeferredConstraints( PropertyAccessor propertyAccessor ) throws IndexEntryConflictException, IOException;
/**
* @return true if index was not shutdown properly and its internal state is dirty, false otherwise
*/
boolean isDirty();
/**
* Validates the {@link Value value tuple} before transaction determines that it can commit.
*/
default void validateBeforeCommit( Value[] tuple )
{
// For most value types there are no specific validations to be made.
}
class Adapter implements IndexAccessor
{
@Override
public void drop()
{
}
@Override
public IndexUpdater newUpdater( IndexUpdateMode mode )
{
return SwallowingIndexUpdater.INSTANCE;
}
@Override
public void force( IOLimiter ioLimiter )
{
}
@Override
public void refresh()
{
}
@Override
public void close()
{
}
@Override
public IndexReader newReader()
{
return IndexReader.EMPTY;
}
@Override
public BoundedIterable newAllEntriesReader()
{
return new BoundedIterable()
{
@Override
public long maxCount()
{
return 0;
}
@Override
public void close()
{
}
@Override
public Iterator iterator()
{
return emptyIterator();
}
};
}
@Override
public ResourceIterator snapshotFiles()
{
return emptyResourceIterator();
}
@Override
public void verifyDeferredConstraints( PropertyAccessor propertyAccessor )
{
}
@Override
public boolean isDirty()
{
return false;
}
}
class Delegator implements IndexAccessor
{
private final IndexAccessor delegate;
public Delegator( IndexAccessor delegate )
{
this.delegate = delegate;
}
@Override
public void drop() throws IOException
{
delegate.drop();
}
@Override
public IndexUpdater newUpdater( IndexUpdateMode mode )
{
return delegate.newUpdater( mode );
}
@Override
public void force( IOLimiter ioLimiter ) throws IOException
{
delegate.force( ioLimiter );
}
@Override
public void refresh() throws IOException
{
delegate.refresh();
}
@Override
public void close() throws IOException
{
delegate.close();
}
@Override
public IndexReader newReader()
{
return delegate.newReader();
}
@Override
public BoundedIterable newAllEntriesReader()
{
return delegate.newAllEntriesReader();
}
@Override
public ResourceIterator snapshotFiles() throws IOException
{
return delegate.snapshotFiles();
}
@Override
public String toString()
{
return delegate.toString();
}
@Override
public void verifyDeferredConstraints( PropertyAccessor propertyAccessor )
throws IndexEntryConflictException, IOException
{
delegate.verifyDeferredConstraints( propertyAccessor );
}
@Override
public boolean isDirty()
{
return delegate.isDirty();
}
@Override
public void validateBeforeCommit( Value[] tuple )
{
delegate.validateBeforeCommit( tuple );
}
}
}