C++ and Python Linting + Formatting + Format checks, including pre-commit and pre-push for GPU-code unit testing + CI

.clang-format

@@ -0,0 +1,119 @@
+---
+Language: Cpp
+Standard: c++20
+BasedOnStyle: LLVM
+IndentWidth: 4
+TabWidth: 4
+UseTab: Never
+ColumnLimit: 100
+AccessModifierOffset: -4
+AlignAfterOpenBracket: Align
+AlignConsecutiveAssignments: false
+AlignConsecutiveDeclarations: false
+AlignEscapedNewlines: Right
+AlignOperands: true
+AlignTrailingComments: true
+AllowShortBlocksOnASingleLine: false
+AllowShortCaseLabelsOnASingleLine: false
+AllowShortFunctionsOnASingleLine: Empty
+AllowShortIfStatementsOnASingleLine: false
+AllowShortLoopsOnASingleLine: false
+AlwaysBreakAfterDefinitionReturnType: None
+AlwaysBreakAfterReturnType: None
+AlwaysBreakBeforeMultilineStrings: false
+AlwaysBreakTemplateDeclarations: Yes
+BinPackArguments: true
+BinPackParameters: true
+BraceWrapping:
+  AfterClass: false
+  AfterControlStatement: false
+  AfterEnum: false
+  AfterFunction: true
+  AfterNamespace: false
+  AfterObjCDeclaration: false
+  AfterStruct: false
+  AfterUnion: false
+  AfterExternBlock: false
+  BeforeCatch: false
+  BeforeElse: false
+  IndentBraces: false
+  SplitEmptyFunction: true
+  SplitEmptyRecord: true
+  SplitEmptyNamespace: true
+BreakBeforeBinaryOperators: None
+BreakBeforeBraces: Attach
+BreakBeforeInheritanceComma: false
+BreakInheritanceList: BeforeColon
+BreakBeforeTernaryOperators: true
+BreakConstructorInitializersBeforeComma: false
+BreakConstructorInitializers: BeforeColon
+BreakAfterJavaFieldAnnotations: false
+BreakStringLiterals: true
+CommentPragmas: '^ IWYU pragma:'
+CompactNamespaces: false
+ConstructorInitializerAllOnOneLineOrOnePerLine: false
+ConstructorInitializerIndentWidth: 4
+ContinuationIndentWidth: 4
+Cpp11BracedListStyle: true
+DerivePointerAlignment: false
+DisableFormat: false
+ExperimentalAutoDetectBinPacking: false
+FixNamespaceComments: true
+ForEachMacros:
+  - foreach
+  - Q_FOREACH
+  - BOOST_FOREACH
+IncludeBlocks: Regroup
+IncludeCategories:
+  - Regex: '^<.*\.(cu|cu\.h|cuh|hpp|h)$'
+    Priority: 1
+  - Regex: '^<.*\.(cpp|cc|c\+\+|cxx|c)$'
+    Priority: 2
+  - Regex: '^<.*'
+    Priority: 3
+  - Regex: '.*'
+    Priority: 4
+IncludeIsMainRegex: '(Test)?$'
+IndentCaseLabels: true
+IndentPPDirectives: None
+IndentWrappedFunctionNames: false
+JavaScriptQuotes: Leave
+JavaScriptWrapImports: true
+KeepEmptyLinesAtTheStartOfBlocks: true
+MacroBlockBegin: ''
+MacroBlockEnd: ''
+MaxEmptyLinesToKeep: 1
+NamespaceIndentation: None
+ObjCBinPackProtocolList: Auto
+ObjCBlockIndentWidth: 2
+ObjCSpaceAfterProperty: false
+ObjCSpaceBeforeProtocolList: true
+PenaltyBreakAssignment: 2
+PenaltyBreakBeforeFirstCallParameter: 1
+PenaltyBreakComment: 300
+PenaltyBreakFirstLessLess: 120
+PenaltyBreakString: 1000
+PenaltyBreakTemplateDeclaration: 10
+PenaltyExcessCharacter: 1000000
+PenaltyReturnTypeOnItsOwnLine: 200
+PointerAlignment: Left
+ReflowComments: true
+SortIncludes: true
+SortUsingDeclarations: true
+SpaceAfterCStyleCast: false
+SpaceAfterTemplateKeyword: true
+SpaceBeforeAssignmentOperators: true
+SpaceBeforeParens: ControlStatements
+SpaceBeforeRangeBasedForLoopColon: true
+SpaceInEmptyParentheses: false
+SpacesBeforeTrailingComments: 1
+SpacesInAngles: false
+SpacesInContainerLiterals: true
+SpacesInCStyleCastParentheses: false
+SpacesInParentheses: false
+SpacesInSquareBrackets: false
+StatementMacros:
+  - Q_UNUSED
+  - QT_REQUIRE_VERSION
+...
+

.clang-tidy

@@ -0,0 +1,37 @@
+---
+# Clang-tidy configuration for CUDA/C++ project
+Checks: >
+  -*,
+  bugprone-*,
+  cert-*,
+  cppcoreguidelines-*,
+  performance-*,
+  readability-*,
+  modernize-*,
+  -bugprone-easily-swappable-parameters,
+  -readability-identifier-length,
+  -readability-identifier-naming,
+  -readability-named-parameter,
+  -readability-implicit-bool-conversion,
+  -cppcoreguidelines-avoid-magic-numbers,
+  -cppcoreguidelines-pro-bounds-array-to-pointer-decay,
+  -cppcoreguidelines-pro-bounds-pointer-arithmetic,
+  -cppcoreguidelines-pro-type-union-access,
+  -cppcoreguidelines-pro-type-vararg,
+  -cppcoreguidelines-owning-memory,
+  -modernize-use-trailing-return-type,
+  -readability-magic-numbers
+
+WarningsAsErrors: ''
+HeaderFilterRegex: '.*'
+FormatStyle: 'file'
+CheckOptions:
+  - key: cppcoreguidelines-special-member-functions.AllowSoleDefaultDtor
+    value: true
+  - key: cppcoreguidelines-special-member-functions.AllowMissingMoveFunctions
+    value: true
+  - key: performance-unnecessary-value-param.AllowedTypes
+    value: 'std::vector;.*Iterator'
+  - key: modernize-use-nodiscard.Macros
+    value: 'CUDA_CHECK'
+

.github/workflows/ci.yml

@@ -0,0 +1,42 @@
+name: Format and Lint
+
+on:
+  pull_request:
+    branches:
+      - main
+      - master
+
+jobs:
+  format-and-lint:
+    name: Format and Lint Check
+    runs-on: ubuntu-latest
+
+    steps:
+      - name: Checkout code
+        uses: actions/checkout@v4
+
+      - name: Install pixi
+        uses: prefix-dev/setup-pixi@v0.9.0
+        with:
+          pixi-version: "latest"
+          cache: true  # Cache pixi installation
+
+      - name: Cache pixi environment
+        uses: actions/cache@v4
+        with:
+          path: |
+            ~/.pixi
+            .pixi
+          key: ${{ runner.os }}-pixi-${{ hashFiles('pixi.lock') }}
+          restore-keys: |
+            ${{ runner.os }}-pixi-
+      
+      - name: Install dependencies
+        run: pixi install
+
+      - name: Check formatting
+        run: pixi run format-check
+
+      - name: Lint code
+        run: pixi run lint
+

.pre-commit-config.yaml

@@ -0,0 +1,44 @@
+# Pre-commit hooks configuration for GenMetaBalls
+# See https://pre-commit.com for more information
+
+repos:
+  # Local hooks for custom formatting and linting
+  - repo: local
+    hooks:
+      # Format files (C++/CUDA + Python)
+      - id: format
+        name: Format files
+        entry: pixi run format
+        language: system
+        pass_filenames: false
+        always_run: true
+        stages: [pre-commit]
+
+      # Check formatting (C++/CUDA + Python)
+      - id: format-check
+        name: Check formatting
+        entry: pixi run format-check
+        language: system
+        pass_filenames: false
+        always_run: true
+        stages: [pre-commit]
+
+      # Lint files (C++/CUDA + Python)
+      - id: lint
+        name: Lint files
+        entry: pixi run lint
+        language: system
+        pass_filenames: false
+        always_run: true
+        stages: [pre-commit]
+
+      # Run tests (pre-push hook)
+      - id: test
+        name: Run all tests
+        entry: scripts/test-quiet.sh
+        language: system
+        pass_filenames: false
+        always_run: true
+        stages: [pre-push]
+        verbose: true
+

CMakeLists.txt

@@ -6,6 +6,9 @@ check_language(CUDA)
 
 set(CMAKE_CXX_STANDARD 20)
 
+# Generate compile_commands.json for clang-tidy and other tools
+set(CMAKE_EXPORT_COMPILE_COMMANDS ON)
+
 ################
 # Core Library #
 ################

README.md

@@ -4,12 +4,34 @@ Let's get the ball rolling with blazing fast CUDA kernels!
 
 ## Installation
 
-For initial installation, run:
+### Usage Setup
+
+To simply run `genmetaballs`:
+
+```bash
+pixi install
+```
+
+### Development Setup
+
+For development:
 
 ```bash
 pixi install
+pixi run dev-setup
 ```
 
+The `dev-setup` task sets up [pre-commit](https://pre-commit.com/) git hooks:
+- **Pre-commit**: Formats and lints code before each commit
+- **Pre-push**: Runs all tests before pushes
+
+Run the hooks manually if needed:
+```bash
+pixi run pre-commit-run        # Run formatting/linting checks
+pixi run pre-commit-run-push   # Run all tests
+```
+
+
 ## Testing
 
 ### C++/CUDA Tests
@@ -35,3 +57,37 @@ To run both C++/CUDA and Python tests together:
 ```bash
 pixi run test
 ```
+
+## Formatting & Linting
+
+### Formatting
+
+Format all files (C++/CUDA + Python):
+```bash
+pixi run format
+```
+
+Check formatting without modifying files:
+```bash
+pixi run format-check
+```
+
+### Linting
+
+Lint all files:
+```bash
+pixi run lint
+```
+
+Auto-fix linting issues:
+```bash
+pixi run lint-fix
+```
+
+**Language-specific commands:**
+- `format-cpp`, `format-python` - Format specific language
+- `format-check-cpp`, `format-check-python` - Check formatting for specific language
+- `lint-cpp`, `lint-python` - Lint specific language
+- `lint-cpp-fix`, `lint-python-fix` - Auto-fix linting issues for specific language
+- `lint-full` - C++/CUDA linting with compile_commands.json (more accurate)
+- `lint-full-fix` - Auto-fix C++/CUDA issues using compile_commands.json

genmetaballs/src/cuda/bindings.cu

@@ -1,5 +1,4 @@
 #include <cstdint>
-
 #include <nanobind/nanobind.h>
 #include <nanobind/stl/vector.h>
 
@@ -11,10 +10,6 @@ constexpr uint32_t BLOCK_DIM = 1024;
 namespace nb = nanobind;
 
 NB_MODULE(_genmetaballs_bindings, m) {
-    m.def(
-        "gpu_add",
-        &gpu_add<GRID_DIM, BLOCK_DIM>,
-        "Add two lists elementwise on the GPU",
-        nb::arg("a"), nb::arg("b")
-    );
+    m.def("gpu_add", &gpu_add<GRID_DIM, BLOCK_DIM>, "Add two lists elementwise on the GPU",
+          nb::arg("a"), nb::arg("b"));
 }

genmetaballs/src/cuda/core/add.cuh

@@ -4,22 +4,14 @@
 
 #include "utils.h"
 
-__global__ void add_kernel(
-    float const *a,
-    float const *b,
-    const uint32_t n,
-    float *sum
-) {
+__global__ void add_kernel(float const* a, float const* b, const uint32_t n, float* sum) {
     const uint32_t i = threadIdx.x + blockIdx.x * blockDim.x;
-    if(i < n)
+    if (i < n)
         sum[i] = a[i] + b[i];
 }
 
-template<uint32_t grid_dim, uint32_t block_dim>
-std::vector<float> gpu_add(
-    const std::vector<float> &a_vec,
-    const std::vector<float> &b_vec
-) {
+template <uint32_t grid_dim, uint32_t block_dim>
+std::vector<float> gpu_add(const std::vector<float>& a_vec, const std::vector<float>& b_vec) {
     const uint32_t n = a_vec.size();
     const uint32_t nbytes = n * sizeof(float);
     float *a, *b, *sum;

genmetaballs/src/cuda/core/utils.cu

@@ -4,10 +4,10 @@
 
 #include "utils.h"
 
-void cuda_check(cudaError_t code, const char *file, int line) {
+void cuda_check(cudaError_t code, const char* file, int line) {
     if (code != cudaSuccess) {
-        std::cerr << "CUDA error at " << file << ":" << line << ": "
-                  << cudaGetErrorString(code) << std::endl;
+        std::cerr << "CUDA error at " << file << ":" << line << ": " << cudaGetErrorString(code)
+                  << std::endl;
         exit(1);
     }
 }

genmetaballs/src/genmetaballs/gpu_add.py

@@ -1,4 +1,5 @@
 from . import _genmetaballs_bindings as _gmbb
 
+
 def gpu_add(a: list[float], b: list[float]) -> list[float]:
     return _gmbb.gpu_add(a, b)