Skip to content

refactor(proto): migrate RepartitionExec serde #23792

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
buraksenn wants to merge 1 commit into
base: main
Choose a base branch
from
+205 −38
Open

refactor(proto): migrate RepartitionExec serde #23792

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
181 changes: 181 additions & 0 deletions datafusion/physical-plan/src/repartition/mod.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1703,6 +1703,187 @@ impl ExecutionPlan for RepartitionExec {
cache: new_properties.into(),
})))
}

#[cfg(feature = "proto")]
fn try_to_proto(
&self,
ctx: &crate::proto::ExecutionPlanEncodeCtx<'_>,
) -> Result<Option<datafusion_proto_models::protobuf::PhysicalPlanNode>> {
use datafusion_proto_models::protobuf;

let input = ctx.encode_child(self.input())?;

// Keep the existing protobuf wire representation unchanged.
let partition_method = match self.partitioning() {
Partitioning::RoundRobinBatch(n) => {
protobuf::partitioning::PartitionMethod::RoundRobin(*n as u64)
}
Partitioning::Hash(exprs, n) => {
let hash_expr = ctx.encode_expressions(exprs)?;
protobuf::partitioning::PartitionMethod::Hash(
protobuf::PhysicalHashRepartition {
hash_expr,
partition_count: *n as u64,
},
)
}
Partitioning::Range(range) => {
let sort_expr = range
.ordering()
.iter()
.map(|sort_expr| {
Ok(protobuf::PhysicalSortExprNode {
expr: Some(Box::new(ctx.encode_expr(&sort_expr.expr)?)),
asc: !sort_expr.options.descending,
nulls_first: sort_expr.options.nulls_first,
})
})
.collect::<Result<Vec<_>>>()?;
let split_point = range
.split_points()
.iter()
.map(|split_point| {
let value = split_point
.values()
.iter()
.map(|value| value.try_into().map_err(Into::into))
.collect::<Result<Vec<_>>>()?;
Ok(protobuf::PhysicalRangeSplitPoint { value })
})
.collect::<Result<Vec<_>>>()?;
protobuf::partitioning::PartitionMethod::Range(
protobuf::PhysicalRangePartitioning {
sort_expr,
split_point,
},
)
}
Partitioning::UnknownPartitioning(n) => {
protobuf::partitioning::PartitionMethod::Unknown(*n as u64)
}
};

Ok(Some(protobuf::PhysicalPlanNode {
physical_plan_type: Some(
protobuf::physical_plan_node::PhysicalPlanType::Repartition(Box::new(
protobuf::RepartitionExecNode {
input: Some(Box::new(input)),
partitioning: Some(protobuf::Partitioning {
partition_method: Some(partition_method),
}),
preserve_order: self.preserve_order(),
},
)),
),
}))
}
}

#[cfg(feature = "proto")]
impl RepartitionExec {
/// Reconstruct a [`RepartitionExec`] from its protobuf representation.
pub fn try_from_proto(
node: &datafusion_proto_models::protobuf::PhysicalPlanNode,
ctx: &crate::proto::ExecutionPlanDecodeCtx<'_>,
) -> Result<Arc<dyn ExecutionPlan>> {
use datafusion_proto_models::protobuf;

let repart = crate::expect_plan_variant!(
node,
protobuf::physical_plan_node::PhysicalPlanType::Repartition,
"RepartitionExec",
);
let input = ctx.decode_required_child(
repart.input.as_deref(),
"RepartitionExec",
"input",
)?;
let input_schema = input.schema();

let partition_method = repart
.partitioning
.as_ref()
.and_then(|p| p.partition_method.as_ref())
.ok_or_else(|| {
datafusion_common::internal_datafusion_err!(
"RepartitionExec is missing required field 'partitioning'"
)
})?;

let partitioning = match partition_method {
protobuf::partitioning::PartitionMethod::RoundRobin(n) => {
Partitioning::RoundRobinBatch(*n as usize)
}
protobuf::partitioning::PartitionMethod::Hash(hash) => {
let exprs = hash
.hash_expr
.iter()
.map(|expr| ctx.decode_expr(expr, input_schema.as_ref()))
.collect::<Result<Vec<_>>>()?;
let partition_count =
usize::try_from(hash.partition_count).map_err(|_| {
datafusion_common::internal_datafusion_err!(
"Hash partition count {} exceeds usize::MAX",
hash.partition_count
)
})?;
Partitioning::Hash(exprs, partition_count)
}
protobuf::partitioning::PartitionMethod::Unknown(n) => {
Partitioning::UnknownPartitioning(*n as usize)
}
protobuf::partitioning::PartitionMethod::Range(range) => {
let sort_exprs = range
.sort_expr
.iter()
.map(|sort_expr| {
let expr = sort_expr.expr.as_ref().ok_or_else(|| {
datafusion_common::internal_datafusion_err!(
"Unexpected empty physical expression"
)
})?;
Ok(PhysicalSortExpr {
expr: ctx.decode_expr(expr, input_schema.as_ref())?,
options: SortOptions {
descending: !sort_expr.asc,
nulls_first: sort_expr.nulls_first,
},
})
})
.collect::<Result<Vec<_>>>()?;
let sort_expr_count = sort_exprs.len();
let ordering = LexOrdering::new(sort_exprs).ok_or_else(|| {
datafusion_common::internal_datafusion_err!(
"Range partitioning requires non-empty ordering"
)
})?;
if ordering.len() != sort_expr_count {
return datafusion_common::internal_err!(
"Range partitioning ordering must not contain duplicate expressions"
);
}
let split_points = range
.split_point
.iter()
.map(|split_point| {
let values = split_point
.value
.iter()
.map(|value| ScalarValue::try_from(value).map_err(Into::into))
.collect::<Result<Vec<_>>>()?;
Ok(SplitPoint::new(values))
})
.collect::<Result<Vec<_>>>()?;
Partitioning::Range(RangePartitioning::try_new(ordering, split_points)?)
}
};

let mut repart_exec = RepartitionExec::try_new(input, partitioning)?;
if repart.preserve_order {
repart_exec = repart_exec.with_preserve_order();
}
Ok(Arc::new(repart_exec))
}
}

impl RepartitionExec {
Expand Down
62 changes: 24 additions & 38 deletions datafusion/proto/src/physical_plan/mod.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -106,8 +106,6 @@ use datafusion_physical_plan::{ExecutionPlan, InputOrderMode, PhysicalExpr, Wind
use prost::Message;
use prost::bytes::BufMut;

use self::from_proto::parse_protobuf_partitioning;
use self::to_proto::serialize_partitioning;
use crate::common::{byte_to_string, str_to_byte};
use crate::convert::{FromProto, TryFromProto};
use crate::convert_required;
Expand Down Expand Up @@ -790,8 +788,8 @@ pub trait PhysicalPlanNodeExt: Sized {
PhysicalPlanType::Merge(_) => {
CoalescePartitionsExec::try_from_proto(self.node(), &decode_ctx)
}
PhysicalPlanType::Repartition(repart) => {
self.try_into_repartition_physical_plan(repart, ctx, proto_converter)
PhysicalPlanType::Repartition(_) => {
RepartitionExec::try_from_proto(self.node(), &decode_ctx)
}
PhysicalPlanType::GlobalLimit(limit) => {
self.try_into_global_limit_physical_plan(limit, ctx, proto_converter)
Expand Down Expand Up @@ -985,14 +983,6 @@ pub trait PhysicalPlanNodeExt: Sized {
return Ok(node);
}

if let Some(exec) = plan.downcast_ref::<RepartitionExec>() {
return protobuf::PhysicalPlanNode::try_from_repartition_exec(
exec,
codec,
proto_converter,
);
}

if let Some(exec) = plan.downcast_ref::<SortExec>() {
return protobuf::PhysicalPlanNode::try_from_sort_exec(
exec,
Expand Down Expand Up @@ -1482,25 +1472,27 @@ pub trait PhysicalPlanNodeExt: Sized {
CoalescePartitionsExec::try_from_proto(&node, &decode_ctx)
}

#[deprecated(
since = "55.0.0",
note = "unused by DataFusion; `RepartitionExec` deserializes itself via `RepartitionExec::try_from_proto`"
)]
fn try_into_repartition_physical_plan(
&self,
repart: &protobuf::RepartitionExecNode,
ctx: &PhysicalPlanDecodeContext<'_>,
proto_converter: &dyn PhysicalProtoConverterExtension,
) -> Result<Arc<dyn ExecutionPlan>> {
let input: Arc<dyn ExecutionPlan> =
into_physical_plan(&repart.input, ctx, proto_converter)?;
let partitioning = parse_protobuf_partitioning(
repart.partitioning.as_ref(),
let node = protobuf::PhysicalPlanNode {
physical_plan_type: Some(PhysicalPlanType::Repartition(Box::new(
repart.clone(),
))),
};
let decoder = ConverterPlanDecoder {
ctx,
input.schema().as_ref(),
proto_converter,
)?;
let mut repart_exec = RepartitionExec::try_new(input, partitioning.unwrap())?;
if repart.preserve_order {
repart_exec = repart_exec.with_preserve_order();
}
Ok(Arc::new(repart_exec))
};
let decode_ctx = ExecutionPlanDecodeCtx::new(&decoder);
RepartitionExec::try_from_proto(&node, &decode_ctx)
}

fn try_into_global_limit_physical_plan(
Expand Down Expand Up @@ -3511,28 +3503,22 @@ pub trait PhysicalPlanNodeExt: Sized {
})
}

#[deprecated(
since = "55.0.0",
note = "unused by DataFusion; `RepartitionExec` serializes itself via `ExecutionPlan::try_to_proto`"
)]
fn try_from_repartition_exec(
exec: &RepartitionExec,
codec: &dyn PhysicalExtensionCodec,
proto_converter: &dyn PhysicalProtoConverterExtension,
) -> Result<protobuf::PhysicalPlanNode> {
let input = protobuf::PhysicalPlanNode::try_from_physical_plan_with_converter(
exec.input().to_owned(),
let encoder = ConverterPlanEncoder {
codec,
proto_converter,
)?;

let pb_partitioning =
serialize_partitioning(exec.partitioning(), codec, proto_converter)?;

Ok(protobuf::PhysicalPlanNode {
physical_plan_type: Some(PhysicalPlanType::Repartition(Box::new(
protobuf::RepartitionExecNode {
input: Some(Box::new(input)),
partitioning: Some(pb_partitioning),
preserve_order: exec.preserve_order(),
},
))),
};
let encode_ctx = ExecutionPlanEncodeCtx::new(&encoder);
exec.try_to_proto(&encode_ctx)?.ok_or_else(|| {
internal_datafusion_err!("RepartitionExec is not serializable")
})
}

Expand Down
Loading