Tabnine Logo
RangePartitionRewriter
Code IndexAdd Tabnine to your IDE (free)

How to use
RangePartitionRewriter
in
org.apache.flink.optimizer.traversals

Best Java code snippets using org.apache.flink.optimizer.traversals.RangePartitionRewriter (Showing top 10 results out of 315)

origin: apache/flink

plan.accept(new RangePartitionRewriter(plan));
origin: apache/flink

@Override
public void postVisit(PlanNode node) {
  if(node instanceof IterationPlanNode) {
    IterationPlanNode iNode = (IterationPlanNode)node;
    if(!visitedIterationNodes.contains(iNode)) {
      visitedIterationNodes.add(iNode);
      iNode.acceptForStepFunction(this);
    }
  }
  final Iterable<Channel> inputChannels = node.getInputs();
  for (Channel channel : inputChannels) {
    ShipStrategyType shipStrategy = channel.getShipStrategy();
    // Make sure we only optimize the DAG for range partition, and do not optimize multi times.
    if (shipStrategy == ShipStrategyType.PARTITION_RANGE) {
      if(channel.getDataDistribution() == null) {
        if (node.isOnDynamicPath()) {
          throw new InvalidProgramException("Range Partitioning not supported within iterations if users do not supply the data distribution.");
        }
        PlanNode channelSource = channel.getSource();
        List<Channel> newSourceOutputChannels = rewriteRangePartitionChannel(channel);
        channelSource.getOutgoingChannels().remove(channel);
        channelSource.getOutgoingChannels().addAll(newSourceOutputChannels);
      }
    }
  }
}
origin: org.apache.flink/flink-optimizer_2.10

@Override
public void postVisit(PlanNode node) {
  if(node instanceof IterationPlanNode) {
    IterationPlanNode iNode = (IterationPlanNode)node;
    if(!visitedIterationNodes.contains(iNode)) {
      visitedIterationNodes.add(iNode);
      iNode.acceptForStepFunction(this);
    }
  }
  final Iterable<Channel> inputChannels = node.getInputs();
  for (Channel channel : inputChannels) {
    ShipStrategyType shipStrategy = channel.getShipStrategy();
    // Make sure we only optimize the DAG for range partition, and do not optimize multi times.
    if (shipStrategy == ShipStrategyType.PARTITION_RANGE) {
      if(channel.getDataDistribution() == null) {
        if (node.isOnDynamicPath()) {
          throw new InvalidProgramException("Range Partitioning not supported within iterations if users do not supply the data distribution.");
        }
        PlanNode channelSource = channel.getSource();
        List<Channel> newSourceOutputChannels = rewriteRangePartitionChannel(channel);
        channelSource.getOutgoingChannels().remove(channel);
        channelSource.getOutgoingChannels().addAll(newSourceOutputChannels);
      }
    }
  }
}
origin: com.alibaba.blink/flink-optimizer

plan.accept(new RangePartitionRewriter(plan));
origin: com.alibaba.blink/flink-optimizer

@Override
public void postVisit(PlanNode node) {
  if(node instanceof IterationPlanNode) {
    IterationPlanNode iNode = (IterationPlanNode)node;
    if(!visitedIterationNodes.contains(iNode)) {
      visitedIterationNodes.add(iNode);
      iNode.acceptForStepFunction(this);
    }
  }
  final Iterable<Channel> inputChannels = node.getInputs();
  for (Channel channel : inputChannels) {
    ShipStrategyType shipStrategy = channel.getShipStrategy();
    // Make sure we only optimize the DAG for range partition, and do not optimize multi times.
    if (shipStrategy == ShipStrategyType.PARTITION_RANGE) {
      if(channel.getDataDistribution() == null) {
        if (node.isOnDynamicPath()) {
          throw new InvalidProgramException("Range Partitioning not supported within iterations if users do not supply the data distribution.");
        }
        PlanNode channelSource = channel.getSource();
        List<Channel> newSourceOutputChannels = rewriteRangePartitionChannel(channel);
        channelSource.getOutgoingChannels().remove(channel);
        channelSource.getOutgoingChannels().addAll(newSourceOutputChannels);
      }
    }
  }
}
origin: org.apache.flink/flink-optimizer_2.11

plan.accept(new RangePartitionRewriter(plan));
origin: org.apache.flink/flink-optimizer

@Override
public void postVisit(PlanNode node) {
  if(node instanceof IterationPlanNode) {
    IterationPlanNode iNode = (IterationPlanNode)node;
    if(!visitedIterationNodes.contains(iNode)) {
      visitedIterationNodes.add(iNode);
      iNode.acceptForStepFunction(this);
    }
  }
  final Iterable<Channel> inputChannels = node.getInputs();
  for (Channel channel : inputChannels) {
    ShipStrategyType shipStrategy = channel.getShipStrategy();
    // Make sure we only optimize the DAG for range partition, and do not optimize multi times.
    if (shipStrategy == ShipStrategyType.PARTITION_RANGE) {
      if(channel.getDataDistribution() == null) {
        if (node.isOnDynamicPath()) {
          throw new InvalidProgramException("Range Partitioning not supported within iterations if users do not supply the data distribution.");
        }
        PlanNode channelSource = channel.getSource();
        List<Channel> newSourceOutputChannels = rewriteRangePartitionChannel(channel);
        channelSource.getOutgoingChannels().remove(channel);
        channelSource.getOutgoingChannels().addAll(newSourceOutputChannels);
      }
    }
  }
}
origin: org.apache.flink/flink-optimizer_2.10

plan.accept(new RangePartitionRewriter(plan));
origin: org.apache.flink/flink-optimizer_2.11

@Override
public void postVisit(PlanNode node) {
  if(node instanceof IterationPlanNode) {
    IterationPlanNode iNode = (IterationPlanNode)node;
    if(!visitedIterationNodes.contains(iNode)) {
      visitedIterationNodes.add(iNode);
      iNode.acceptForStepFunction(this);
    }
  }
  final Iterable<Channel> inputChannels = node.getInputs();
  for (Channel channel : inputChannels) {
    ShipStrategyType shipStrategy = channel.getShipStrategy();
    // Make sure we only optimize the DAG for range partition, and do not optimize multi times.
    if (shipStrategy == ShipStrategyType.PARTITION_RANGE) {
      if(channel.getDataDistribution() == null) {
        if (node.isOnDynamicPath()) {
          throw new InvalidProgramException("Range Partitioning not supported within iterations if users do not supply the data distribution.");
        }
        PlanNode channelSource = channel.getSource();
        List<Channel> newSourceOutputChannels = rewriteRangePartitionChannel(channel);
        channelSource.getOutgoingChannels().remove(channel);
        channelSource.getOutgoingChannels().addAll(newSourceOutputChannels);
      }
    }
  }
}
origin: org.apache.flink/flink-optimizer

plan.accept(new RangePartitionRewriter(plan));
org.apache.flink.optimizer.traversalsRangePartitionRewriter

Most used methods

  • <init>
  • rewriteRangePartitionChannel

Popular in Java

  • Creating JSON documents from java classes using gson
  • setScale (BigDecimal)
  • scheduleAtFixedRate (ScheduledExecutorService)
  • runOnUiThread (Activity)
  • OutputStream (java.io)
    A writable sink for bytes.Most clients will use output streams that write data to the file system (
  • ConnectException (java.net)
    A ConnectException is thrown if a connection cannot be established to a remote host on a specific po
  • ArrayList (java.util)
    ArrayList is an implementation of List, backed by an array. All optional operations including adding
  • Enumeration (java.util)
    A legacy iteration interface.New code should use Iterator instead. Iterator replaces the enumeration
  • Notification (javax.management)
  • StringUtils (org.apache.commons.lang)
    Operations on java.lang.String that arenull safe. * IsEmpty/IsBlank - checks if a String contains
  • Github Copilot alternatives
Tabnine Logo
  • Products

    Search for Java codeSearch for JavaScript code
  • IDE Plugins

    IntelliJ IDEAWebStormVisual StudioAndroid StudioEclipseVisual Studio CodePyCharmSublime TextPhpStormVimGoLandRubyMineEmacsJupyter NotebookJupyter LabRiderDataGripAppCode
  • Company

    About UsContact UsCareers
  • Resources

    FAQBlogTabnine AcademyTerms of usePrivacy policyJava Code IndexJavascript Code Index
Get Tabnine for your IDE now