Skip to content

Reapply "Add ExecutionPlan::apply_expressions() (apache#20337)" (apache#22437) #24018

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
jayshrivastava wants to merge 3 commits into
base: main
Choose a base branch
from
Open

Reapply "Add ExecutionPlan::apply_expressions() (apache#20337)" (apache#22437) #24018

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
58f0e1a
Reapply "Add `ExecutionPlan::apply_expressions()` (#20337)" (#22437)
jayshrivastava Jul 30, 2026
380f5c7
Refine physical expression traversal API
jayshrivastava Jul 30, 2026
5514194
Use apply_expressions for aggregate dynamic filters
jayshrivastava Jul 30, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
10 changes: 10 additions & 0 deletions datafusion-examples/examples/custom_data_source/custom_datasource.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -27,6 +27,7 @@ use datafusion::arrow::array::{UInt8Builder, UInt64Builder};
use datafusion::arrow::datatypes::{DataType, Field, Schema, SchemaRef};
use datafusion::arrow::record_batch::RecordBatch;
use datafusion::common::assert_batches_eq;
use datafusion::common::tree_node::TreeNodeRecursion;
use datafusion::datasource::{TableProvider, TableType, provider_as_source};
use datafusion::error::Result;
use datafusion::execution::context::TaskContext;
Expand Down Expand Up @@ -314,4 +315,13 @@ impl ExecutionPlan for CustomExec {
None,
)?))
}

fn apply_expressions(
&self,
_f: &mut dyn FnMut(
&Arc<dyn datafusion::physical_plan::PhysicalExpr>,
) -> Result<TreeNodeRecursion>,
) -> Result<TreeNodeRecursion> {
Ok(TreeNodeRecursion::Continue)
}
}
10 changes: 10 additions & 0 deletions datafusion-examples/examples/execution_monitoring/memory_pool_execution_plan.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -29,6 +29,7 @@
use arrow::array::record_batch;
use arrow::record_batch::RecordBatch;
use arrow_schema::SchemaRef;
use datafusion::common::tree_node::TreeNodeRecursion;
use datafusion::common::{exec_datafusion_err, internal_err};
use datafusion::datasource::{DefaultTableSource, memory::MemTable};
use datafusion::error::Result;
Expand Down Expand Up @@ -291,4 +292,13 @@ impl ExecutionPlan for BufferingExecutionPlan {
}),
)))
}

fn apply_expressions(
&self,
_f: &mut dyn FnMut(
&Arc<dyn datafusion::physical_plan::PhysicalExpr>,
) -> Result<TreeNodeRecursion>,
) -> Result<TreeNodeRecursion> {
Ok(TreeNodeRecursion::Continue)
}
}
19 changes: 19 additions & 0 deletions datafusion-examples/examples/proto/composed_extension_codec.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -37,6 +37,7 @@ use std::sync::Arc;

use datafusion::common::Result;
use datafusion::common::internal_err;
use datafusion::common::tree_node::TreeNodeRecursion;
use datafusion::execution::TaskContext;
use datafusion::physical_plan::{DisplayAs, ExecutionPlan};
use datafusion::prelude::SessionContext;
Expand Down Expand Up @@ -124,6 +125,15 @@ impl ExecutionPlan for ParentExec {
) -> Result<datafusion::physical_plan::SendableRecordBatchStream> {
unreachable!()
}

fn apply_expressions(
&self,
_f: &mut dyn FnMut(
&Arc<dyn datafusion::physical_plan::PhysicalExpr>,
) -> Result<TreeNodeRecursion>,
) -> Result<TreeNodeRecursion> {
Ok(TreeNodeRecursion::Continue)
}
}

/// A PhysicalExtensionCodec that can serialize and deserialize ParentExec
Expand Down Expand Up @@ -202,6 +212,15 @@ impl ExecutionPlan for ChildExec {
) -> Result<datafusion::physical_plan::SendableRecordBatchStream> {
unreachable!()
}

fn apply_expressions(
&self,
_f: &mut dyn FnMut(
&Arc<dyn datafusion::physical_plan::PhysicalExpr>,
) -> Result<TreeNodeRecursion>,
) -> Result<TreeNodeRecursion> {
Ok(TreeNodeRecursion::Continue)
}
}

/// A PhysicalExtensionCodec that can serialize and deserialize ChildExec
Expand Down
18 changes: 17 additions & 1 deletion datafusion-examples/examples/relation_planner/table_sample.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -117,7 +117,7 @@ use datafusion::{
};
use datafusion_common::{
DFSchemaRef, DataFusionError, Result, Statistics, internal_err, not_impl_err,
plan_datafusion_err, plan_err,
plan_datafusion_err, plan_err, tree_node::TreeNodeRecursion,
};
use datafusion_expr::physical_planning_context::PhysicalPlanningContext;
use datafusion_expr::{
Expand Down Expand Up @@ -749,6 +749,22 @@ impl ExecutionPlan for SampleExec {

Ok(Arc::new(stats))
}

fn apply_expressions(
&self,
f: &mut dyn FnMut(
&Arc<dyn datafusion::physical_plan::PhysicalExpr>,
) -> Result<TreeNodeRecursion>,
) -> Result<TreeNodeRecursion> {
datafusion::physical_plan::apply_expression_roots(
self.cache
.output_ordering()
.into_iter()
.flatten()
.map(|sort_expr| &sort_expr.expr),
f,
)

@jayshrivastava jayshrivastava Jul 30, 2026

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Question for reviewers - this pattern repeats a lot. Is it worth adding a helper?

apply_expression_roots(
    properties
        .output_ordering()
        .into_iter()
        .flatten()
        .map(|sort_expr| &sort_expr.expr),
    f
)

The projection pattern also repeats a bit

      crate::apply_expression_roots(
          self.projector
              .projection()
              .as_ref()
              .iter()
              .map(|proj_expr| &proj_expr.expr),
          f,
      )

}
}

/// Bernoulli sampler: includes each row with probability `(upper - lower)`.
Expand Down
14 changes: 11 additions & 3 deletions datafusion/catalog/src/memory/table.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -31,6 +31,7 @@ use arrow::compute::{and, filter_record_batch};
use arrow::datatypes::{DataType, Field, Schema, SchemaRef};
use arrow::record_batch::RecordBatch;
use datafusion_common::error::Result;
use datafusion_common::tree_node::TreeNodeRecursion;
use datafusion_common::{Constraints, DFSchema, SchemaExt, not_impl_err, plan_err};
use datafusion_datasource::memory::{MemSink, MemorySourceConfig};
use datafusion_datasource::sink::DataSinkExec;
Expand All @@ -39,13 +40,13 @@ use datafusion_expr::dml::InsertOp;
use datafusion_expr::physical_planning_context::PhysicalPlanningContext;
use datafusion_expr::{Expr, SortExpr, TableType};
use datafusion_physical_expr::{
LexOrdering, PhysicalExpr, create_physical_expr, create_physical_sort_exprs,
LexOrdering, create_physical_expr, create_physical_sort_exprs,
};
use datafusion_physical_plan::repartition::RepartitionExec;
use datafusion_physical_plan::stream::RecordBatchStreamAdapter;
use datafusion_physical_plan::{
DisplayAs, DisplayFormatType, ExecutionPlan, Partitioning, PlanProperties,
collect_partitioned,
DisplayAs, DisplayFormatType, ExecutionPlan, Partitioning, PhysicalExpr,
PlanProperties, collect_partitioned,
};
use datafusion_session::Session;

Expand Down Expand Up @@ -597,4 +598,11 @@ impl ExecutionPlan for DmlResultExec {
stream,
)))
}

fn apply_expressions(
&self,
_f: &mut dyn FnMut(&Arc<dyn PhysicalExpr>) -> Result<TreeNodeRecursion>,
) -> Result<TreeNodeRecursion> {
Ok(TreeNodeRecursion::Continue)
}
}
25 changes: 25 additions & 0 deletions datafusion/core/src/physical_planner.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4711,6 +4711,13 @@ mod tests {
) -> Result<SendableRecordBatchStream> {
unimplemented!("NoOpExecutionPlan::execute");
}

fn apply_expressions(
&self,
_f: &mut dyn FnMut(&Arc<dyn PhysicalExpr>) -> Result<TreeNodeRecursion>,
) -> Result<TreeNodeRecursion> {
Ok(TreeNodeRecursion::Continue)
}
}

struct ExpressionExtensionPlanner;
Expand Down Expand Up @@ -4878,6 +4885,12 @@ digraph {
) -> Result<SendableRecordBatchStream> {
unimplemented!()
}
fn apply_expressions(
&self,
_f: &mut dyn FnMut(&Arc<dyn PhysicalExpr>) -> Result<TreeNodeRecursion>,
) -> Result<TreeNodeRecursion> {
Ok(TreeNodeRecursion::Continue)
}
}
impl DisplayAs for OkExtensionNode {
fn fmt_as(&self, _t: DisplayFormatType, f: &mut fmt::Formatter) -> fmt::Result {
Expand Down Expand Up @@ -4924,6 +4937,12 @@ digraph {
) -> Result<SendableRecordBatchStream> {
unimplemented!()
}
fn apply_expressions(
&self,
_f: &mut dyn FnMut(&Arc<dyn PhysicalExpr>) -> Result<TreeNodeRecursion>,
) -> Result<TreeNodeRecursion> {
Ok(TreeNodeRecursion::Continue)
}
}
impl DisplayAs for InvariantFailsExtensionNode {
fn fmt_as(&self, _t: DisplayFormatType, f: &mut fmt::Formatter) -> fmt::Result {
Expand Down Expand Up @@ -5048,6 +5067,12 @@ digraph {
) -> Result<SendableRecordBatchStream> {
unimplemented!()
}
fn apply_expressions(
&self,
_f: &mut dyn FnMut(&Arc<dyn PhysicalExpr>) -> Result<TreeNodeRecursion>,
) -> Result<TreeNodeRecursion> {
Ok(TreeNodeRecursion::Continue)
}
}
impl DisplayAs for ExecutableInvariantFails {
fn fmt_as(&self, _t: DisplayFormatType, f: &mut fmt::Formatter) -> fmt::Result {
Expand Down
17 changes: 17 additions & 0 deletions datafusion/core/tests/custom_sources_cases/mod.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -38,6 +38,7 @@ use datafusion_catalog::Session;
use datafusion_common::cast::as_primitive_array;
use datafusion_common::project_schema;
use datafusion_common::stats::Precision;
use datafusion_common::tree_node::TreeNodeRecursion;
use datafusion_physical_expr::EquivalenceProperties;
use datafusion_physical_plan::PlanProperties;
use datafusion_physical_plan::StatisticsArgs;
Expand Down Expand Up @@ -209,6 +210,22 @@ impl ExecutionPlan for CustomExecutionPlan {
.collect(),
}))
}

fn apply_expressions(
&self,
f: &mut dyn FnMut(
&Arc<dyn datafusion::physical_plan::PhysicalExpr>,
) -> Result<TreeNodeRecursion>,
) -> Result<TreeNodeRecursion> {
datafusion::physical_plan::apply_expression_roots(
self.cache
.output_ordering()
.into_iter()
.flatten()
.map(|sort_expr| &sort_expr.expr),
f,
)
}
}

#[async_trait]
Expand Down
17 changes: 17 additions & 0 deletions datafusion/core/tests/custom_sources_cases/provider_filter_pushdown.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -35,6 +35,7 @@ use datafusion::prelude::*;
use datafusion::scalar::ScalarValue;
use datafusion_catalog::Session;
use datafusion_common::cast::as_primitive_array;
use datafusion_common::tree_node::TreeNodeRecursion;
use datafusion_common::{DataFusionError, internal_err, not_impl_err};
use datafusion_expr::expr::{BinaryExpr, Cast};
use datafusion_functions_aggregate::expr_fn::count;
Expand Down Expand Up @@ -148,6 +149,22 @@ impl ExecutionPlan for CustomPlan {
})),
)))
}

fn apply_expressions(
&self,
f: &mut dyn FnMut(
&Arc<dyn datafusion::physical_plan::PhysicalExpr>,
) -> Result<TreeNodeRecursion>,
) -> Result<TreeNodeRecursion> {
datafusion::physical_plan::apply_expression_roots(
self.cache
.output_ordering()
.into_iter()
.flatten()
.map(|sort_expr| &sort_expr.expr),
f,
)
}
}

#[derive(Clone, Debug)]
Expand Down
17 changes: 17 additions & 0 deletions datafusion/core/tests/custom_sources_cases/statistics.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -33,6 +33,7 @@ use datafusion::{
scalar::ScalarValue,
};
use datafusion_catalog::Session;
use datafusion_common::tree_node::TreeNodeRecursion;
use datafusion_common::{project_schema, stats::Precision};
use datafusion_physical_expr::EquivalenceProperties;
use datafusion_physical_plan::execution_plan::{Boundedness, EmissionType};
Expand Down Expand Up @@ -185,6 +186,22 @@ impl ExecutionPlan for StatisticsValidation {
Ok(Arc::new(self.stats.clone()))
}
}

fn apply_expressions(
&self,
f: &mut dyn FnMut(
&Arc<dyn datafusion::physical_plan::PhysicalExpr>,
) -> Result<TreeNodeRecursion>,
) -> Result<TreeNodeRecursion> {
datafusion::physical_plan::apply_expression_roots(
self.cache
.output_ordering()
.into_iter()
.flatten()
.map(|sort_expr| &sort_expr.expr),
f,
)
}
}

fn init_ctx(stats: Statistics, schema: Schema) -> Result<SessionContext> {
Expand Down
17 changes: 17 additions & 0 deletions datafusion/core/tests/fuzz_cases/once_exec.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -17,6 +17,7 @@

use arrow_schema::SchemaRef;
use datafusion_common::internal_datafusion_err;
use datafusion_common::tree_node::TreeNodeRecursion;
use datafusion_execution::{SendableRecordBatchStream, TaskContext};
use datafusion_physical_expr::{EquivalenceProperties, Partitioning};
use datafusion_physical_plan::execution_plan::{Boundedness, EmissionType};
Expand Down Expand Up @@ -105,4 +106,20 @@ impl ExecutionPlan for OnceExec {

stream.ok_or_else(|| internal_datafusion_err!("Stream already consumed"))
}

fn apply_expressions(
&self,
f: &mut dyn FnMut(
&Arc<dyn datafusion_physical_plan::PhysicalExpr>,
) -> datafusion_common::Result<TreeNodeRecursion>,
) -> datafusion_common::Result<TreeNodeRecursion> {
datafusion_physical_plan::apply_expression_roots(
self.cache
.output_ordering()
.into_iter()
.flatten()
.map(|sort_expr| &sort_expr.expr),
f,
)
}
}
25 changes: 24 additions & 1 deletion datafusion/core/tests/physical_optimizer/enforce_distribution.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -40,7 +40,9 @@ use datafusion::prelude::{SessionConfig, SessionContext};
use datafusion_common::ScalarValue;
use datafusion_common::config::CsvOptions;
use datafusion_common::error::Result;
use datafusion_common::tree_node::{Transformed, TransformedResult, TreeNode};
use datafusion_common::tree_node::{
Transformed, TransformedResult, TreeNode, TreeNodeRecursion,
};
use datafusion_datasource::file_groups::FileGroup;
use datafusion_datasource::file_scan_config::FileScanConfigBuilder;
use datafusion_expr::{JoinType, Operator};
Expand Down Expand Up @@ -203,6 +205,20 @@ impl ExecutionPlan for SortRequiredExec {
)))
}

fn apply_expressions(
&self,
f: &mut dyn FnMut(&Arc<dyn PhysicalExpr>) -> Result<TreeNodeRecursion>,
) -> Result<TreeNodeRecursion> {
datafusion_physical_plan::apply_expression_roots(
self.cache
.output_ordering()
.into_iter()
.flatten()
.map(|sort_expr| &sort_expr.expr),
f,
)
}

fn execute(
&self,
_partition: usize,
Expand Down Expand Up @@ -290,6 +306,13 @@ impl ExecutionPlan for SinglePartitionMaintainsOrderExec {
Ok(Arc::new(Self::new(child)))
}

fn apply_expressions(
&self,
_f: &mut dyn FnMut(&Arc<dyn PhysicalExpr>) -> Result<TreeNodeRecursion>,
) -> Result<TreeNodeRecursion> {
Ok(TreeNodeRecursion::Continue)
}

fn execute(
&self,
_partition: usize,
Expand Down
Loading
Loading