[
https://issues.apache.org/jira/browse/FLINK-7?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15044496#comment-15044496
]
ASF GitHub Bot commented on FLINK-7:
------------------------------------
Github user ChengXiangLi commented on a diff in the pull request:
https://github.com/apache/flink/pull/1255#discussion_r46792344
--- Diff:
flink-optimizer/src/main/java/org/apache/flink/optimizer/traversals/RangePartitionRewriter.java
---
@@ -0,0 +1,194 @@
+/*
+ * 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.flink.optimizer.traversals;
+
+import org.apache.flink.api.common.distributions.CommonRangeBoundaries;
+import org.apache.flink.api.common.operators.UnaryOperatorInformation;
+import org.apache.flink.api.common.operators.base.GroupReduceOperatorBase;
+import org.apache.flink.api.common.operators.base.MapOperatorBase;
+import org.apache.flink.api.common.operators.base.MapPartitionOperatorBase;
+import org.apache.flink.api.common.operators.util.FieldList;
+import org.apache.flink.api.common.typeinfo.BasicTypeInfo;
+import org.apache.flink.api.common.typeinfo.TypeInformation;
+import org.apache.flink.api.common.typeutils.TypeComparatorFactory;
+import org.apache.flink.api.java.typeutils.RecordTypeInfo;
+import org.apache.flink.runtime.io.network.DataExchangeMode;
+import org.apache.flink.runtime.operators.udf.AssignRangeIndex;
+import org.apache.flink.runtime.operators.udf.PartitionIDRemoveWrapper;
+import org.apache.flink.runtime.operators.udf.RangeBoundaryBuilder;
+import org.apache.flink.api.java.functions.SampleInCoordinator;
+import org.apache.flink.api.java.functions.SampleInPartition;
+import org.apache.flink.api.java.sampling.IntermediateSampleData;
+import org.apache.flink.api.java.tuple.Tuple2;
+import org.apache.flink.api.java.typeutils.TupleTypeInfo;
+import org.apache.flink.api.java.typeutils.TypeExtractor;
+import org.apache.flink.optimizer.dag.GroupReduceNode;
+import org.apache.flink.optimizer.dag.MapNode;
+import org.apache.flink.optimizer.dag.MapPartitionNode;
+import org.apache.flink.optimizer.dag.TempMode;
+import org.apache.flink.optimizer.plan.Channel;
+import org.apache.flink.optimizer.plan.NamedChannel;
+import org.apache.flink.optimizer.plan.OptimizedPlan;
+import org.apache.flink.optimizer.plan.PlanNode;
+import org.apache.flink.optimizer.plan.SingleInputPlanNode;
+import org.apache.flink.optimizer.util.Utils;
+import org.apache.flink.runtime.operators.DriverStrategy;
+import org.apache.flink.runtime.operators.shipping.ShipStrategyType;
+import org.apache.flink.util.Visitor;
+
+import java.util.ArrayList;
+import java.util.LinkedList;
+import java.util.List;
+
+public class RangePartitionRewriter implements Visitor<PlanNode> {
+ final static long SEED = org.apache.flink.api.java.Utils.RNG.nextLong();
+
+ final OptimizedPlan plan;
+
+ public RangePartitionRewriter(OptimizedPlan plan) {
+ this.plan = plan;
+ }
+
+ @Override
+ public boolean preVisit(PlanNode visitable) {
+ return true;
+ }
+
+ @Override
+ public void postVisit(PlanNode visitable) {
+ final List<Channel> outgoingChannels =
visitable.getOutgoingChannels();
+ final List<Channel> newOutGoingChannels = new LinkedList<>();
+ final List<Channel> toBeRemoveChannels = new ArrayList<>();
+ for (Channel channel : outgoingChannels) {
+ ShipStrategyType shipStrategy =
channel.getShipStrategy();
+ if (shipStrategy == ShipStrategyType.PARTITION_RANGE) {
+ TypeInformation<?> outputType =
channel.getSource().getProgramOperator().getOperatorInfo().getOutputType();
+ // Do not optimize for record type, it's a
special case for range partitioner, and should be removed later.
+ if (!(outputType instanceof RecordTypeInfo)) {
--- End diff --
Thanks for the remind, @fhueske , I've added verification on
`PartitionOperator` level and related unit test. BTW, would the `Record` type
be fully discarded finally? it seems been broken on many features, but there
are still lots of related code remain in Flink.
> [GitHub] Enable Range Partitioner
> ---------------------------------
>
> Key: FLINK-7
> URL: https://issues.apache.org/jira/browse/FLINK-7
> Project: Flink
> Issue Type: Sub-task
> Components: Distributed Runtime
> Reporter: GitHub Import
> Assignee: Chengxiang Li
> Fix For: pre-apache
>
>
> The range partitioner is currently disabled. We need to implement the
> following aspects:
> 1) Distribution information, if available, must be propagated back together
> with the ordering property.
> 2) A generic bucket lookup structure (currently specific to PactRecord).
> Tests to re-enable after fixing this issue:
> - TeraSortITCase
> - GlobalSortingITCase
> - GlobalSortingMixedOrderITCase
> ---------------- Imported from GitHub ----------------
> Url: https://github.com/stratosphere/stratosphere/issues/7
> Created by: [StephanEwen|https://github.com/StephanEwen]
> Labels: core, enhancement, optimizer,
> Milestone: Release 0.4
> Assignee: [fhueske|https://github.com/fhueske]
> Created at: Fri Apr 26 13:48:24 CEST 2013
> State: open
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
