apache · Jefffrey · Dec 31, 2025 · Dec 31, 2025 · Dec 31, 2025 · Dec 31, 2025
diff --git a/Cargo.toml b/Cargo.toml
@@ -201,6 +201,8 @@ uninlined_format_args = "warn"
 inefficient_to_string = "warn"
 # https://github.com/apache/datafusion/issues/18503
 needless_pass_by_value = "warn"
+# https://github.com/apache/datafusion/issues/18881
+allow_attributes = "warn"
 
 [workspace.lints.rust]
 unexpected_cfgs = { level = "warn", check-cfg = [

diff --git a/datafusion-examples/examples/custom_data_source/file_stream_provider.rs b/datafusion-examples/examples/custom_data_source/file_stream_provider.rs
@@ -22,7 +22,7 @@
 ///
 /// On non-Windows systems, this example creates a named pipe (FIFO) and
 /// writes rows into it asynchronously while DataFusion reads the data
-/// through a `FileStreamProvider`.  
+/// through a `FileStreamProvider`.
 ///
 /// This illustrates how to integrate dynamically updated data sources
 /// with DataFusion without needing to reload the entire dataset each time.
@@ -126,7 +126,6 @@ mod non_windows {
         let broken_pipe_timeout = Duration::from_secs(10);
         let sa = file_path;
         // Spawn a new thread to write to the FIFO file
-        #[allow(clippy::disallowed_methods)] // spawn allowed only in tests
         tasks.spawn_blocking(move || {
             let file = OpenOptions::new().write(true).open(sa).unwrap();
             // Reference time to use when deciding to fail the test

diff --git a/datafusion-examples/examples/data_io/catalog.rs b/datafusion-examples/examples/data_io/catalog.rs
@@ -140,7 +140,6 @@ struct DirSchemaOpts<'a> {
 /// Schema where every file with extension `ext` in a given `dir` is a table.
 #[derive(Debug)]
 struct DirSchema {
-    ext: String,
     tables: RwLock<HashMap<String, Arc<dyn TableProvider>>>,
 }
 
@@ -173,14 +172,8 @@ impl DirSchema {
         }
         Ok(Arc::new(Self {
             tables: RwLock::new(tables),
-            ext: ext.to_string(),
         }))
     }
-
-    #[allow(unused)]
-    fn name(&self) -> &str {
-        &self.ext
-    }
 }
 
 #[async_trait]
@@ -217,7 +210,6 @@ impl SchemaProvider for DirSchema {
 
     /// If supported by the implementation, removes an existing table from this schema and returns it.
     /// If no table of that name exists, returns Ok(None).
-    #[allow(unused_variables)]
     fn deregister_table(&self, name: &str) -> Result<Option<Arc<dyn TableProvider>>> {
         let mut tables = self.tables.write().unwrap();
         log::info!("dropping table {name}");

diff --git a/datafusion-examples/examples/flight/sql_server.rs b/datafusion-examples/examples/flight/sql_server.rs
@@ -115,7 +115,6 @@ impl FlightSqlServiceImpl {
         Ok(uuid)
     }
 
-    #[allow(clippy::result_large_err)]
     fn get_ctx<T>(&self, req: &Request<T>) -> Result<Arc<SessionContext>, Status> {
         // get the token from the authorization header on Request
         let auth = req
@@ -141,7 +140,6 @@ impl FlightSqlServiceImpl {
         }
     }
 
-    #[allow(clippy::result_large_err)]
     fn get_plan(&self, handle: &str) -> Result<LogicalPlan, Status> {
         if let Some(plan) = self.statements.get(handle) {
             Ok(plan.clone())
@@ -150,7 +148,6 @@ impl FlightSqlServiceImpl {
         }
     }
 
-    #[allow(clippy::result_large_err)]
     fn get_result(&self, handle: &str) -> Result<Vec<RecordBatch>, Status> {
         if let Some(result) = self.results.get(handle) {
             Ok(result.clone())
@@ -198,13 +195,11 @@ impl FlightSqlServiceImpl {
         .unwrap()
     }
 
-    #[allow(clippy::result_large_err)]
     fn remove_plan(&self, handle: &str) -> Result<(), Status> {
         self.statements.remove(&handle.to_string());
         Ok(())
     }
 
-    #[allow(clippy::result_large_err)]
     fn remove_result(&self, handle: &str) -> Result<(), Status> {
         self.results.remove(&handle.to_string());
         Ok(())

diff --git a/datafusion/catalog-listing/src/mod.rs b/datafusion/catalog-listing/src/mod.rs
@@ -15,7 +15,6 @@
 // specific language governing permissions and limitations
 // under the License.
 
-#![deny(clippy::allow_attributes)]
 #![cfg_attr(test, allow(clippy::needless_pass_by_value))]
 #![doc(
     html_logo_url = "https://raw.githubusercontent.com/apache/datafusion/19fe44cf2f30cbdd63d4a4f52c74055163c6cc38/docs/logos/standalone_logo/logo_original.svg",

diff --git a/datafusion/catalog/src/lib.rs b/datafusion/catalog/src/lib.rs
@@ -24,7 +24,6 @@
 // https://github.com/apache/datafusion/issues/11143
 #![cfg_attr(not(test), deny(clippy::clone_on_ref_ptr))]
 #![cfg_attr(test, allow(clippy::needless_pass_by_value))]
-#![deny(clippy::allow_attributes)]
 
 //! Interfaces and default implementations of catalogs and schemas.
 //!

diff --git a/datafusion/common-runtime/src/lib.rs b/datafusion/common-runtime/src/lib.rs
@@ -16,7 +16,6 @@
 // under the License.
 
 #![cfg_attr(test, allow(clippy::needless_pass_by_value))]
-#![deny(clippy::allow_attributes)]
 #![doc(
     html_logo_url = "https://raw.githubusercontent.com/apache/datafusion/19fe44cf2f30cbdd63d4a4f52c74055163c6cc38/docs/logos/standalone_logo/logo_original.svg",
     html_favicon_url = "https://raw.githubusercontent.com/apache/datafusion/19fe44cf2f30cbdd63d4a4f52c74055163c6cc38/docs/logos/standalone_logo/logo_original.svg"

diff --git a/datafusion/common/src/lib.rs b/datafusion/common/src/lib.rs
@@ -24,7 +24,6 @@
 // https://github.com/apache/datafusion/issues/11143
 #![deny(clippy::clone_on_ref_ptr)]
 #![cfg_attr(test, allow(clippy::needless_pass_by_value))]
-#![deny(clippy::allow_attributes)]
 
 mod column;
 mod dfschema;

diff --git a/datafusion/core/benches/data_utils/mod.rs b/datafusion/core/benches/data_utils/mod.rs
@@ -36,6 +36,7 @@ use std::sync::Arc;
 
 /// create an in-memory table given the partition len, array len, and batch size,
 /// and the result table will be of array_len in total, and then partitioned, and batched.
+#[expect(clippy::allow_attributes)] // some issue where expect(dead_code) doesn't fire properly
 #[allow(dead_code)]
 pub fn create_table_provider(
     partitions_len: usize,
@@ -183,6 +184,7 @@ impl TraceIdBuilder {
 
 /// Create time series data with `partition_cnt` partitions and `sample_cnt` rows per partition
 /// in ascending order, if `asc` is true, otherwise randomly sampled using a Pareto distribution
+#[expect(clippy::allow_attributes)] // some issue where expect(dead_code) doesn't fire properly
 #[allow(dead_code)]
 pub(crate) fn make_data(
     partition_cnt: i32,

diff --git a/datafusion/core/benches/preserve_file_partitioning.rs b/datafusion/core/benches/preserve_file_partitioning.rs
@@ -322,7 +322,7 @@ async fn save_plans(
     }
 }
 
-#[allow(clippy::too_many_arguments)]
+#[expect(clippy::too_many_arguments)]
 fn run_benchmark(
     c: &mut Criterion,
     rt: &Runtime,

diff --git a/datafusion/core/src/bin/print_functions_docs.rs b/datafusion/core/src/bin/print_functions_docs.rs
@@ -84,30 +84,6 @@ fn print_window_docs() -> Result<String> {
     print_docs(providers, window_doc_sections::doc_sections())
 }
 
-// Temporary method useful to semi automate
-// the migration of UDF documentation generation from code based
-// to attribute based
-// To be removed
-#[allow(dead_code)]
-fn save_doc_code_text(documentation: &Documentation, name: &str) {
-    let attr_text = documentation.to_doc_attribute();
-
-    let file_path = format!("{name}.txt");
-    if std::path::Path::new(&file_path).exists() {
-        std::fs::remove_file(&file_path).unwrap();
-    }
-
-    // Open the file in append mode, create it if it doesn't exist
-    let mut file = std::fs::OpenOptions::new()
-        .append(true) // Open in append mode
-        .create(true) // Create the file if it doesn't exist
-        .open(file_path)
-        .unwrap();
-
-    use std::io::Write;
-    file.write_all(attr_text.as_bytes()).unwrap();
-}
-
 #[expect(clippy::needless_pass_by_value)]
 fn print_docs(
     providers: Vec<Box<dyn DocProvider>>,
@@ -306,8 +282,7 @@ impl DocProvider for WindowUDF {
     }
 }
 
-#[allow(clippy::borrowed_box)]
-#[allow(clippy::ptr_arg)]
+#[expect(clippy::borrowed_box)]
 fn get_names_and_aliases(functions: &Vec<&Box<dyn DocProvider>>) -> Vec<String> {
     functions
         .iter()

diff --git a/datafusion/core/src/lib.rs b/datafusion/core/src/lib.rs
@@ -15,7 +15,6 @@
 // specific language governing permissions and limitations
 // under the License.
 
-#![deny(clippy::allow_attributes)]
 #![doc(
     html_logo_url = "https://raw.githubusercontent.com/apache/datafusion/19fe44cf2f30cbdd63d4a4f52c74055163c6cc38/docs/logos/standalone_logo/logo_original.svg",
     html_favicon_url = "https://raw.githubusercontent.com/apache/datafusion/19fe44cf2f30cbdd63d4a4f52c74055163c6cc38/docs/logos/standalone_logo/logo_original.svg"

diff --git a/datafusion/core/tests/custom_sources_cases/dml_planning.rs b/datafusion/core/tests/custom_sources_cases/dml_planning.rs
@@ -94,7 +94,7 @@ impl TableProvider for CaptureDeleteProvider {
 }
 
 /// A TableProvider that captures filters and assignments passed to update().
-#[allow(clippy::type_complexity)]
+#[expect(clippy::type_complexity)]
 struct CaptureUpdateProvider {
     schema: SchemaRef,
     received_filters: Arc<Mutex<Option<Vec<Expr>>>>,

diff --git a/datafusion/core/tests/execution/coop.rs b/datafusion/core/tests/execution/coop.rs
@@ -764,7 +764,7 @@ async fn hash_join_without_repartition_and_no_agg(
 #[derive(Debug)]
 enum Yielded {
     ReadyOrPending,
-    Err(#[allow(dead_code)] DataFusionError),
+    Err(#[expect(dead_code)] DataFusionError),
     Timeout,
 }
 

diff --git a/datafusion/core/tests/fifo/mod.rs b/datafusion/core/tests/fifo/mod.rs
@@ -94,7 +94,6 @@ mod unix_test {
     /// This function creates a writing task for the FIFO file. To verify
     /// incremental processing, it waits for a signal to continue writing after
     /// a certain number of lines are written.
-    #[allow(clippy::disallowed_methods)]
     fn create_writing_task(
         file_path: PathBuf,
         header: String,
@@ -105,6 +104,7 @@ mod unix_test {
         // Timeout for a long period of BrokenPipe error
         let broken_pipe_timeout = Duration::from_secs(10);
         // Spawn a new task to write to the FIFO file
+        #[expect(clippy::disallowed_methods)]
         tokio::spawn(async move {
             let mut file = tokio::fs::OpenOptions::new()
                 .write(true)
@@ -357,7 +357,7 @@ mod unix_test {
             (sink_fifo_path.clone(), sink_fifo_path.display());
 
         // Spawn a new thread to read sink EXTERNAL TABLE.
-        #[allow(clippy::disallowed_methods)] // spawn allowed only in tests
+        #[expect(clippy::disallowed_methods)] // spawn allowed only in tests
         tasks.push(spawn_blocking(move || {
             let file = File::open(sink_fifo_path_thread).unwrap();
             let schema = Arc::new(Schema::new(vec![

diff --git a/datafusion/core/tests/fuzz_cases/aggregation_fuzzer/context_generator.rs b/datafusion/core/tests/fuzz_cases/aggregation_fuzzer/context_generator.rs
@@ -214,7 +214,7 @@ impl GeneratedSessionContextBuilder {
 
 /// The generated params for [`SessionContext`]
 #[derive(Debug)]
-#[allow(dead_code)]
+#[expect(dead_code)]
 pub struct SessionContextParams {
     batch_size: usize,
     target_partitions: usize,

diff --git a/datafusion/core/tests/fuzz_cases/aggregation_fuzzer/query_builder.rs b/datafusion/core/tests/fuzz_cases/aggregation_fuzzer/query_builder.rs
@@ -182,13 +182,13 @@ impl QueryBuilder {
 
     /// Add max columns num in group by(default: 3), for example if it is set to 1,
     /// the generated sql will group by at most 1 column
-    #[allow(dead_code)]
+    #[expect(dead_code)]
     pub fn with_max_group_by_columns(mut self, max_group_by_columns: usize) -> Self {
         self.max_group_by_columns = max_group_by_columns;
         self
     }
 
-    #[allow(dead_code)]
+    #[expect(dead_code)]
     pub fn with_min_group_by_columns(mut self, min_group_by_columns: usize) -> Self {
         self.min_group_by_columns = min_group_by_columns;
         self
@@ -202,7 +202,7 @@ impl QueryBuilder {
     }
 
     /// Add if also test the no grouping aggregation case(default: true)
-    #[allow(dead_code)]
+    #[expect(dead_code)]
     pub fn with_no_grouping(mut self, no_grouping: bool) -> Self {
         self.no_grouping = no_grouping;
         self

diff --git a/datafusion/core/tests/fuzz_cases/join_fuzz.rs b/datafusion/core/tests/fuzz_cases/join_fuzz.rs
@@ -1086,7 +1086,7 @@ impl JoinFuzzTestCase {
     /// Files can be of different sizes
     /// The method can be useful to read partitions have been saved by `save_partitioned_batches_as_parquet`
     /// for test debugging purposes
-    #[allow(dead_code)]
+    #[expect(dead_code)]
     async fn load_partitioned_batches_from_parquet(
         dir: &str,
     ) -> std::io::Result<Vec<RecordBatch>> {

diff --git a/datafusion/core/tests/fuzz_cases/sort_preserving_repartition_fuzz.rs b/datafusion/core/tests/fuzz_cases/sort_preserving_repartition_fuzz.rs
@@ -301,7 +301,7 @@ mod sp_repartition_fuzz_tests {
                 let mut handles = Vec::new();
 
                 for seed in seed_start..seed_end {
-                    #[allow(clippy::disallowed_methods)] // spawn allowed only in tests
+                    #[expect(clippy::disallowed_methods)] // spawn allowed only in tests
                     let job = tokio::spawn(run_sort_preserving_repartition_test(
                         make_staggered_batches::<true>(n_row, n_distinct, seed as u64),
                         is_first_roundrobin,

diff --git a/datafusion/core/tests/macro_hygiene/mod.rs b/datafusion/core/tests/macro_hygiene/mod.rs
@@ -73,7 +73,7 @@ mod config_field {
     #[test]
     fn test_macro() {
         #[derive(Debug)]
-        #[allow(dead_code)]
+        #[expect(dead_code)]
         struct E;
 
         impl std::fmt::Display for E {
@@ -84,7 +84,7 @@ mod config_field {
 
         impl std::error::Error for E {}
 
-        #[allow(dead_code)]
+        #[expect(dead_code)]
         #[derive(Default)]
         struct S;
 

diff --git a/datafusion/core/tests/parquet/filter_pushdown.rs b/datafusion/core/tests/parquet/filter_pushdown.rs
@@ -220,7 +220,6 @@ async fn single_file() {
 }
 
 #[tokio::test]
-#[allow(dead_code)]
 async fn single_file_small_data_pages() {
     let batches = read_parquet_test_data(
         "tests/data/filter_pushdown/single_file_small_pages.gz.parquet",

diff --git a/datafusion/datasource-arrow/src/mod.rs b/datafusion/datasource-arrow/src/mod.rs
@@ -19,7 +19,6 @@
 // Make sure fast / cheap clones on Arc are explicit:
 // https://github.com/apache/datafusion/issues/11143
 #![cfg_attr(not(test), deny(clippy::clone_on_ref_ptr))]
-#![deny(clippy::allow_attributes)]
 
 //! [`ArrowFormat`]: Apache Arrow file format abstractions
 

diff --git a/datafusion/datasource-avro/src/mod.rs b/datafusion/datasource-avro/src/mod.rs
@@ -24,7 +24,6 @@
 // https://github.com/apache/datafusion/issues/11143
 #![cfg_attr(not(test), deny(clippy::clone_on_ref_ptr))]
 #![cfg_attr(test, allow(clippy::needless_pass_by_value))]
-#![deny(clippy::allow_attributes)]
 
 //! An [Avro](https://avro.apache.org/) based [`FileSource`](datafusion_datasource::file::FileSource) implementation and related functionality.
 

diff --git a/datafusion/datasource-csv/src/mod.rs b/datafusion/datasource-csv/src/mod.rs
@@ -19,7 +19,6 @@
 // Make sure fast / cheap clones on Arc are explicit:
 // https://github.com/apache/datafusion/issues/11143
 #![cfg_attr(not(test), deny(clippy::clone_on_ref_ptr))]
-#![deny(clippy::allow_attributes)]
 
 pub mod file_format;
 pub mod source;

diff --git a/datafusion/datasource-json/src/mod.rs b/datafusion/datasource-json/src/mod.rs
@@ -19,7 +19,6 @@
 // Make sure fast / cheap clones on Arc are explicit:
 // https://github.com/apache/datafusion/issues/11143
 #![cfg_attr(not(test), deny(clippy::clone_on_ref_ptr))]
-#![deny(clippy::allow_attributes)]
 
 pub mod file_format;
 pub mod source;

diff --git a/datafusion/datasource-parquet/src/mod.rs b/datafusion/datasource-parquet/src/mod.rs
@@ -19,7 +19,6 @@
 // https://github.com/apache/datafusion/issues/11143
 #![cfg_attr(not(test), deny(clippy::clone_on_ref_ptr))]
 #![cfg_attr(test, allow(clippy::needless_pass_by_value))]
-#![deny(clippy::allow_attributes)]
 
 pub mod access_plan;
 pub mod file_format;

diff --git a/datafusion/datasource/src/mod.rs b/datafusion/datasource/src/mod.rs
@@ -24,7 +24,6 @@
 // https://github.com/apache/datafusion/issues/11143
 #![cfg_attr(not(test), deny(clippy::clone_on_ref_ptr))]
 #![cfg_attr(test, allow(clippy::needless_pass_by_value))]
-#![deny(clippy::allow_attributes)]
 
 //! A table that uses the `ObjectStore` listing capability
 //! to get the list of files to process.

diff --git a/datafusion/doc/src/lib.rs b/datafusion/doc/src/lib.rs
@@ -15,7 +15,6 @@
 // specific language governing permissions and limitations
 // under the License.
 
-#![deny(clippy::allow_attributes)]
 #![cfg_attr(test, allow(clippy::needless_pass_by_value))]
 #![doc(
     html_logo_url = "https://raw.githubusercontent.com/apache/datafusion/19fe44cf2f30cbdd63d4a4f52c74055163c6cc38/docs/logos/standalone_logo/logo_original.svg",

diff --git a/datafusion/execution/src/lib.rs b/datafusion/execution/src/lib.rs
@@ -24,7 +24,6 @@
 // https://github.com/apache/datafusion/issues/11143
 #![deny(clippy::clone_on_ref_ptr)]
 #![cfg_attr(test, allow(clippy::needless_pass_by_value))]
-#![deny(clippy::allow_attributes)]
 
 //! DataFusion execution configuration and runtime structures
 

diff --git a/datafusion/expr-common/src/lib.rs b/datafusion/expr-common/src/lib.rs
@@ -32,7 +32,6 @@
 // https://github.com/apache/datafusion/issues/11143
 #![deny(clippy::clone_on_ref_ptr)]
 #![cfg_attr(test, allow(clippy::needless_pass_by_value))]
-#![deny(clippy::allow_attributes)]
 
 pub mod accumulator;
 pub mod casts;