001/*
002 * Copyright (c) 2007-2022 The Cascading Authors. All Rights Reserved.
003 *
004 * Project and contact information: https://cascading.wensel.net/
005 *
006 * This file is part of the Cascading project.
007 *
008 * Licensed under the Apache License, Version 2.0 (the "License");
009 * you may not use this file except in compliance with the License.
010 * You may obtain a copy of the License at
011 *
012 *     http://www.apache.org/licenses/LICENSE-2.0
013 *
014 * Unless required by applicable law or agreed to in writing, software
015 * distributed under the License is distributed on an "AS IS" BASIS,
016 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
017 * See the License for the specific language governing permissions and
018 * limitations under the License.
019 */
020
021package cascading.flow.hadoop.planner;
022
023import cascading.flow.hadoop.planner.rule.partitioner.ConsecutiveTapsNodePartitioner;
024import cascading.flow.hadoop.planner.rule.partitioner.ConsecutiveTapsStepPartitioner;
025import cascading.flow.hadoop.planner.rule.partitioner.GroupTapNodePartitioner;
026import cascading.flow.hadoop.planner.rule.partitioner.MultiTapGroupNodePartitioner;
027import cascading.flow.hadoop.planner.rule.partitioner.StreamedAccumulatedTapsHashJoinPipelinePartitioner;
028import cascading.flow.hadoop.planner.rule.partitioner.StreamedAccumulatedTapsPipelinePartitioner;
029import cascading.flow.hadoop.planner.rule.partitioner.StreamedOnlySourcesPipelinePartitioner;
030import cascading.flow.hadoop.planner.rule.partitioner.StreamedSelfJoinSourcesPipelinePartitioner;
031import cascading.flow.hadoop.planner.rule.partitioner.TapGroupTapStepPartitioner;
032import cascading.flow.hadoop.planner.rule.transformer.RemoveMalformedHashJoinPipelineTransformer;
033import cascading.flow.hadoop.planner.rule.transformer.ReplaceAccumulateTapWithDistCacheTransformer;
034import cascading.flow.hadoop.planner.rule.transformer.TapBalanceCheckpointTransformer;
035import cascading.flow.hadoop.planner.rule.transformer.TapBalanceGroupBlockingHashJoinTransformer;
036import cascading.flow.hadoop.planner.rule.transformer.TapBalanceGroupGroupTransformer;
037import cascading.flow.hadoop.planner.rule.transformer.TapBalanceGroupMergeGroupTransformer;
038import cascading.flow.hadoop.planner.rule.transformer.TapBalanceGroupNonBlockingHashJoinTransformer;
039import cascading.flow.hadoop.planner.rule.transformer.TapBalanceGroupSplitJoinTransformer;
040import cascading.flow.hadoop.planner.rule.transformer.TapBalanceGroupSplitMergeGroupTransformer;
041import cascading.flow.hadoop.planner.rule.transformer.TapBalanceGroupSplitMergeTransformer;
042import cascading.flow.hadoop.planner.rule.transformer.TapBalanceGroupSplitTransformer;
043import cascading.flow.hadoop.planner.rule.transformer.TapBalanceGroupSplitTriangleTransformer;
044import cascading.flow.hadoop.planner.rule.transformer.TapBalanceHashJoinBlockingHashJoinTransformer;
045import cascading.flow.hadoop.planner.rule.transformer.TapBalanceHashJoinSameSourceTransformer;
046import cascading.flow.hadoop.planner.rule.transformer.TapBalanceNonSafePipeSplitTransformer;
047import cascading.flow.hadoop.planner.rule.transformer.TapBalanceNonSafeSplitTransformer;
048import cascading.flow.hadoop.planner.rule.transformer.TapBalanceSameSourceStreamedAccumulatedTransformer;
049import cascading.flow.planner.rule.RuleRegistry;
050import cascading.flow.planner.rule.assertion.BufferAfterEveryAssert;
051import cascading.flow.planner.rule.assertion.EveryAfterBufferAssert;
052import cascading.flow.planner.rule.assertion.LoneGroupAssert;
053import cascading.flow.planner.rule.assertion.MissingGroupAssert;
054import cascading.flow.planner.rule.assertion.SplitBeforeEveryAssert;
055import cascading.flow.planner.rule.transformer.ApplyAssertionLevelTransformer;
056import cascading.flow.planner.rule.transformer.ApplyDebugLevelTransformer;
057import cascading.flow.planner.rule.transformer.RemoveNoOpPipeTransformer;
058
059/**
060 *
061 */
062public class MapReduceHadoopRuleRegistry extends RuleRegistry
063  {
064  public MapReduceHadoopRuleRegistry()
065    {
066//    enableDebugLogging();
067
068    // PreBalance
069    addRule( new LoneGroupAssert() );
070    addRule( new MissingGroupAssert() );
071    addRule( new BufferAfterEveryAssert() );
072    addRule( new EveryAfterBufferAssert() );
073    addRule( new SplitBeforeEveryAssert() );
074
075    // Balance with temporary Taps
076    addRule( new TapBalanceGroupSplitTransformer() );
077    addRule( new TapBalanceGroupSplitJoinTransformer() );
078    addRule( new TapBalanceGroupSplitTriangleTransformer() ); // testGroupSplitToCoGroupsTriangle
079    addRule( new TapBalanceGroupSplitMergeGroupTransformer() );
080    addRule( new TapBalanceGroupSplitMergeTransformer() );
081    addRule( new TapBalanceGroupMergeGroupTransformer() );
082    addRule( new TapBalanceGroupGroupTransformer() );
083    addRule( new TapBalanceCheckpointTransformer() );
084    addRule( new TapBalanceHashJoinSameSourceTransformer() );
085    addRule( new TapBalanceHashJoinBlockingHashJoinTransformer() );
086    addRule( new TapBalanceGroupBlockingHashJoinTransformer() );
087    addRule( new TapBalanceGroupNonBlockingHashJoinTransformer() );
088    addRule( new TapBalanceSameSourceStreamedAccumulatedTransformer() );
089    addRule( new TapBalanceNonSafeSplitTransformer() );
090    addRule( new TapBalanceNonSafePipeSplitTransformer() );
091
092    // PreResolve
093    addRule( new RemoveNoOpPipeTransformer() );
094    addRule( new ApplyAssertionLevelTransformer() );
095    addRule( new ApplyDebugLevelTransformer() );
096    addRule( new ReplaceAccumulateTapWithDistCacheTransformer() );
097
098    // PostResolve
099//    addRule( new CombineAdjacentTapTransformer() );
100
101    // PartitionSteps
102    addRule( new ConsecutiveTapsStepPartitioner() );
103    addRule( new TapGroupTapStepPartitioner() );
104
105    // PartitionNodes
106    addRule( new ConsecutiveTapsNodePartitioner() );
107    addRule( new MultiTapGroupNodePartitioner() );
108    addRule( new GroupTapNodePartitioner() );
109
110    // PartitionPipelines
111    addRule( new StreamedAccumulatedTapsHashJoinPipelinePartitioner() ); // solves streamed/accumulated annotation hiding
112    addRule( new StreamedAccumulatedTapsPipelinePartitioner() );
113    addRule( new StreamedSelfJoinSourcesPipelinePartitioner() );
114    addRule( new StreamedOnlySourcesPipelinePartitioner() );
115
116    // PostPipelines
117    addRule( new RemoveMalformedHashJoinPipelineTransformer() );
118    }
119  }