Merge remote-tracking branch 'apache/main' into alamb/backport_instructions

Author: alamb

.github/workflows/extended.yml

@@ -173,10 +173,10 @@ jobs:
           ref: ${{ github.event.inputs.pr_head_sha }} # will be empty if triggered by push
           submodules: true
           fetch-depth: 1
-      - name: Setup Rust toolchain
-        uses: ./.github/actions/setup-builder
-        with:
-          rust-version: stable
+      # Don't use setup-builder to avoid configuring RUST_BACKTRACE which is expensive
+      - name: Install protobuf compiler
+        run: |
+          apt-get update && apt-get install -y protobuf-compiler
       - name: Run sqllogictest
         run: |
           cargo test --features backtrace,parquet_encryption --profile ci-optimized --test sqllogictests -- --include-sqlite

AGENTS.md

@@ -2,33 +2,22 @@
 
 ## Developer Documentation
 
+- [Quick Start Setup](docs/source/contributor-guide/development_environment.md#quick-start)
+- [Testing Quick Start](docs/source/contributor-guide/testing.md#testing-quick-start)
+- [Before Submitting a PR](docs/source/contributor-guide/index.md#before-submitting-a-pr)
 - [Contributor Guide](docs/source/contributor-guide/index.md)
 - [Architecture Guide](docs/source/contributor-guide/architecture.md)
 
 ## Before Committing
 
-Before committing any changes, you **must** run the following checks and fix any issues:
+Before committing any changes, you MUST follow the instructions in
+[Before Submitting a PR](docs/source/contributor-guide/index.md#before-submitting-a-pr)
+and ensure the required checks listed there pass. Do not commit code that
+fails any of those checks.
 
-```bash
-cargo fmt --all
-cargo clippy --all-targets --all-features -- -D warnings
-```
-
-- `cargo fmt` ensures consistent code formatting across the project.
-- `cargo clippy` catches common mistakes and enforces idiomatic Rust patterns. All warnings must be resolved (treated as errors via `-D warnings`).
-
-Do not commit code that fails either of these checks.
+When creating a PR, you MUST follow the [PR template](.github/pull_request_template.md).
 
 ## Testing
 
-Run relevant tests before submitting changes:
-
-```bash
-cargo test --all-features
-```
-
-For SQL logic tests:
-
-```bash
-cargo test -p datafusion-sqllogictest
-```
+See the [Testing Quick Start](docs/source/contributor-guide/testing.md#testing-quick-start)
+for the recommended pre-PR test commands.

Cargo.lock

@@ -462,7 +462,7 @@ mod tests {
     use std::ops::Not;
 
     use super::*;
-    use datafusion_expr::{Expr, case, col, lit};
+    use datafusion_expr::{case, col};
 
     #[test]
     fn test_split_files() {

datafusion/catalog-listing/src/helpers.rs

@@ -920,7 +920,6 @@ mod tests {
     use arrow::compute::SortOptions;
     use datafusion_physical_expr::expressions::Column;
     use datafusion_physical_expr_common::sort_expr::PhysicalSortExpr;
-    use std::sync::Arc;
 
     /// Helper to create a PhysicalSortExpr
     fn sort_expr(

datafusion/catalog-listing/src/table.rs

@@ -486,7 +486,7 @@ pub trait TableProviderFactory: Debug + Sync + Send {
 }
 
 /// A trait for table function implementations
-pub trait TableFunctionImpl: Debug + Sync + Send {
+pub trait TableFunctionImpl: Debug + Sync + Send + Any {
     /// Create a table provider
     fn call(&self, args: &[Expr]) -> Result<Arc<dyn TableProvider>>;
 }

datafusion/catalog/src/table.rs

@@ -1088,6 +1088,12 @@ config_namespace! {
         /// process to reorder the join keys
         pub top_down_join_key_reordering: bool, default = true
 
+        /// When set to true, the physical plan optimizer may swap join inputs
+        /// based on statistics. When set to false, statistics-driven join
+        /// input reordering is disabled and the original join order in the
+        /// query is used.
+        pub join_reordering: bool, default = true
+
         /// When set to true, the physical plan optimizer will prefer HashJoin over SortMergeJoin.
         /// HashJoin can work more efficiently than SortMergeJoin but consumes more memory
         pub prefer_hash_join: bool, default = true