org.apache.cassandra.db.SuperColumns Maven / Gradle / Ivy
Go to download
Show more of this group Show more artifacts with this name
Show all versions of cassandra-all Show documentation
Show all versions of cassandra-all Show documentation
Palantir open source project
/*
* 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.
*/
package org.apache.cassandra.db;
import java.io.DataInput;
import java.io.IOError;
import java.io.IOException;
import java.nio.ByteBuffer;
import java.util.ArrayList;
import java.util.Iterator;
import java.util.List;
import java.util.SortedSet;
import java.util.TreeSet;
import org.apache.cassandra.config.CFMetaData;
import org.apache.cassandra.db.composites.*;
import org.apache.cassandra.db.filter.*;
import org.apache.cassandra.db.marshal.AbstractType;
import org.apache.cassandra.utils.ByteBufferUtil;
public class SuperColumns
{
public static Iterator onDiskIterator(DataInput in, int superColumnCount, ColumnSerializer.Flag flag, int expireBefore, CellNameType type)
{
return new SCIterator(in, superColumnCount, flag, expireBefore, type);
}
public static void deserializerSuperColumnFamily(DataInput in, ColumnFamily cf, ColumnSerializer.Flag flag, int version) throws IOException
{
// Note that there was no way to insert a range tombstone in a SCF in 1.2
cf.delete(cf.getComparator().deletionInfoSerializer().deserialize(in, version));
assert !cf.deletionInfo().rangeIterator().hasNext();
Iterator iter = onDiskIterator(in, in.readInt(), flag, Integer.MIN_VALUE, cf.getComparator());
while (iter.hasNext())
cf.addAtom(iter.next());
}
private static class SCIterator implements Iterator
{
private final DataInput in;
private final int scCount;
private final ColumnSerializer.Flag flag;
private final int expireBefore;
private final CellNameType type;
private int read;
private ByteBuffer scName;
private Iterator subColumnsIterator;
private SCIterator(DataInput in, int superColumnCount, ColumnSerializer.Flag flag, int expireBefore, CellNameType type)
{
this.in = in;
this.scCount = superColumnCount;
this.flag = flag;
this.expireBefore = expireBefore;
this.type = type;
}
public boolean hasNext()
{
return (subColumnsIterator != null && subColumnsIterator.hasNext()) || read < scCount;
}
public OnDiskAtom next()
{
try
{
if (subColumnsIterator != null && subColumnsIterator.hasNext())
{
Cell c = subColumnsIterator.next();
return c.withUpdatedName(type.makeCellName(scName, c.name().toByteBuffer()));
}
// Read one more super column
++read;
scName = ByteBufferUtil.readWithShortLength(in);
DeletionInfo delInfo = new DeletionInfo(DeletionTime.serializer.deserialize(in));
/* read the number of columns */
int size = in.readInt();
List subCells = new ArrayList<>(size);
ColumnSerializer colSer = subType(type).columnSerializer();
for (int i = 0; i < size; ++i)
subCells.add(colSer.deserialize(in, flag, expireBefore));
subColumnsIterator = subCells.iterator();
// If the SC was deleted, return that first, otherwise return the first subcolumn
DeletionTime dtime = delInfo.getTopLevelDeletion();
if (!dtime.equals(DeletionTime.LIVE))
return new RangeTombstone(startOf(scName), endOf(scName), dtime);
return next();
}
catch (IOException e)
{
throw new IOError(e);
}
}
public void remove()
{
throw new UnsupportedOperationException();
}
}
private static CellNameType subType(CellNameType type)
{
return new SimpleDenseCellNameType(type.subtype(1));
}
public static CellNameType scNameType(CellNameType type)
{
return new SimpleDenseCellNameType(type.subtype(0));
}
public static AbstractType> getComparatorFor(CFMetaData metadata, ByteBuffer superColumn)
{
return getComparatorFor(metadata, superColumn != null);
}
public static AbstractType> getComparatorFor(CFMetaData metadata, boolean subColumn)
{
return metadata.isSuper()
? metadata.comparator.subtype(subColumn ? 1 : 0)
: metadata.comparator.asAbstractType();
}
// Extract the first component of a columnName, i.e. the super column name
public static ByteBuffer scName(Composite columnName)
{
return columnName.get(0);
}
// Extract the 2nd component of a columnName, i.e. the sub-column name
public static ByteBuffer subName(Composite columnName)
{
return columnName.get(1);
}
public static Composite startOf(ByteBuffer scName)
{
return CellNames.compositeDense(scName).start();
}
public static Composite endOf(ByteBuffer scName)
{
return CellNames.compositeDense(scName).end();
}
public static IDiskAtomFilter fromSCFilter(CellNameType type, ByteBuffer scName, IDiskAtomFilter filter)
{
if (filter instanceof NamesQueryFilter)
return fromSCNamesFilter(type, scName, (NamesQueryFilter)filter);
else
return fromSCSliceFilter(type, scName, (SliceQueryFilter)filter);
}
public static IDiskAtomFilter fromSCNamesFilter(CellNameType type, ByteBuffer scName, NamesQueryFilter filter)
{
if (scName == null)
{
ColumnSlice[] slices = new ColumnSlice[filter.columns.size()];
int i = 0;
for (CellName name : filter.columns)
{
// Note that, because the filter in argument is the one from thrift, 'name' are SimpleDenseCellName.
// So calling name.slice() would be incorrect, as simple cell names don't handle the EOC properly.
// This is why we call buffer() and rebuild a Composite of the right type before call slice().
slices[i++] = type.make(name.toByteBuffer()).slice();
}
return new SliceQueryFilter(slices, false, slices.length, 1);
}
else
{
SortedSet newColumns = new TreeSet<>(type);
for (CellName c : filter.columns)
newColumns.add(type.makeCellName(scName, c.toByteBuffer()));
return filter.withUpdatedColumns(newColumns);
}
}
public static SliceQueryFilter fromSCSliceFilter(CellNameType type, ByteBuffer scName, SliceQueryFilter filter)
{
assert filter.slices.length == 1;
if (scName == null)
{
// The filter is on the super column name
CBuilder builder = type.builder();
Composite start = filter.start().isEmpty()
? Composites.EMPTY
: builder.buildWith(filter.start().toByteBuffer()).withEOC(filter.reversed ? Composite.EOC.END : Composite.EOC.START);
Composite finish = filter.finish().isEmpty()
? Composites.EMPTY
: builder.buildWith(filter.finish().toByteBuffer()).withEOC(filter.reversed ? Composite.EOC.START : Composite.EOC.END);
return new SliceQueryFilter(start, finish, filter.reversed, filter.count, 1);
}
else
{
CBuilder builder = type.builder().add(scName);
Composite start = filter.start().isEmpty()
? builder.build().withEOC(filter.reversed ? Composite.EOC.END : Composite.EOC.START)
: builder.buildWith(filter.start().toByteBuffer());
Composite end = filter.finish().isEmpty()
? builder.build().withEOC(filter.reversed ? Composite.EOC.START : Composite.EOC.END)
: builder.buildWith(filter.finish().toByteBuffer());
return new SliceQueryFilter(start, end, filter.reversed, filter.count);
}
}
}
| |