Skip to content
Draft
Show file tree
Hide file tree
Changes from all commits
Commits
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
4 changes: 4 additions & 0 deletions Cargo.lock

Some generated files are not rendered by default. Learn more about how customized files appear on GitHub.

8 changes: 8 additions & 0 deletions datafusion/datasource-csv/Cargo.toml
Original file line number Diff line number Diff line change
Expand Up @@ -30,6 +30,13 @@ version.workspace = true
[package.metadata.docs.rs]
all-features = true

[features]
proto = [
"dep:datafusion-proto-models",
"datafusion-datasource/proto",
"datafusion-physical-plan/proto",
]

[dependencies]
arrow = { workspace = true }
async-trait = { workspace = true }
Expand All @@ -41,6 +48,7 @@ datafusion-execution = { workspace = true }
datafusion-expr = { workspace = true }
datafusion-physical-expr-common = { workspace = true }
datafusion-physical-plan = { workspace = true }
datafusion-proto-models = { workspace = true, optional = true }
datafusion-session = { workspace = true }
futures = { workspace = true }
object_store = { workspace = true }
Expand Down
87 changes: 87 additions & 0 deletions datafusion/datasource-csv/src/file_format.rs
Original file line number Diff line number Diff line change
Expand Up @@ -826,6 +826,93 @@ impl DataSink for CsvSink {
) -> Result<u64> {
FileSink::write_all(self, data, context).await
}

#[cfg(feature = "proto")]
fn try_to_proto(
&self,
input: datafusion_proto_models::protobuf::PhysicalPlanNode,
sort_order: Option<
datafusion_proto_models::protobuf::PhysicalSortExprNodeCollection,
>,
sink_schema: &Schema,
) -> Result<Option<datafusion_proto_models::protobuf::PhysicalPlanNode>> {
use datafusion_proto_models::protobuf;
use protobuf::physical_plan_node::PhysicalPlanType;

let sink = protobuf::CsvSink {
config: Some(self.config.to_proto()?),
writer_options: Some(self.writer_options().try_into()?),
};
let node = protobuf::CsvSinkExecNode {
input: Some(Box::new(input)),
sink: Some(sink),
sink_schema: Some(sink_schema.try_into()?),
sort_order,
};
Ok(Some(protobuf::PhysicalPlanNode {
physical_plan_type: Some(PhysicalPlanType::CsvSink(Box::new(node))),
}))
}
}

#[cfg(feature = "proto")]
impl CsvSink {
/// Reconstructs a [`DataSinkExec`] containing a `CsvSink` from protobuf.
pub fn try_from_proto(
node: &datafusion_proto_models::protobuf::PhysicalPlanNode,
ctx: &datafusion_physical_plan::proto::ExecutionPlanDecodeCtx<'_>,
) -> Result<Arc<dyn ExecutionPlan>> {
use datafusion_datasource::file_sink_config::parse_sink_sort_order;
use datafusion_proto_models::protobuf;

let sink_node = match &node.physical_plan_type {
Some(protobuf::physical_plan_node::PhysicalPlanType::CsvSink(sink)) => {
sink.as_ref()
}
_ => {
return datafusion_common::internal_err!(
"PhysicalPlanNode is not a CsvSink"
);
}
};
let input = ctx.decode_required_child(
sink_node.input.as_deref(),
"CsvSinkExecNode",
"input",
)?;
let proto_sink = sink_node.sink.as_ref().ok_or_else(|| {
datafusion_common::internal_datafusion_err!(
"CsvSinkExecNode is missing required field 'sink'"
)
})?;
let config =
FileSinkConfig::from_proto(proto_sink.config.as_ref().ok_or_else(|| {
datafusion_common::internal_datafusion_err!(
"CsvSink is missing required field 'config'"
)
})?)?;
let writer_options = proto_sink
.writer_options
.as_ref()
.ok_or_else(|| {
datafusion_common::internal_datafusion_err!(
"CsvSink is missing required field 'writer_options'"
)
})?
.try_into()?;
let data_sink = CsvSink::new(config, writer_options);
let sort_order = parse_sink_sort_order(
sink_node.sort_order.as_ref(),
ctx,
input.schema().as_ref(),
)?;

Ok(Arc::new(DataSinkExec::new(
input,
Arc::new(data_sink),
sort_order,
)))
}
}

#[cfg(test)]
Expand Down
8 changes: 8 additions & 0 deletions datafusion/datasource-json/Cargo.toml
Original file line number Diff line number Diff line change
Expand Up @@ -30,6 +30,13 @@ version.workspace = true
[package.metadata.docs.rs]
all-features = true

[features]
proto = [
"dep:datafusion-proto-models",
"datafusion-datasource/proto",
"datafusion-physical-plan/proto",
]

[dependencies]
arrow = { workspace = true }
async-trait = { workspace = true }
Expand All @@ -41,6 +48,7 @@ datafusion-execution = { workspace = true }
datafusion-expr = { workspace = true }
datafusion-physical-expr-common = { workspace = true }
datafusion-physical-plan = { workspace = true }
datafusion-proto-models = { workspace = true, optional = true }
datafusion-session = { workspace = true }
futures = { workspace = true }
object_store = { workspace = true }
Expand Down
87 changes: 87 additions & 0 deletions datafusion/datasource-json/src/file_format.rs
Original file line number Diff line number Diff line change
Expand Up @@ -490,6 +490,93 @@ impl DataSink for JsonSink {
) -> Result<u64> {
FileSink::write_all(self, data, context).await
}

#[cfg(feature = "proto")]
fn try_to_proto(
&self,
input: datafusion_proto_models::protobuf::PhysicalPlanNode,
sort_order: Option<
datafusion_proto_models::protobuf::PhysicalSortExprNodeCollection,
>,
sink_schema: &Schema,
) -> Result<Option<datafusion_proto_models::protobuf::PhysicalPlanNode>> {
use datafusion_proto_models::protobuf;
use protobuf::physical_plan_node::PhysicalPlanType;

let sink = protobuf::JsonSink {
config: Some(self.config.to_proto()?),
writer_options: Some(self.writer_options().try_into()?),
};
let node = protobuf::JsonSinkExecNode {
input: Some(Box::new(input)),
sink: Some(sink),
sink_schema: Some(sink_schema.try_into()?),
sort_order,
};
Ok(Some(protobuf::PhysicalPlanNode {
physical_plan_type: Some(PhysicalPlanType::JsonSink(Box::new(node))),
}))
}
}

#[cfg(feature = "proto")]
impl JsonSink {
/// Reconstructs a [`DataSinkExec`] containing a `JsonSink` from protobuf.
pub fn try_from_proto(
node: &datafusion_proto_models::protobuf::PhysicalPlanNode,
ctx: &datafusion_physical_plan::proto::ExecutionPlanDecodeCtx<'_>,
) -> Result<Arc<dyn ExecutionPlan>> {
use datafusion_datasource::file_sink_config::parse_sink_sort_order;
use datafusion_proto_models::protobuf;

let sink_node = match &node.physical_plan_type {
Some(protobuf::physical_plan_node::PhysicalPlanType::JsonSink(sink)) => {
sink.as_ref()
}
_ => {
return datafusion_common::internal_err!(
"PhysicalPlanNode is not a JsonSink"
);
}
};
let input = ctx.decode_required_child(
sink_node.input.as_deref(),
"JsonSinkExecNode",
"input",
)?;
let proto_sink = sink_node.sink.as_ref().ok_or_else(|| {
datafusion_common::internal_datafusion_err!(
"JsonSinkExecNode is missing required field 'sink'"
)
})?;
let config =
FileSinkConfig::from_proto(proto_sink.config.as_ref().ok_or_else(|| {
datafusion_common::internal_datafusion_err!(
"JsonSink is missing required field 'config'"
)
})?)?;
let writer_options = proto_sink
.writer_options
.as_ref()
.ok_or_else(|| {
datafusion_common::internal_datafusion_err!(
"JsonSink is missing required field 'writer_options'"
)
})?
.try_into()?;
let data_sink = JsonSink::new(config, writer_options);
let sort_order = parse_sink_sort_order(
sink_node.sort_order.as_ref(),
ctx,
input.schema().as_ref(),
)?;

Ok(Arc::new(DataSinkExec::new(
input,
Arc::new(data_sink),
sort_order,
)))
}
}

#[derive(Debug)]
Expand Down
6 changes: 6 additions & 0 deletions datafusion/datasource-parquet/Cargo.toml
Original file line number Diff line number Diff line change
Expand Up @@ -46,6 +46,7 @@ datafusion-physical-expr = { workspace = true }
datafusion-physical-expr-adapter = { workspace = true }
datafusion-physical-expr-common = { workspace = true }
datafusion-physical-plan = { workspace = true }
datafusion-proto-models = { workspace = true, optional = true }
datafusion-pruning = { workspace = true }
datafusion-session = { workspace = true }
futures = { workspace = true }
Expand Down Expand Up @@ -74,6 +75,11 @@ name = "datafusion_datasource_parquet"
path = "src/mod.rs"

[features]
proto = [
"dep:datafusion-proto-models",
"datafusion-datasource/proto",
"datafusion-physical-plan/proto",
]
parquet_encryption = [
"parquet/encryption",
"datafusion-common/parquet_encryption",
Expand Down
91 changes: 91 additions & 0 deletions datafusion/datasource-parquet/src/sink.rs
Original file line number Diff line number Diff line change
Expand Up @@ -33,13 +33,17 @@ use datafusion_datasource::display::FileGroupDisplay;
use datafusion_datasource::file_compression_type::FileCompressionType;
use datafusion_datasource::file_sink_config::{FileSink, FileSinkConfig};
use datafusion_datasource::sink::DataSink;
#[cfg(feature = "proto")]
use datafusion_datasource::sink::DataSinkExec;
use datafusion_datasource::write::demux::DemuxedStreamReceiver;
use datafusion_datasource::write::{
ObjectWriterBuilder, SharedBuffer, get_writer_schema,
};
use datafusion_execution::memory_pool::{MemoryConsumer, MemoryPool, MemoryReservation};
use datafusion_execution::runtime_env::RuntimeEnv;
use datafusion_execution::{SendableRecordBatchStream, TaskContext};
#[cfg(feature = "proto")]
use datafusion_physical_plan::ExecutionPlan;
use datafusion_physical_plan::metrics::{
ElapsedComputeFutureExt, ExecutionPlanMetricsSet, MetricBuilder, MetricCategory,
MetricsSet, Time,
Expand Down Expand Up @@ -411,6 +415,93 @@ impl DataSink for ParquetSink {
) -> Result<u64> {
FileSink::write_all(self, data, context).await
}

#[cfg(feature = "proto")]
fn try_to_proto(
&self,
input: datafusion_proto_models::protobuf::PhysicalPlanNode,
sort_order: Option<
datafusion_proto_models::protobuf::PhysicalSortExprNodeCollection,
>,
sink_schema: &Schema,
) -> Result<Option<datafusion_proto_models::protobuf::PhysicalPlanNode>> {
use datafusion_proto_models::protobuf;
use protobuf::physical_plan_node::PhysicalPlanType;

let sink = protobuf::ParquetSink {
config: Some(self.config.to_proto()?),
parquet_options: Some(self.parquet_options().try_into()?),
};
let node = protobuf::ParquetSinkExecNode {
input: Some(Box::new(input)),
sink: Some(sink),
sink_schema: Some(sink_schema.try_into()?),
sort_order,
};
Ok(Some(protobuf::PhysicalPlanNode {
physical_plan_type: Some(PhysicalPlanType::ParquetSink(Box::new(node))),
}))
}
}

#[cfg(feature = "proto")]
impl ParquetSink {
/// Reconstructs a [`DataSinkExec`] containing a `ParquetSink` from protobuf.
pub fn try_from_proto(
node: &datafusion_proto_models::protobuf::PhysicalPlanNode,
ctx: &datafusion_physical_plan::proto::ExecutionPlanDecodeCtx<'_>,
) -> Result<Arc<dyn ExecutionPlan>> {
use datafusion_datasource::file_sink_config::parse_sink_sort_order;
use datafusion_proto_models::protobuf;

let sink_node = match &node.physical_plan_type {
Some(protobuf::physical_plan_node::PhysicalPlanType::ParquetSink(sink)) => {
sink.as_ref()
}
_ => {
return datafusion_common::internal_err!(
"PhysicalPlanNode is not a ParquetSink"
);
}
};
let input = ctx.decode_required_child(
sink_node.input.as_deref(),
"ParquetSinkExecNode",
"input",
)?;
let proto_sink = sink_node.sink.as_ref().ok_or_else(|| {
datafusion_common::internal_datafusion_err!(
"ParquetSinkExecNode is missing required field 'sink'"
)
})?;
let config =
FileSinkConfig::from_proto(proto_sink.config.as_ref().ok_or_else(|| {
datafusion_common::internal_datafusion_err!(
"ParquetSink is missing required field 'config'"
)
})?)?;
let parquet_options = proto_sink
.parquet_options
.as_ref()
.ok_or_else(|| {
datafusion_common::internal_datafusion_err!(
"ParquetSink is missing required field 'parquet_options'"
)
})?
.try_into()?;
let data_sink = ParquetSink::new(config, parquet_options);
let sort_order = parse_sink_sort_order(
sink_node.sort_order.as_ref(),
ctx,
input.schema().as_ref(),
)?;

Ok(Arc::new(DataSinkExec::new(
input,
Arc::new(data_sink),
sort_order,
)))
}
}

/// Consumes a stream of [ArrowLeafColumn] via a channel and serializes them using an [ArrowColumnWriter]
Expand Down
Loading
Loading