We have now the CI ensure all doc strings remain formatted

.github/workflows/ci.yml

@@ -0,0 +1,18 @@
+name: CI
+
+on:
+  push:
+    branches: [ main ]
+  pull_request:
+
+jobs:
+  fmt:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v3
+
+      - name: Install Rust nightly
+        run: rustup toolchain install nightly
+
+      - name: Run rustfmt (nightly)
+        run: cargo +nightly fmt --all -- --config format_code_in_doc_comments=true

datafusion-examples/examples/advanced_parquet_index.rs

@@ -121,7 +121,6 @@ use url::Url;
 ///         │ ╚═══════════════════╝ │      1. With cached ParquetMetadata, so
 ///         └───────────────────────┘      the ParquetSource does not re-read /
 ///          Parquet File                  decode the thrift footer
-///
 /// ```
 ///
 /// Within a Row Group, Column Chunks store data in DataPages. This example also

datafusion-examples/examples/external_dependency/query-aws-s3.rs

@@ -28,7 +28,6 @@ use url::Url;
 ///
 /// - AWS_ACCESS_KEY_ID
 /// - AWS_SECRET_ACCESS_KEY
-///
 #[tokio::main]
 async fn main() -> Result<()> {
     let ctx = SessionContext::new();

datafusion-examples/examples/flight/flight_sql_server.rs

@@ -68,7 +68,6 @@ macro_rules! status {
 ///
 /// Based heavily on Ballista's implementation: https://github.com/apache/datafusion-ballista/blob/main/ballista/scheduler/src/flight_sql.rs
 /// and the example in arrow-rs: https://github.com/apache/arrow-rs/blob/master/arrow-flight/examples/flight_sql_server.rs
-///
 #[tokio::main]
 async fn main() -> Result<(), Box<dyn std::error::Error>> {
     env_logger::init();

datafusion-examples/examples/parquet_index.rs

@@ -99,7 +99,6 @@ use url::Url;
 ///                   Thus some parquet files are      │             │
 ///                   "pruned" and thus are not        └─────────────┘
 ///                   scanned at all                   Parquet Files
-///
 /// ```
 ///
 /// [`ListingTable`]: datafusion::datasource::listing::ListingTable

datafusion-examples/examples/sql_query.rs

@@ -32,7 +32,6 @@ use std::sync::Arc;
 ///
 /// [`query_memtable`]: a simple query against a [`MemTable`]
 /// [`query_parquet`]: a simple query against a directory with multiple Parquet files
-///
 #[tokio::main]
 async fn main() -> Result<()> {
     query_memtable().await?;

datafusion-examples/examples/thread_pools.rs

@@ -342,7 +342,7 @@ impl CpuRuntime {
     /// message such as:
     ///
     /// ```text
-    ///A Tokio 1.x context was found, but IO is disabled.
+    /// A Tokio 1.x context was found, but IO is disabled.
     /// ```
     pub fn handle(&self) -> &Handle {
         &self.handle

datafusion/common/src/config.rs

@@ -55,7 +55,7 @@ use hex;
 ///        /// Field 3 doc
 ///        field3: Option<usize>, default = None
 ///    }
-///}
+/// }
 /// ```
 ///
 /// Will generate
@@ -1139,36 +1139,35 @@ impl ConfigOptions {
 /// # Example
 /// ```
 /// use datafusion_common::{
-///     config::ConfigExtension, extensions_options,
-///     config::ConfigOptions,
+///     config::ConfigExtension, config::ConfigOptions, extensions_options,
 /// };
-///  // Define a new configuration struct using the `extensions_options` macro
-///  extensions_options! {
-///     /// My own config options.
-///     pub struct MyConfig {
-///         /// Should "foo" be replaced by "bar"?
-///         pub foo_to_bar: bool, default = true
+/// // Define a new configuration struct using the `extensions_options` macro
+/// extensions_options! {
+///    /// My own config options.
+///    pub struct MyConfig {
+///        /// Should "foo" be replaced by "bar"?
+///        pub foo_to_bar: bool, default = true
 ///
-///         /// How many "baz" should be created?
-///         pub baz_count: usize, default = 1337
-///     }
-///  }
+///        /// How many "baz" should be created?
+///        pub baz_count: usize, default = 1337
+///    }
+/// }
 ///
-///  impl ConfigExtension for MyConfig {
+/// impl ConfigExtension for MyConfig {
 ///     const PREFIX: &'static str = "my_config";
-///  }
+/// }
 ///
-///  // set up config struct and register extension
-///  let mut config = ConfigOptions::default();
-///  config.extensions.insert(MyConfig::default());
+/// // set up config struct and register extension
+/// let mut config = ConfigOptions::default();
+/// config.extensions.insert(MyConfig::default());
 ///
-///  // overwrite config default
-///  config.set("my_config.baz_count", "42").unwrap();
+/// // overwrite config default
+/// config.set("my_config.baz_count", "42").unwrap();
 ///
-///  // check config state
-///  let my_config = config.extensions.get::<MyConfig>().unwrap();
-///  assert!(my_config.foo_to_bar,);
-///  assert_eq!(my_config.baz_count, 42,);
+/// // check config state
+/// let my_config = config.extensions.get::<MyConfig>().unwrap();
+/// assert!(my_config.foo_to_bar,);
+/// assert_eq!(my_config.baz_count, 42,);
 /// ```
 ///
 /// # Note:

datafusion/common/src/dfschema.rs

@@ -56,12 +56,10 @@ pub type DFSchemaRef = Arc<DFSchema>;
 /// an Arrow schema.
 ///
 /// ```rust
-/// use datafusion_common::{DFSchema, Column};
 /// use arrow::datatypes::{DataType, Field, Schema};
+/// use datafusion_common::{Column, DFSchema};
 ///
-/// let arrow_schema = Schema::new(vec![
-///    Field::new("c1", DataType::Int32, false),
-/// ]);
+/// let arrow_schema = Schema::new(vec![Field::new("c1", DataType::Int32, false)]);
 ///
 /// let df_schema = DFSchema::try_from_qualified_schema("t1", &arrow_schema).unwrap();
 /// let column = Column::from_qualified_name("t1.c1");
@@ -77,12 +75,10 @@ pub type DFSchemaRef = Arc<DFSchema>;
 /// Create an unqualified schema using TryFrom:
 ///
 /// ```rust
-/// use datafusion_common::{DFSchema, Column};
 /// use arrow::datatypes::{DataType, Field, Schema};
+/// use datafusion_common::{Column, DFSchema};
 ///
-/// let arrow_schema = Schema::new(vec![
-///    Field::new("c1", DataType::Int32, false),
-/// ]);
+/// let arrow_schema = Schema::new(vec![Field::new("c1", DataType::Int32, false)]);
 ///
 /// let df_schema = DFSchema::try_from(arrow_schema).unwrap();
 /// let column = Column::new_unqualified("c1");
@@ -94,13 +90,15 @@ pub type DFSchemaRef = Arc<DFSchema>;
 /// Use the `Into` trait to convert `DFSchema` into an Arrow schema:
 ///
 /// ```rust
+/// use arrow::datatypes::{Field, Schema};
 /// use datafusion_common::DFSchema;
-/// use arrow::datatypes::{Schema, Field};
 /// use std::collections::HashMap;
 ///
-/// let df_schema = DFSchema::from_unqualified_fields(vec![
-///    Field::new("c1", arrow::datatypes::DataType::Int32, false),
-/// ].into(),HashMap::new()).unwrap();
+/// let df_schema = DFSchema::from_unqualified_fields(
+///     vec![Field::new("c1", arrow::datatypes::DataType::Int32, false)].into(),
+///     HashMap::new(),
+/// )
+/// .unwrap();
 /// let schema = Schema::from(df_schema);
 /// assert_eq!(schema.fields().len(), 1);
 /// ```

datafusion/common/src/diagnostic.rs

@@ -30,8 +30,11 @@ use crate::Span;
 /// ```rust
 /// # use datafusion_common::{Location, Span, Diagnostic};
 /// let span = Some(Span {
-///     start: Location{ line: 2, column: 1 },
-///     end: Location{ line: 4, column: 15 }
+///     start: Location { line: 2, column: 1 },
+///     end: Location {
+///         line: 4,
+///         column: 15,
+///     },
 /// });
 /// let diagnostic = Diagnostic::new_error("Something went wrong", span)
 ///     .with_help("Have you tried turning it on and off again?", None);

datafusion/common/src/error.rs

@@ -677,7 +677,10 @@ impl DataFusionError {
 /// let mut builder = DataFusionError::builder();
 /// builder.add_error(DataFusionError::Internal("foo".to_owned()));
 /// // ok_or returns the value if no errors have been added
-/// assert_contains!(builder.error_or(42).unwrap_err().to_string(), "Internal error: foo");
+/// assert_contains!(
+///     builder.error_or(42).unwrap_err().to_string(),
+///     "Internal error: foo"
+/// );
 /// ```
 #[derive(Debug, Default)]
 pub struct DataFusionErrorBuilder(Vec<DataFusionError>);
@@ -695,7 +698,10 @@ impl DataFusionErrorBuilder {
     /// # use datafusion_common::{assert_contains, DataFusionError};
     /// let mut builder = DataFusionError::builder();
     /// builder.add_error(DataFusionError::Internal("foo".to_owned()));
-    /// assert_contains!(builder.error_or(42).unwrap_err().to_string(), "Internal error: foo");
+    /// assert_contains!(
+    ///     builder.error_or(42).unwrap_err().to_string(),
+    ///     "Internal error: foo"
+    /// );
     /// ```
     pub fn add_error(&mut self, error: DataFusionError) {
         self.0.push(error);
@@ -707,8 +713,11 @@ impl DataFusionErrorBuilder {
     /// ```
     /// # use datafusion_common::{assert_contains, DataFusionError};
     /// let builder = DataFusionError::builder()
-    ///   .with_error(DataFusionError::Internal("foo".to_owned()));
-    /// assert_contains!(builder.error_or(42).unwrap_err().to_string(), "Internal error: foo");
+    ///     .with_error(DataFusionError::Internal("foo".to_owned()));
+    /// assert_contains!(
+    ///     builder.error_or(42).unwrap_err().to_string(),
+    ///     "Internal error: foo"
+    /// );
     /// ```
     pub fn with_error(mut self, error: DataFusionError) -> Self {
         self.0.push(error);