This is an automated email from the ASF dual-hosted git repository.
alamb pushed a commit to branch 57_maintenance
in repository https://gitbox.apache.org/repos/asf/arrow-rs.git
The following commit(s) were added to refs/heads/57_maintenance by this push:
new f42d0c42ad [57_maintenance] fix(arrow-schema): validate `UnionFields`
when deserializing (#11813)
f42d0c42ad is described below
commit f42d0c42add18f99bd96bb8bb1cba16c8c446f5a
Author: Jeffrey Vo <[email protected]>
AuthorDate: Fri Oct 23 00:23:26 2026 +0900
[57_maintenance] fix(arrow-schema): validate `UnionFields` when
deserializing (#11813)
- Backport https://github.com/apache/arrow-rs/pull/11638
- For inclusion in https://github.com/apache/arrow-rs/issues/11666
Cherry-pick of
https://github.com/apache/arrow-rs/commit/a2b405fe22b6657fd4f69eb139e6836cafa3b55f
Co-authored-by: ump45nose <[email protected]>
Co-authored-by: ump45nose <[email protected]>
---
Cargo.lock | 34 +++++++++++++++++++++++++++++++
arrow-schema/Cargo.toml | 2 ++
arrow-schema/src/fields.rs | 51 +++++++++++++++++++++++++++++++++++++++++++++-
3 files changed, 86 insertions(+), 1 deletion(-)
diff --git a/Cargo.lock b/Cargo.lock
index c7967e9d3c..eec34b17b7 100644
--- a/Cargo.lock
+++ b/Cargo.lock
@@ -479,6 +479,7 @@ dependencies = [
"bitflags",
"criterion",
"insta",
+ "postcard",
"serde",
"serde_core",
"serde_json",
@@ -922,6 +923,15 @@ dependencies = [
"cc",
]
+[[package]]
+name = "cobs"
+version = "0.3.0"
+source = "registry+https://github.com/rust-lang/crates.io-index";
+checksum = "0fa961b519f0b462e3a3b4a34b64d119eeaca1d59af726fe450bbba07a9fc0a1"
+dependencies = [
+ "thiserror 2.0.21",
+]
+
[[package]]
name = "colorchoice"
version = "1.0.5"
@@ -1172,6 +1182,18 @@ version = "1.19.0"
source = "registry+https://github.com/rust-lang/crates.io-index";
checksum = "0e9c71c2167ca323c882b99918929403426e2373ea17242ff5653e0d5e1058be"
+[[package]]
+name = "embedded-io"
+version = "0.4.0"
+source = "registry+https://github.com/rust-lang/crates.io-index";
+checksum = "ef1a6892d9eef45c8fa6b9e0086428a2cca8491aca8f787c534a3d6d0bcb3ced"
+
+[[package]]
+name = "embedded-io"
+version = "0.6.1"
+source = "registry+https://github.com/rust-lang/crates.io-index";
+checksum = "edd0f118536f44f5ccd48bcb8b111bdc3de888b58c74639dfb034a357d0f206d"
+
[[package]]
name = "encode_unicode"
version = "1.0.0"
@@ -2468,6 +2490,18 @@ version = "1.15.0"
source = "registry+https://github.com/rust-lang/crates.io-index";
checksum = "05c8b63e8d9609db387f0324918f81d68fe27748f084ef092fb35954d0539a85"
+[[package]]
+name = "postcard"
+version = "1.1.3"
+source = "registry+https://github.com/rust-lang/crates.io-index";
+checksum = "6764c3b5dd454e283a30e6dfe78e9b31096d9e32036b5d1eaac7a6119ccb9a24"
+dependencies = [
+ "cobs",
+ "embedded-io 0.4.0",
+ "embedded-io 0.6.1",
+ "serde",
+]
+
[[package]]
name = "potential_utf"
version = "0.1.6"
diff --git a/arrow-schema/Cargo.toml b/arrow-schema/Cargo.toml
index 8164652843..4e307b3992 100644
--- a/arrow-schema/Cargo.toml
+++ b/arrow-schema/Cargo.toml
@@ -58,6 +58,8 @@ bincode = { version = "2.0.1", default-features = false,
features = [
] }
criterion = { workspace = true, default-features = false }
insta = "1.43.1"
+postcard = { version = "1.0.10", default-features = false, features =
["use-std"] }
+serde_json = "1.0"
[[bench]]
name = "ffi"
diff --git a/arrow-schema/src/fields.rs b/arrow-schema/src/fields.rs
index 3b33d161b7..1dad368990 100644
--- a/arrow-schema/src/fields.rs
+++ b/arrow-schema/src/fields.rs
@@ -324,10 +324,23 @@ impl<'a> IntoIterator for &'a Fields {
/// - All type ids are unique
/// - All type ids are non-negative
#[derive(Clone, Eq, PartialEq, Ord, PartialOrd, Hash)]
-#[cfg_attr(feature = "serde", derive(serde::Serialize, serde::Deserialize))]
+#[cfg_attr(feature = "serde", derive(serde::Serialize))]
#[cfg_attr(feature = "serde", serde(transparent))]
pub struct UnionFields(Arc<[(i8, FieldRef)]>);
+// Manual implementation to enforce invariants per docstring above
+#[cfg(feature = "serde")]
+impl<'de> serde::Deserialize<'de> for UnionFields {
+ fn deserialize<D>(deserializer: D) -> Result<Self, D::Error>
+ where
+ D: serde::Deserializer<'de>,
+ {
+ let entries = Vec::<(i8, FieldRef)>::deserialize(deserializer)?;
+ let (type_ids, fields): (Vec<i8>, Vec<FieldRef>) =
entries.into_iter().unzip();
+ Self::try_new(type_ids, fields).map_err(serde::de::Error::custom)
+ }
+}
+
impl std::fmt::Debug for UnionFields {
fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
self.0.as_ref().fmt(f)
@@ -1069,4 +1082,40 @@ mod tests {
];
UnionFields::from_iter(fields);
}
+
+ #[cfg(feature = "serde")]
+ #[test]
+ fn test_union_fields_serde_round_trip() {
+ let fields = UnionFields::try_new(
+ vec![1, 3],
+ vec![
+ Field::new("a", DataType::Int32, true),
+ Field::new("b", DataType::Utf8, true),
+ ],
+ )
+ .unwrap();
+ let serialized = postcard::to_stdvec(&fields).unwrap();
+ let deserialized: UnionFields =
postcard::from_bytes(&serialized).unwrap();
+ assert_eq!(fields, deserialized);
+ }
+
+ #[cfg(feature = "serde")]
+ #[test]
+ fn test_union_fields_deserialize_should_validate() {
+ let field = Arc::new(Field::new("a", DataType::Int32, true));
+
+ let duplicate: Vec<(i8, FieldRef)> = vec![(1, Arc::clone(&field)), (1,
Arc::clone(&field))];
+ let serialized = serde_json::to_string(&duplicate).unwrap();
+ let err =
serde_json::from_str::<UnionFields>(&serialized).unwrap_err();
+ assert!(err.to_string().contains("duplicate type id: 1"), "{err}");
+
+ let negative: Vec<(i8, FieldRef)> = vec![(0, Arc::clone(&field)), (-1,
field)];
+ let serialized = serde_json::to_string(&negative).unwrap();
+ let err =
serde_json::from_str::<UnionFields>(&serialized).unwrap_err();
+ assert!(
+ err.to_string()
+ .contains("type ids must be non-negative: -1"),
+ "{err}"
+ );
+ }
}
