Merge branch 'main' into fix-exclude

Author: xiedeyantu

.github/workflows/codeql.yml

@@ -32,7 +32,7 @@ permissions:
 jobs:
   analyze:
     name: Analyze Actions
-    runs-on: ubuntu-latest
+    runs-on: ubuntu-slim
     permissions:
       contents: read
       security-events: write

.github/workflows/dev.yml

@@ -41,7 +41,7 @@ jobs:
 
   prettier:
     name: Use prettier to check formatting of documents
-    runs-on: ubuntu-latest
+    runs-on: ubuntu-slim
     steps:
       - uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd  # v6.0.2
       - uses: actions/setup-node@6044e13b5dc448c55e2357c09f80417699197238  # v6.2.0

.github/workflows/extended.yml

@@ -44,15 +44,10 @@ on:
       - 'datafusion/physical*/**/*.rs'
       - 'datafusion/expr*/**/*.rs'
       - 'datafusion/optimizer/**/*.rs'
+      - 'datafusion/sql/**/*.rs'
       - 'datafusion-testing'
   workflow_dispatch:
     inputs:
-      pr_number:
-        description: 'Pull request number'
-        type: string
-      check_run_id:
-        description: 'Check run ID for status updates'
-        type: string
       pr_head_sha:
         description: 'PR head SHA'
         type: string
@@ -93,8 +88,7 @@ jobs:
   linux-test-extended:
     name: cargo test 'extended_tests' (amd64)
     needs: [linux-build-lib]
-    runs-on: ${{ github.repository_owner == 'apache' && format('runs-on={0},family=m8a+m7a+c8a,cpu=32,image=ubuntu24-full-x64,extras=s3-cache,disk=large,tag=datafusion,spot=false', github.run_id) || 'ubuntu-latest' }}
-    # spot=false because the tests are long, https://runs-on.com/configuration/spot-instances/#disable-spot-pricing
+    runs-on: ${{ github.repository_owner == 'apache' && format('runs-on={0},family=m8a+m7a+c8a,cpu=32,image=ubuntu24-full-x64,extras=s3-cache,disk=large,tag=datafusion', github.run_id) || 'ubuntu-latest' }}
     # note: do not use amd/rust container to preserve disk space
     steps:
       - uses: runs-on/action@cd2b598b0515d39d78c38a02d529db87d2196d1e  # v2.0.3
@@ -162,8 +156,7 @@ jobs:
 
   sqllogictest-sqlite:
     name: "Run sqllogictests with the sqlite test suite"
-    runs-on: ${{ github.repository_owner == 'apache' && format('runs-on={0},family=m8a+m7a+c8a,cpu=48,image=ubuntu24-full-x64,extras=s3-cache,disk=large,tag=datafusion,spot=false', github.run_id) || 'ubuntu-latest' }}
-    # spot=false because the tests are long, https://runs-on.com/configuration/spot-instances/#disable-spot-pricing
+    runs-on: ${{ github.repository_owner == 'apache' && format('runs-on={0},family=m8a+m7a+c8a,cpu=32,image=ubuntu24-full-x64,extras=s3-cache,disk=large,tag=datafusion', github.run_id) || 'ubuntu-latest' }}
     container:
       image: amd64/rust
     steps:

.github/workflows/labeler.yml

@@ -31,7 +31,7 @@ on:
 jobs:
   process:
     name: Process
-    runs-on: ubuntu-latest
+    runs-on: ubuntu-slim
     # only run for users whose permissions allow them to update PRs
     # otherwise labeler is failing:
     # https://github.com/apache/datafusion/issues/3743

.github/workflows/large_files.yml

@@ -27,7 +27,7 @@ on:
 
 jobs:
   check-files:
-    runs-on: ubuntu-latest
+    runs-on: ubuntu-slim
     steps:
       - uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd  # v6.0.2
         with:

.github/workflows/stale.yml

@@ -22,7 +22,7 @@ on:
 
 jobs:
   close-stale-prs:
-    runs-on: ubuntu-latest
+    runs-on: ubuntu-slim
     permissions:
       issues: write
       pull-requests: write

.github/workflows/take.yml

@@ -25,7 +25,7 @@ permissions:
 
 jobs:
   issue_assign:
-    runs-on: ubuntu-latest
+    runs-on: ubuntu-slim
     if: (!github.event.issue.pull_request) && (github.event.comment.body == 'take' || github.event.comment.body == 'untake')
     concurrency:
       group: ${{ github.actor }}-issue-assign

.gitignore

@@ -75,3 +75,6 @@ rat.txt
 
 # data generated by examples
 datafusion-examples/examples/datafusion-examples/
+
+# Samply profile data
+profile.json.gz

Cargo.lock

@@ -224,62 +224,57 @@ unused_qualifications = "deny"
 # --------------------
 # Compilation Profiles
 # --------------------
-#  A Cargo profile is a preset for the compiler/linker knobs that trade off:
+# A Cargo profile is a preset for the compiler/linker knobs that trades off:
 # - Build time: how quickly code compiles and links
 # - Runtime performance: how fast the resulting binaries execute
 # - Binary size: how large the executables end up
 # - Debuggability: how much debug information is preserved for debugging and profiling
 #
+# To use a profile: `cargo [ build | run | ... ] --profile <profile-name>`
+#
 # Profiles available:
-# - dev: default debug build; fastest to compile, slowest to run, full debug info
-#     for everyday development.
-#     Run: cargo run
-# - release: optimized build; slowest to compile, fastest to run, smallest
-#     binaries for public releases.
-#     Run: cargo run --release
-# - release-nonlto: skips LTO, so it builds quicker while staying close to
-#     release performance. It is useful when developing performance optimizations.
-#     Run: cargo run --profile release-nonlto
+# - dev: default debug build; fastest to compile, slowest to run, full debug info.
+#     For everyday development; default for "cargo [ build | test | run ]".
+# - release: fully optimized build; slowest to compile, fastest to run, smallest
+#     binaries. For public releases; default for "cargo [ bench | install ]".
+# - release-nonlto: skips LTO, so it builds much faster while staying close to
+#     release performance. Useful when developing performance optimizations.
 # - profiling: inherits release optimizations but retains debug info to support
 #     profiling tools and flamegraphs.
-#     Run: cargo run --profile profiling
-# - ci: derived from `dev` but disables incremental builds and strips dependency
-#     symbols to keep CI artifacts small and reproducible.
-#     Run: cargo run --profile ci
-# - ci-optimized: derived from `release` but enables debug assertions, and uses
-#     lighter optimizations. Used for long-running CI tasks.
-#     Run: cargo run --profile ci-release
+# - ci: derived from `dev` but disables debug info and incremental builds to keep
+#     CI artifacts small and reproducible.
+# - ci-optimized: derived from `release` but enables debug assertions and uses
+#     less aggressive optimizations for faster builds.  Used for long-running CI
+#     tasks.
 #
 # If you want to optimize compilation, the `compile_profile` benchmark can be useful.
 # See `benchmarks/README.md` for more details.
 [profile.release]
 codegen-units = 1
 lto = true
-strip = true      # Eliminate debug information to minimize binary size
+strip = true      # Eliminate debug info to minimize binary size
 
 [profile.release-nonlto]
-codegen-units = 16
-debug-assertions = false
-incremental = false
 inherits = "release"
+codegen-units = 16
 lto = false
-opt-level = 3
-overflow-checks = false
-rpath = false
-strip = false            # Retain debug info for flamegraphs
+strip = false        # Retain debug info for flamegraphs
+
+[profile.profiling]
+inherits = "release"
+debug = true
+strip = false
 
 [profile.ci-optimized]
 inherits = "release"
 debug-assertions = true
 codegen-units = 16
 lto = "thin"
-strip = true
 
 [profile.ci]
-debug = false
 inherits = "dev"
+debug = false
 incremental = false
-debug-assertions = true
 
 # This rule applies to every package except workspace members (dependencies
 # such as `arrow` and `tokio`). It disables debug info and related features on
@@ -289,8 +284,3 @@ debug = false
 debug-assertions = false
 strip = "debuginfo"
 incremental = false
-
-[profile.profiling]
-inherits = "release"
-debug = true
-strip = false