ifesdjeen commented on code in PR #3060: URL: https://github.com/apache/cassandra/pull/3060#discussion_r1531710977
########## test/harry/main/org/apache/cassandra/harry/dsl/OverridingCkGenerator.java: ########## @@ -0,0 +1,187 @@ +/* + * 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.harry.dsl; + +import java.util.ArrayList; +import java.util.HashMap; +import java.util.List; +import java.util.Map; + +import org.apache.cassandra.harry.ddl.ColumnSpec; +import org.apache.cassandra.harry.gen.Bijections; +import org.apache.cassandra.harry.gen.DataGenerators; + +/** + * A class that helps to override parts of clustering key. The tricky part about CK overrides is that Harry model makes + * an assumption about the ordering of clustering keys, which means clusterings have to be sorted in the same way their + * descriptors are. This, combined with reverse types, makes managing this state somewhat tricky at times. + * + * Additionally, Relation in delete/select query receives individual CK descriptors (i.e. after they have been sliced), + * while most other queries usually operate on inflated clustering key. All of this is required for efficient _stateless_ + * validation, but makes overrides a bit less intuitive. + * + * To summarise: overrides for inflating are done for individual clustering key columns. Overrides for deflating a clustering + * operate on an entire key. Main reason for this is to allow having same string in several rows for the same column. + */ +public class OverridingCkGenerator extends DataGenerators.KeyGenerator +{ + private final DataGenerators.KeyGenerator delegate; + private final KeyPartOverride[] columnValueOverrides; + private final List<ColumnSpec<?>> columnSpecOverrides; + private final Map<ArrayWrapper, Long> valueToDescriptor; + + // Had to be a static method because you can not call super after you have initialised any fields + public static OverridingCkGenerator make(DataGenerators.KeyGenerator delegate) + { + KeyPartOverride[] columnValueOverrides = new KeyPartOverride[delegate.columns.size()]; + List<ColumnSpec<?>> columnSpecOverrides = new ArrayList<>(); + for (int i = 0; i < delegate.columns.size(); i++) + { + columnValueOverrides[i] = new KeyPartOverride<>(delegate.columns.get(i).generator()); + columnSpecOverrides.add(delegate.columns.get(i).override(columnValueOverrides[i])); + } + assert columnValueOverrides.length == columnSpecOverrides.size(); + return new OverridingCkGenerator(delegate, columnValueOverrides, columnSpecOverrides); + } + + private OverridingCkGenerator(DataGenerators.KeyGenerator delegate, + KeyPartOverride[] columnValueOverrides, + List<ColumnSpec<?>> columnSpecOverrides) + { + super(columnSpecOverrides); + this.columnValueOverrides = columnValueOverrides; + this.columnSpecOverrides = columnSpecOverrides; + this.delegate = delegate; + this.valueToDescriptor = new HashMap<>(); + } + + public void override(long descriptor, Object[] value) + { + long[] slices = delegate.slice(descriptor); + for (int i = 0; i < slices.length; i++) + columnValueOverrides[i].override(slices[i], value[i]); + + // We _always_ deflate clustering key as a package, since we can not validate a clustering key without all components anyways. + valueToDescriptor.put(new ArrayWrapper(value), descriptor); + } + + @Override + public Object[] inflate(long descriptor) + { + return DataGenerators.inflateKey(columnSpecOverrides, descriptor, slice(descriptor)); + } + + @Override + public long deflate(Object[] value) + { + Long descriptor = valueToDescriptor.get(new ArrayWrapper(value)); + if (descriptor != null) + return descriptor; + + return delegate.deflate(value); + } + + @Override + public int byteSize() + { + return delegate.byteSize(); + } + + @Override + public int compare(long l, long r) + { + return delegate.byteSize(); + } + + @Override + public long[] slice(long descriptor) + { + return delegate.slice(descriptor); + } + + @Override + public long stitch(long[] parts) + { + return delegate.stitch(parts); + } + + @Override + public long minValue(int idx) + { + return delegate.minValue(idx); + } + + @Override + public long maxValue(int idx) + { + return delegate.maxValue(idx); + } + + public static class KeyPartOverride<T> implements Bijections.Bijection<T> Review Comment: Yeah, good point. Probably the overhead of `valueToDescriptor` is not worth custom class. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
