Skip to content

refactor(proto): migrate CsvSource serde #24177

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

Merged
adriangb merged 1 commit into from
Aug 8, 2026
Merged

refactor(proto): migrate CsvSource serde #24177

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
1 change: 1 addition & 0 deletions datafusion/datasource-csv/Cargo.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -31,6 +31,7 @@ version.workspace = true
all-features = true

[features]
# Enables protobuf serialization hooks for CSV sources and sinks.
proto = [
"dep:datafusion-proto-models",
"datafusion-datasource/proto",
Expand Down
137 changes: 137 additions & 0 deletions datafusion/datasource-csv/src/source.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -308,6 +308,49 @@ impl FileSource for CsvSource {
DisplayFormatType::TreeRender => Ok(()),
}
}

/// Emit a `CsvScan` node wrapping the shared base config and CSV options.
#[cfg(feature = "proto")]
fn try_to_proto(
&self,
base: &FileScanConfig,
ctx: &datafusion_physical_plan::proto::ExecutionPlanEncodeCtx<'_>,
) -> Result<Option<datafusion_proto_models::protobuf::PhysicalPlanNode>> {
use datafusion_proto_models::protobuf;
use protobuf::physical_plan_node::PhysicalPlanType;

let node = protobuf::CsvScanExecNode {
base_conf: Some(base.try_to_proto(ctx)?),
has_header: self.has_header(),
delimiter: proto_byte_to_string(self.delimiter(), "delimiter")?,
quote: proto_byte_to_string(self.quote(), "quote")?,
optional_escape: self
.escape()
.map(|escape| {
Ok::<_, DataFusionError>(
protobuf::csv_scan_exec_node::OptionalEscape::Escape(
proto_byte_to_string(escape, "escape")?,
),
)
})
.transpose()?,
optional_comment: self
.comment()
.map(|comment| {
Ok::<_, DataFusionError>(
protobuf::csv_scan_exec_node::OptionalComment::Comment(
proto_byte_to_string(comment, "comment")?,
),
)
})
.transpose()?,
newlines_in_values: self.newlines_in_values(),
truncate_rows: self.truncate_rows(),
};
Ok(Some(protobuf::PhysicalPlanNode {
physical_plan_type: Some(PhysicalPlanType::CsvScan(node)),
}))
}
}

impl FileOpener for CsvOpener {
Expand Down Expand Up @@ -501,3 +544,97 @@ pub async fn plan_to_csv(

Ok(())
}

#[cfg(feature = "proto")]
fn proto_byte_to_string(b: u8, description: &str) -> Result<String> {
let bytes = &[b];
let s = std::str::from_utf8(bytes).map_err(|_| {
datafusion_common::internal_datafusion_err!(
"Invalid CSV {description}: can not represent {bytes:0x?} as utf8"
)
})?;
Ok(s.to_owned())
}

#[cfg(feature = "proto")]
fn proto_str_to_byte(s: &str, description: &str) -> Result<u8> {
datafusion_common::assert_eq_or_internal_err!(
s.len(),
1,
"Invalid CSV {description}: expected single character, got {s}"
);
Ok(s.as_bytes()[0])
}

#[cfg(feature = "proto")]
impl CsvSource {
/// Reconstructs a `DataSourceExec` from a protobuf `CsvScan`.
///
/// Custom line terminators are not represented in the wire format.
pub fn try_from_proto(
node: &datafusion_proto_models::protobuf::PhysicalPlanNode,
ctx: &datafusion_physical_plan::proto::ExecutionPlanDecodeCtx<'_>,
) -> Result<Arc<dyn ExecutionPlan>> {
use datafusion_common::config::CsvOptions;
use datafusion_datasource::file_compression_type::FileCompressionType;
use datafusion_datasource::file_scan_config::{
FileScanConfig, FileScanConfigBuilder,
};
use datafusion_datasource::source::DataSourceExec;
use datafusion_proto_models::protobuf;

let scan = match &node.physical_plan_type {
Some(protobuf::physical_plan_node::PhysicalPlanType::CsvScan(scan)) => scan,
_ => {
return datafusion_common::internal_err!(
"PhysicalPlanNode is not a CsvScan"
);
}
};

let base_conf = scan.base_conf.as_ref().ok_or_else(|| {
datafusion_common::internal_datafusion_err!(
"CsvScanExecNode is missing required field 'base_conf'"
)
})?;

let escape = match &scan.optional_escape {
Some(protobuf::csv_scan_exec_node::OptionalEscape::Escape(escape)) => {
Some(proto_str_to_byte(escape, "escape")?)
}
None => None,
};
let comment = match &scan.optional_comment {
Some(protobuf::csv_scan_exec_node::OptionalComment::Comment(comment)) => {
Some(proto_str_to_byte(comment, "comment")?)
}
None => None,
};

let table_schema = FileScanConfig::parse_table_schema_from_proto(base_conf)?;

let csv_options = CsvOptions {
has_header: Some(scan.has_header),
delimiter: proto_str_to_byte(&scan.delimiter, "delimiter")?,
quote: proto_str_to_byte(&scan.quote, "quote")?,
newlines_in_values: Some(scan.newlines_in_values),
truncated_rows: Some(scan.truncate_rows),
..Default::default()
};
let source = Arc::new(
CsvSource::new(table_schema)
.with_csv_options(csv_options)
.with_escape(escape)
.with_comment(comment),
);

// The compression type is not on the wire; CSV scans always
// deserialize as uncompressed.
let conf = FileScanConfigBuilder::from(FileScanConfig::try_from_proto(
base_conf, ctx, source,
)?)
.with_file_compression_type(FileCompressionType::UNCOMPRESSED)
.build();
Ok(DataSourceExec::from_data_source(conf))
}
}
21 changes: 0 additions & 21 deletions datafusion/proto/src/common.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -15,27 +15,6 @@
// specific language governing permissions and limitations
// under the License.

use datafusion_common::{Result, assert_eq_or_internal_err, internal_datafusion_err};

pub(crate) fn str_to_byte(s: &String, description: &str) -> Result<u8> {
assert_eq_or_internal_err!(
s.len(),
1,
"Invalid CSV {description}: expected single character, got {s}"
);
Ok(s.as_bytes()[0])
}

pub(crate) fn byte_to_string(b: u8, description: &str) -> Result<String> {
let b = &[b];
let b = std::str::from_utf8(b).map_err(|_| {
internal_datafusion_err!(
"Invalid CSV {description}: can not represent {b:0x?} as utf8"
)
})?;
Ok(b.to_owned())
}

#[macro_export]
macro_rules! convert_required {
($PB:expr) => {{
Expand Down
105 changes: 15 additions & 90 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 @@ -23,14 +23,12 @@ use std::sync::Arc;

use arrow::datatypes::{IntervalMonthDayNanoType, Schema, SchemaRef};
use datafusion_catalog::memory::MemorySourceConfig;
use datafusion_common::config::CsvOptions;
use datafusion_common::{
DataFusionError, Result, internal_datafusion_err, internal_err, not_impl_err,
};
#[cfg(feature = "parquet")]
use datafusion_datasource::file::FileSource;
use datafusion_datasource::file_compression_type::FileCompressionType;
use datafusion_datasource::file_scan_config::{FileScanConfig, FileScanConfigBuilder};
use datafusion_datasource::file_scan_config::FileScanConfig;
use datafusion_datasource::sink::DataSinkExec;
use datafusion_datasource::source::{DataSource, DataSourceExec};
use datafusion_datasource_arrow::source::ArrowSource;
Expand Down Expand Up @@ -94,7 +92,6 @@ use datafusion_physical_plan::{ExecutionPlan, PhysicalExpr};
use prost::Message;
use prost::bytes::BufMut;

use crate::common::{byte_to_string, str_to_byte};
use crate::convert_required;
use crate::physical_plan::from_proto::{
parse_physical_expr_with_converter, parse_physical_sort_exprs,
Expand Down Expand Up @@ -126,6 +123,7 @@ mod file_scan_config_serde {
use datafusion_common::{Constraint, Constraints, ScalarValue, Statistics};
use datafusion_datasource::file::FileSource;
use datafusion_datasource::file_groups::FileGroup;
use datafusion_datasource::file_scan_config::FileScanConfigBuilder;
use datafusion_datasource::file_stream::FileOpener;
use datafusion_datasource::{PartitionedFile, TableSchema};
use datafusion_execution::object_store::ObjectStoreUrl;
Expand Down Expand Up @@ -1084,8 +1082,8 @@ pub trait PhysicalPlanNodeExt: Sized {
PhysicalPlanType::Filter(_) => {
FilterExec::try_from_proto(self.node(), &decode_ctx)
}
PhysicalPlanType::CsvScan(scan) => {
self.try_into_csv_scan_physical_plan(scan, ctx, proto_converter)
PhysicalPlanType::CsvScan(_) => {
CsvSource::try_from_proto(self.node(), &decode_ctx)
}
PhysicalPlanType::JsonScan(scan) => {
self.try_into_json_scan_physical_plan(scan, ctx, proto_converter)
Expand Down Expand Up @@ -1340,57 +1338,25 @@ pub trait PhysicalPlanNodeExt: Sized {
FilterExec::try_from_proto(&node, &decode_ctx)
}

#[deprecated(
since = "55.0.0",
note = "unused by DataFusion; `CsvSource` deserializes itself via `CsvSource::try_from_proto`"
)]
fn try_into_csv_scan_physical_plan(
&self,
scan: &protobuf::CsvScanExecNode,
ctx: &PhysicalPlanDecodeContext<'_>,
proto_converter: &dyn PhysicalProtoConverterExtension,
) -> Result<Arc<dyn ExecutionPlan>> {
let escape =
if let Some(protobuf::csv_scan_exec_node::OptionalEscape::Escape(escape)) =
&scan.optional_escape
{
Some(str_to_byte(escape, "escape")?)
} else {
None
};

let comment = if let Some(
protobuf::csv_scan_exec_node::OptionalComment::Comment(comment),
) = &scan.optional_comment
{
Some(str_to_byte(comment, "comment")?)
} else {
None
};

// Parse table schema with partition columns
let table_schema =
parse_table_schema_from_proto(scan.base_conf.as_ref().unwrap())?;

let csv_options = CsvOptions {
has_header: Some(scan.has_header),
delimiter: str_to_byte(&scan.delimiter, "delimiter")?,
quote: str_to_byte(&scan.quote, "quote")?,
newlines_in_values: Some(scan.newlines_in_values),
..Default::default()
};
let source = Arc::new(
CsvSource::new(table_schema)
.with_csv_options(csv_options)
.with_escape(escape)
.with_comment(comment),
);

let conf = FileScanConfigBuilder::from(parse_protobuf_file_scan_config(
scan.base_conf.as_ref().unwrap(),
let node = protobuf::PhysicalPlanNode {
physical_plan_type: Some(PhysicalPlanType::CsvScan(scan.clone())),
};
let decoder = ConverterPlanDecoder {
ctx,
proto_converter,
source,
)?)
.with_file_compression_type(FileCompressionType::UNCOMPRESSED)
.build();
Ok(DataSourceExec::from_data_source(conf))
};
let decode_ctx = ExecutionPlanDecodeCtx::new(&decoder);
CsvSource::try_from_proto(&node, &decode_ctx)
}

fn try_into_json_scan_physical_plan(
Expand Down Expand Up @@ -2561,47 +2527,6 @@ pub trait PhysicalPlanNodeExt: Sized {
proto_converter: &dyn PhysicalProtoConverterExtension,
) -> Result<Option<protobuf::PhysicalPlanNode>> {
let data_source = data_source_exec.data_source();
if let Some(maybe_csv) = data_source.downcast_ref::<FileScanConfig>() {
let source = maybe_csv.file_source();
if let Some(csv_config) = source.downcast_ref::<CsvSource>() {
return Ok(Some(protobuf::PhysicalPlanNode {
physical_plan_type: Some(PhysicalPlanType::CsvScan(
protobuf::CsvScanExecNode {
base_conf: Some(serialize_file_scan_config(
maybe_csv,
codec,
proto_converter,
)?),
has_header: csv_config.has_header(),
delimiter: byte_to_string(
csv_config.delimiter(),
"delimiter",
)?,
quote: byte_to_string(csv_config.quote(), "quote")?,
optional_escape: if let Some(escape) = csv_config.escape() {
Some(
protobuf::csv_scan_exec_node::OptionalEscape::Escape(
byte_to_string(escape, "escape")?,
),
)
} else {
None
},
optional_comment: if let Some(comment) = csv_config.comment()
{
Some(protobuf::csv_scan_exec_node::OptionalComment::Comment(
byte_to_string(comment, "comment")?,
))
} else {
None
},
newlines_in_values: csv_config.newlines_in_values(),
truncate_rows: csv_config.truncate_rows(),
},
)),
}));
}
}

if let Some(scan_conf) = data_source.downcast_ref::<FileScanConfig>() {
let source = scan_conf.file_source();
Expand Down
Loading
Loading