Skip to content

Deprecate RecordBatchOptions::with_match_field_names #7406

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
wants to merge 4 commits into
base: main
Choose a base branch
from
Open

Deprecate RecordBatchOptions::with_match_field_names #7406

Show file tree
Hide file tree
Changes from 2 commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
4bb7b63
Deprecate RecordBatchOptions::with_match_field_names
tustvold Apr 11, 2025
ca4553d
Cleanup
tustvold Apr 11, 2025
0d56f16
Soften wording
tustvold Apr 11, 2025
7399f4e
Clippy
tustvold Apr 11, 2025
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
17 changes: 9 additions & 8 deletions arrow-array/src/record_batch.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -312,6 +312,7 @@ impl RecordBatch {

// function for comparing column type and field type
// return true if 2 types are not matched
#[allow(deprecated)]
let type_not_match = if options.match_field_names {
|(_, (col_type, field_type)): &(usize, (&DataType, &DataType))| col_type != field_type
} else {
Expand Down Expand Up @@ -400,10 +401,7 @@ impl RecordBatch {
RecordBatch::try_new_with_options(
SchemaRef::new(projected_schema),
batch_fields,
&RecordBatchOptions {
match_field_names: true,
row_count: Some(self.row_count),
},
&RecordBatchOptions::new().with_row_count(Some(self.row_count)),
)
}

Expand Down Expand Up @@ -742,6 +740,7 @@ impl RecordBatch {
#[non_exhaustive]
pub struct RecordBatchOptions {
/// Match field names of structs and lists. If set to `true`, the names must match.
#[deprecated(note = "match_field_names is unsound")]
Copy link
Contributor Author

Choose a reason for hiding this comment

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

This is marked non_exhaustive so the churn should be fairly minimal

Copy link
Contributor

Choose a reason for hiding this comment

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

if we mark it unsound, I think it would help to provide a link / reference with an explanation about why it is unsound

For example, it is not clear to me why mismatched names is unsound 🤔

pub match_field_names: bool,

/// Optional row count, useful for specifying a row count for a RecordBatch with no columns
Expand All @@ -750,6 +749,7 @@ pub struct RecordBatchOptions {

impl RecordBatchOptions {
/// Creates a new `RecordBatchOptions`
#[allow(deprecated)]
pub fn new() -> Self {
Self {
match_field_names: true,
Expand All @@ -764,6 +764,8 @@ impl RecordBatchOptions {
}

/// Sets the `match_field_names` of `RecordBatchOptions` and returns this [`RecordBatch`]
#[deprecated(note = "match_field_names is unsound")]
Copy link
Contributor

Choose a reason for hiding this comment

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

Should be there any reason it is unsound so the user can accept exact risks when using it?

Copy link
Contributor Author

@tustvold tustvold Apr 11, 2025 •
edited
Loading

Choose a reason for hiding this comment

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

Tbh I've not sat down and worked out a precise exploit chain, if I had this would flagged as a security vulnerability. However, it is breaking a pretty fundamental invariant that is assumed in a number of places. The worst it is probably going to do is cause something to panic, or produce invalid output, but the potential is there and I'd sleep happier not having it being used 😆

Copy link
Contributor

Choose a reason for hiding this comment

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

I had the same question above -- if we are going to claim something is unsound I think we should justify why and provide some hints for an alternative

#[allow(deprecated)]
pub fn with_match_field_names(mut self, match_field_names: bool) -> Self {
self.match_field_names = match_field_names;
self
Expand Down Expand Up @@ -1059,6 +1061,7 @@ mod tests {
}

#[test]
#[allow(deprecated)]
fn create_record_batch_field_name_mismatch() {
let fields = vec![
Field::new("a1", DataType::Int32, false),
Expand Down Expand Up @@ -1514,10 +1517,7 @@ mod tests {
let expected = RecordBatch::try_new_with_options(
Arc::new(Schema::empty()),
vec![],
&RecordBatchOptions {
match_field_names: true,
row_count: Some(3),
},
&RecordBatchOptions::new().with_row_count(Some(3)),
)
.expect("valid conversion");

Expand Down Expand Up @@ -1558,6 +1558,7 @@ mod tests {
assert_eq!("Invalid argument error: Column 'a' is declared as non-nullable but contains null values", format!("{}", maybe_batch.err().unwrap()));
}
#[test]
#[allow(deprecated)]
fn test_record_batch_options() {
let options = RecordBatchOptions::new()
.with_match_field_names(false)
Expand Down
4 changes: 1 addition & 3 deletions arrow-csv/src/reader/mod.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -866,9 +866,7 @@ fn parse(
RecordBatch::try_new_with_options(
projected_schema,
arr,
&RecordBatchOptions::new()
.with_match_field_names(true)
.with_row_count(Some(rows.len())),
&RecordBatchOptions::new().with_row_count(Some(rows.len())),
)
})
}
Expand Down
4 changes: 1 addition & 3 deletions arrow-ipc/src/reader.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2400,9 +2400,7 @@ mod tests {
#[test]
fn test_no_columns_batch() {
let schema = Arc::new(Schema::empty());
let options = RecordBatchOptions::new()
.with_match_field_names(true)
.with_row_count(Some(10));
let options = RecordBatchOptions::new().with_row_count(Some(10));
let input_batch = RecordBatch::try_new_with_options(schema, vec![], &options).unwrap();
let output_batch = roundtrip_ipc_stream(&input_batch);
assert_eq!(input_batch, output_batch);
Expand Down
6 changes: 1 addition & 5 deletions arrow/src/util/data_gen.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -46,11 +46,7 @@ pub fn create_random_batch(
.map(|field| create_random_array(field, size, null_density, true_density))
.collect::<Result<Vec<ArrayRef>>>()?;

RecordBatch::try_new_with_options(
schema,
columns,
&RecordBatchOptions::new().with_match_field_names(false),
Copy link
Contributor Author

Choose a reason for hiding this comment

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

I'm not sure why this was here, as the below code will always create the correct types AFAICT - perhaps a workaround for a since fixed bug?

)
RecordBatch::try_new(schema, columns)
}

/// Create a random [ArrayRef] from a [DataType] with a length,
Expand Down
2 changes: 1 addition & 1 deletion parquet/benches/arrow_writer.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -198,7 +198,7 @@ fn create_float_bench_batch_with_nans(size: usize, nan_density: f32) -> Result<R
Ok(RecordBatch::try_new_with_options(
Arc::new(schema),
columns,
&RecordBatchOptions::new().with_match_field_names(false),
&RecordBatchOptions::new(),
)?)
}

Expand Down
Loading