Check if the node is supposed to produce limit not reached checks.

Cleans up the resources used by the physical operator (if any). In general, all the resources should be cleaned up when the task finishes but operators like SortMergeJoinExec and LimitExec may want eager cleanup to free up tight resources (e.g., memory).

Consume the generated columns or row from current SparkPlan, call its parent's doConsume().

Note that outputVars and row can't both be null.

Generate the Java source code to process the rows from child SparkPlan. This should only be called from consume.

This should be override by subclass to support codegen.

Note: The operator should not assume the existence of an outer processing loop, which it can jump from with "continue;"!

For example, filter could generate this: # code to evaluate the predicate expression, result is isNull1 and value2 if (!isNull1 && value2) { # call consume(), which will call parent.doConsume() }

Note: A plan can either consume the rows as UnsafeRow (row), or a list of variables (input). When consuming as a listing of variables, the code to produce the input is already generated and CodegenContext.currentVars is already set. When consuming as UnsafeRow, implementations need to put row.code in the generated code and set CodegenContext.INPUT_ROW manually. Some plans may need more tweaks as they have different inputs(join build side, aggregate buffer, etc.), or other special cases.

The generated code for doConsume call when aggregate has grouping keys.

Produces the result of the query as an RDD[InternalRow]

Overridden by concrete implementations of SparkPlan.

Produces the result of the query as a broadcast variable.

Overridden by concrete implementations of SparkPlan.

Produces the result of the query as an RDD[ColumnarBatch] if supportsColumnar returns true. By convention the executor that creates a ColumnarBatch is responsible for closing it when it is no longer needed. This allows input formats to be able to reuse batches if needed.

Produces the result of the writes as an RDD[WriterCommitMessage]

Overridden by concrete implementations of SparkPlan.

Overridden by concrete implementations of SparkPlan. It is guaranteed to run before any execute of SparkPlan. This is helpful if we want to set up some state before executing the query, e.g., BroadcastHashJoin uses it to broadcast asynchronously.

Generate the Java source code to process, should be overridden by subclass to support codegen.

doProduce() usually generate the framework, for example, aggregation could generate this:

if (!initialized) { # create a hash map, then build the aggregation hash map # call child.produce() initialized = true; } while (hashmap.hasNext()) { row = hashmap.next(); # build the aggregation results # create variables for results # call consume(), which will call parent.doConsume() if (shouldStop()) return; }

The generated code for doProduce call when aggregate has grouping keys.

Returns source code to evaluate the variables for non-deterministic expressions, and clear the code of evaluated variables, to prevent them to be evaluated twice.

Returns source code to evaluate the variables for required attributes, and clear the code of evaluated variables, to prevent them to be evaluated twice.

Returns source code to evaluate all the variables, and clear the code of them, to prevent them to be evaluated twice.

Returns the result of this query as an RDD[InternalRow] by delegating to doExecute after preparations.

Concrete implementations of SparkPlan should override doExecute.

Returns the result of this query as a broadcast variable by delegating to doExecuteBroadcast after preparations.

Concrete implementations of SparkPlan should override doExecuteBroadcast.

Runs this query returning the result as an array.

Runs this query returning the result as an array, using external Row format.

Returns the result of this query as an RDD[ColumnarBatch] by delegating to doColumnarExecute after preparations.

Concrete implementations of SparkPlan should override doColumnarExecute if supportsColumnar returns true.

Executes a query after preparing the query and adding query plan information to created RDDs for visualization.

Runs this query returning the last n rows as an array.

This is modeled after RDD.take but never runs any job locally on the driver.

Runs this query returning the first n rows as an array.

This is modeled after RDD.take but never runs any job locally on the driver.

Runs this query returning the result as an iterator of InternalRow.

Returns the result of writes as an RDD[WriterCommitMessage] variable by delegating to doExecuteWrite after preparations.

Concrete implementations of SparkPlan should override doExecuteWrite.

The generated code to evaluate aggregate functions.

Returns all the RDDs of InternalRow which generates the input rows.

A sequence of checks which evaluate to true if the downstream Limit operators have not received enough records and reached the limit. If current node is a data producing node, it can leverage this information to stop producing data and complete the data flow earlier. Common data producing nodes are leaf nodes like Range and Scan, and blocking nodes like Sort and Aggregate. These checks should be put into the loop condition of the data producing loop.

A helper method to generate the data producing loop condition according to the limit-not-reached checks.

Overridden make copy also propagates sqlContext to copied plan.

Creates a metric using the specified name.

All the modes of aggregate expressions.

Whether or not the result rows of this operator should be copied before putting into a buffer.

If any operator inside WholeStageCodegen generate multiple rows from a single row (for example, Join), this should be true.

If an operator starts a new pipeline, this should be false.

Whether this operator needs to build hash table.

Whether or not the children of this operator should generate a stop check when consuming input rows. This is used to suppress shouldStop() in a loop of WholeStageCodegen.

This should be false if an operator starts a new pipeline, which means it consumes all rows produced by children but doesn't output row to buffer by calling append(), so the children don't require shouldStop() in the loop of producing rows.

Specifies how data is partitioned across different nodes in the cluster. Note this method may fail if it is invoked before EnsureRequirements is applied since PartitioningCollection requires all its partitionings to have the same number of partitions.

Which SparkPlan is calling produce() of this one. It's itself for the first SparkPlan.

Prepares this SparkPlan for execution. It's idempotent.

Finds scalar subquery expressions in this plan node and starts evaluating them.

Returns Java source code to process the rows from input RDD.

Specifies the data distribution requirements of all the children for this operator. By default it's UnspecifiedDistribution for each child, which means each child can have any distribution.

If an operator overwrites this method, and specifies distribution requirements(excluding UnspecifiedDistribution and BroadcastDistribution) for more than one child, Spark guarantees that the outputs of these children will have same number of partitions, so that the operator can safely zip partitions of these children's result RDDs. Some operators can leverage this guarantee to satisfy some interesting requirement, e.g., non-broadcast joins can specify HashClusteredDistribution(a,b) for its left child, and specify HashClusteredDistribution(c,d) for its right child, then it's guaranteed that left and right child are co-partitioned by a,b/c,d, which means tuples of same value are in the partitions of same index, e.g., (a=1,b=2) and (c=1,d=2) are both in the second partition of left and right child.

Specifies sort order for each partition requirements on the input data for this operator.

Resets all the metrics.

Set logical plan link recursively if unset.

Helper default should stop check code.

Whether this SparkPlan supports whole stage codegen or not.

Return true if this stage of the plan supports columnar execution. A plan can also support row-based execution (see supportsRowBased). Spark will decide which execution to be called during query planning.

Return true if this stage of the plan supports row-based execution. A plan can also support columnar execution (see supportsColumnar). Spark will decide which execution to be called during query planning.

Converts the output of this plan to row-based if it is columnar plan.

The corresponding SortAggregateExec to get same result as this node.

The subset of inputSet those should be evaluated before this plan.

We will use this to insert some code to access those columns that are actually used by current plan before calling doConsume().

The exact java types of the columns that are output in columnar processing mode. This is a performance optimization for code generation and is optional.

Blocks the thread until all subqueries finish evaluation and update the results.