Derive Clone for more ExecutionPlans (#13203)

* Derive `Clone` for more ExecutionPlans

* improve docs

Author: alamb

datafusion/physical-plan/src/aggregates/mod.rs

@@ -344,7 +344,7 @@ impl From<StreamType> for SendableRecordBatchStream {
 }
 
 /// Hash aggregate execution plan
-#[derive(Debug)]
+#[derive(Debug, Clone)]
 pub struct AggregateExec {
     /// Aggregation mode (full, partial)
     mode: AggregateMode,

datafusion/physical-plan/src/coalesce_batches.rs

@@ -48,7 +48,7 @@ use futures::stream::{Stream, StreamExt};
 /// reaches the `fetch` value.
 ///
 /// See [`BatchCoalescer`] for more information
-#[derive(Debug)]
+#[derive(Debug, Clone)]
 pub struct CoalesceBatchesExec {
     /// The input plan
     input: Arc<dyn ExecutionPlan>,

datafusion/physical-plan/src/coalesce_partitions.rs

@@ -36,7 +36,7 @@ use datafusion_execution::TaskContext;
 
 /// Merge execution plan executes partitions in parallel and combines them into a single
 /// partition. No guarantees are made about the order of the resulting partition.
-#[derive(Debug)]
+#[derive(Debug, Clone)]
 pub struct CoalescePartitionsExec {
     /// Input execution plan
     input: Arc<dyn ExecutionPlan>,

datafusion/physical-plan/src/empty.rs

@@ -35,7 +35,7 @@ use datafusion_physical_expr::EquivalenceProperties;
 use log::trace;
 
 /// Execution plan for empty relation with produce_one_row=false
-#[derive(Debug)]
+#[derive(Debug, Clone)]
 pub struct EmptyExec {
     /// The schema for the produced row
     schema: SchemaRef,

datafusion/physical-plan/src/filter.rs

@@ -54,7 +54,7 @@ use log::trace;
 
 /// FilterExec evaluates a boolean predicate against all input batches to determine which rows to
 /// include in its output batches.
-#[derive(Debug)]
+#[derive(Debug, Clone)]
 pub struct FilterExec {
     /// The expression to filter on. This expression must evaluate to a boolean value.
     predicate: Arc<dyn PhysicalExpr>,

datafusion/physical-plan/src/insert.rs

@@ -79,6 +79,7 @@ pub type FileSinkExec = DataSinkExec;
 /// Execution plan for writing record batches to a [`DataSink`]
 ///
 /// Returns a single row with the number of values written
+#[derive(Clone)]
 pub struct DataSinkExec {
     /// Input plan that produces the record batches to be written.
     input: Arc<dyn ExecutionPlan>,

datafusion/physical-plan/src/joins/cross_join.rs

@@ -49,8 +49,13 @@ use futures::{ready, Stream, StreamExt, TryStreamExt};
 /// Data of the left side
 type JoinLeftData = (RecordBatch, MemoryReservation);
 
+#[allow(rustdoc::private_intra_doc_links)]
 /// executes partitions in parallel and combines them into a set of
 /// partitions by combining all values from the left with all values on the right
+///
+/// Note that the `Clone` trait is not implemented for this struct due to the
+/// `left_fut` [`OnceAsync`], which is used to coordinate the loading of the
+/// left side with the processing in each output stream.
 #[derive(Debug)]
 pub struct CrossJoinExec {
     /// left (build) side which gets loaded in memory

datafusion/physical-plan/src/joins/hash_join.rs

@@ -136,6 +136,7 @@ impl JoinLeftData {
     }
 }
 
+#[allow(rustdoc::private_intra_doc_links)]
 /// Join execution plan: Evaluates eqijoin predicates in parallel on multiple
 /// partitions using a hash table and an optional filter list to apply post
 /// join.
@@ -293,6 +294,10 @@ impl JoinLeftData {
 ///                       │  "dimension"  │     │    "fact"     │
 ///                       └───────────────┘     └───────────────┘
 /// ```
+///
+/// Note that the `Clone` trait is not implemented for this struct due to the
+/// `left_fut` [`OnceAsync`], which is used to coordinate the loading of the
+/// left side with the processing in each output stream.
 #[derive(Debug)]
 pub struct HashJoinExec {
     /// left (build) side which gets hashed

datafusion/physical-plan/src/joins/nested_loop_join.rs

@@ -105,6 +105,7 @@ impl JoinLeftData {
     }
 }
 
+#[allow(rustdoc::private_intra_doc_links)]
 /// NestedLoopJoinExec is build-probe join operator, whose main task is to
 /// perform joins without any equijoin conditions in `ON` clause.
 ///
@@ -140,6 +141,9 @@ impl JoinLeftData {
 /// "reports" about probe phase completion (which means that "visited" bitmap won't be
 /// updated anymore), and only the last thread, reporting about completion, will return output.
 ///
+/// Note that the `Clone` trait is not implemented for this struct due to the
+/// `left_fut` [`OnceAsync`], which is used to coordinate the loading of the
+/// left side with the processing in each output stream.
 #[derive(Debug)]
 pub struct NestedLoopJoinExec {
     /// left side

datafusion/physical-plan/src/joins/sort_merge_join.rs

@@ -71,7 +71,7 @@ use crate::{
 
 /// join execution plan executes partitions in parallel and combines them into a set of
 /// partitions.
-#[derive(Debug)]
+#[derive(Debug, Clone)]
 pub struct SortMergeJoinExec {
     /// Left sorted joining execution plan
     pub left: Arc<dyn ExecutionPlan>,