Expose seismic data as numpy array

.clang-format

@@ -0,0 +1,81 @@
+# Google C/C++ Code Style settings
+# https://clang.llvm.org/docs/ClangFormatStyleOptions.html
+Language: Cpp
+BasedOnStyle: Google
+AccessModifierOffset: -1
+AlignAfterOpenBracket: Align
+AlignConsecutiveAssignments: None
+AlignOperands: Align
+AllowAllArgumentsOnNextLine: true
+AllowAllConstructorInitializersOnNextLine: true
+AllowAllParametersOfDeclarationOnNextLine: false
+AllowShortBlocksOnASingleLine: Empty
+AllowShortCaseLabelsOnASingleLine: false
+AllowShortFunctionsOnASingleLine: Inline
+AllowShortIfStatementsOnASingleLine: Never  # To avoid conflict, set this "Never" and each "if statement" should include brace when coding
+AllowShortLambdasOnASingleLine: Inline
+AllowShortLoopsOnASingleLine: false
+AlwaysBreakAfterReturnType: None
+AlwaysBreakTemplateDeclarations: Yes
+BinPackArguments: true
+BreakBeforeBraces: Custom
+BraceWrapping:
+  AfterCaseLabel: false
+  AfterClass: false
+  AfterStruct: false
+  AfterControlStatement: Never
+  AfterEnum: false
+  AfterFunction: false
+  AfterNamespace: false
+  AfterUnion: false
+  AfterExternBlock: false
+  BeforeCatch: false
+  BeforeElse: false
+  BeforeLambdaBody: false
+  IndentBraces: false
+  SplitEmptyFunction: false
+  SplitEmptyRecord: false
+  SplitEmptyNamespace: false
+BreakBeforeBinaryOperators: None
+BreakBeforeTernaryOperators: true
+BreakConstructorInitializers: BeforeColon
+BreakInheritanceList: BeforeColon
+ColumnLimit: 100
+CompactNamespaces: false
+ContinuationIndentWidth: 4
+Cpp11BracedListStyle: true
+DerivePointerAlignment: false  # Make sure the * or & align on the left
+EmptyLineBeforeAccessModifier: LogicalBlock
+FixNamespaceComments: true
+IncludeBlocks: Preserve
+IndentCaseLabels: true
+IndentPPDirectives: None
+IndentWidth: 2
+KeepEmptyLinesAtTheStartOfBlocks: true
+MaxEmptyLinesToKeep: 1
+NamespaceIndentation: None
+ObjCSpaceAfterProperty: false
+ObjCSpaceBeforeProtocolList: true
+PointerAlignment: Left
+ReflowComments: false
+# SeparateDefinitionBlocks: Always   # Only support since clang-format 14
+SpaceAfterCStyleCast: false
+SpaceAfterLogicalNot: false
+SpaceAfterTemplateKeyword: true
+SpaceBeforeAssignmentOperators: true
+SpaceBeforeCpp11BracedList: false
+SpaceBeforeCtorInitializerColon: true
+SpaceBeforeInheritanceColon: true
+SpaceBeforeParens: ControlStatements
+SpaceBeforeRangeBasedForLoopColon: true
+SpaceBeforeSquareBrackets: false
+SpaceInEmptyParentheses: false
+SpacesBeforeTrailingComments: 2
+SpacesInAngles: false
+SpacesInCStyleCastParentheses: false
+SpacesInContainerLiterals: false
+SpacesInParentheses: false
+SpacesInSquareBrackets: false
+Standard: c++11
+TabWidth: 4
+UseTab: Never

.github/workflows/CI.yml

@@ -3,7 +3,7 @@ name: CI
 on:
   push:
     branches:
-    - "main"
+      - "main"
   pull_request:
 
 env:
@@ -37,6 +37,23 @@ jobs:
         # See https://cmake.org/cmake/help/latest/manual/ctest.1.html for more detail
         run: ctest -C ${{env.BUILD_TYPE}}
 
+  lint_cpp:
+    name: "Lint C++"
+    strategy:
+      matrix:
+        os: [ ubuntu-latest ]
+    runs-on: ${{ matrix.os }}
+
+    steps:
+      - uses: actions/checkout@v4
+
+      - name: 'Run clang-format.'
+        uses: jidicula/[email protected]
+        with:
+          clang-format-version: '20'
+          check-path: 'src'
+
+
   python_tests:
     name: "Python Tests"
     strategy:

.pre-commit-config.yaml

@@ -17,49 +17,49 @@ ci:
   autofix_commit_msg: "style: pre-commit fixes"
 
 repos:
-# Standard hooks
-- repo: https://github.com/pre-commit/pre-commit-hooks
-  rev: v5.0.0
-  hooks:
-  - id: check-added-large-files
-  - id: check-case-conflict
-  - id: check-merge-conflict
-  - id: check-symlinks
-  - id: check-yaml
-    exclude: ^conda\.recipe/meta\.yaml$
-  - id: debug-statements
-  - id: end-of-file-fixer
-  - id: mixed-line-ending
-  - id: requirements-txt-fixer
-  - id: trailing-whitespace
+  # Standard hooks
+  - repo: https://github.com/pre-commit/pre-commit-hooks
+    rev: v5.0.0
+    hooks:
+      - id: check-added-large-files
+      - id: check-case-conflict
+      - id: check-merge-conflict
+      - id: check-symlinks
+      - id: check-yaml
+        exclude: ^conda\.recipe/meta\.yaml$
+      - id: debug-statements
+      - id: end-of-file-fixer
+      - id: mixed-line-ending
+      - id: requirements-txt-fixer
+      - id: trailing-whitespace
 
-# Check linting and style issues
-- repo: https://github.com/astral-sh/ruff-pre-commit
-  rev: "v0.11.12"
-  hooks:
-    - id: ruff
-      args: ["--fix", "--show-fixes"]
-    - id: ruff-format
-      exclude: ^(docs)
+  # Check linting and style issues
+  - repo: https://github.com/astral-sh/ruff-pre-commit
+    rev: "v0.11.12"
+    hooks:
+      - id: ruff
+        args: [ "--fix", "--show-fixes" ]
+      - id: ruff-format
+        exclude: ^(docs)
 
-# Changes tabs to spaces
-- repo: https://github.com/Lucas-C/pre-commit-hooks
-  rev: v1.5.5
-  hooks:
-  - id: remove-tabs
-    exclude: ^(docs)
+  # Changes tabs to spaces
+  - repo: https://github.com/Lucas-C/pre-commit-hooks
+    rev: v1.5.5
+    hooks:
+      - id: remove-tabs
+        exclude: ^(docs)
 
-# CMake formatting
-- repo: https://github.com/cheshirekow/cmake-format-precommit
-  rev: v0.6.13
-  hooks:
-  - id: cmake-format
-    additional_dependencies: [pyyaml]
-    types: [file]
-    files: (\.cmake|CMakeLists.txt)(.in)?$
+  # CMake formatting
+  - repo: https://github.com/cheshirekow/cmake-format-precommit
+    rev: v0.6.13
+    hooks:
+      - id: cmake-format
+        additional_dependencies: [ pyyaml ]
+        types: [ file ]
+        files: (\.cmake|CMakeLists.txt)(.in)?$
 
-# .clang-format file
-- repo: https://github.com/pre-commit/mirrors-clang-format
-  rev: v13.0.0
-  hooks:
-    - id: clang-format
+  # .clang-format file
+  - repo: https://github.com/pre-commit/mirrors-clang-format
+    rev: v20.1.6
+    hooks:
+      - id: clang-format

CMakeLists.txt

@@ -4,6 +4,7 @@ project(
   ${SKBUILD_PROJECT_NAME}
   VERSION ${SKBUILD_PROJECT_VERSION}
   LANGUAGES CXX)
+set(CMAKE_POSITION_INDEPENDENT_CODE ON)
 
 # Include core modules
 add_subdirectory(src/core)

pyproject.toml

@@ -17,6 +17,9 @@ classifiers = [
     "Programming Language :: Python :: 3.13",
     "Private :: Do Not Upload",
 ]
+dependencies = [
+    "numpy>=2.2.6",
+]
 
 
 [dependency-groups]

src/core/add/add.cpp

@@ -5,5 +5,7 @@
 #include "add.hpp"
 
 namespace libseis {
-int add(const int i, const int j) { return i + j; }
-} // namespace libseis
+int add(const int i, const int j) {
+  return i + j;
+}
+}  // namespace libseis

src/core/add/add.hpp

@@ -9,4 +9,4 @@ namespace libseis {
 int add(int i, int j);
 }
 
-#endif // CPPLIB_LIBRARY_H
+#endif  // CPPLIB_LIBRARY_H

src/core/bindings.cpp

@@ -2,9 +2,9 @@
 // Created by David Graham on 6/8/25.
 //
 
-#include "add/add.hpp"
+#include <pybind11/numpy.h>
 #include <pybind11/pybind11.h>
-#include <pybind11/stl.h>
+#include "add/add.hpp"
 
 #include "gather.hpp"
 
@@ -40,13 +40,29 @@ PYBIND11_MODULE(_num, m) {
 PYBIND11_MODULE(_gather, m) {
   py::class_<Gather>(m, "Gather")
       // .def(py::init<int, int, int, double>())
-      .def(py::init<int, int, int, double>(), py::arg("id"), py::arg("nt"),
-           py::arg("nx"), py::arg("dt"))
+      .def(py::init<int, int, int, double>(), py::arg("id"), py::arg("nt"), py::arg("nx"),
+           py::arg("dt"))
       .def_readwrite("id", &Gather::id)
+      .def_readwrite("dt", &Gather::dt)
       .def_readwrite("nt", &Gather::nt)
       .def_readwrite("nx", &Gather::nx)
-      .def_readwrite("dt", &Gather::dt)
-      .def_readonly("data", &Gather::data)
+      .def_property(
+          "data",
+          // Getter: return numpy array (zero-copy)
+          [](Gather& self) -> py::array {
+            return py::array({self.nt, self.nx}, {sizeof(double) * self.nx, sizeof(double)},
+                             self.data.data(), py::cast(&self));
+          },
+          // Setter: accept 2D NumPy array and copy into C++ vector
+          [](Gather& self,
+             const py::array_t<double, py::array::c_style | py::array::forcecast>& arr) {
+            if (arr.ndim() != 2)
+              throw std::runtime_error("data must be 2D");
+            if (arr.shape(0) != self.nt || arr.shape(1) != self.nx)
+              throw std::runtime_error("Shape mismatch with (nt, nx)");
+
+            std::memcpy(self.data.data(), arr.data(), sizeof(double) * self.nt * self.nx);
+          })
       .def("__str__", &Gather::str)
       .def("from_bin_file", &Gather::from_bin_file);
 }

src/core/gather/gather.cpp

@@ -11,66 +11,27 @@
 #include <string>
 #include <vector>
 
-// struct Gather {
-//   int id;
-//   int nt;
-//   int nx;
-//   double dt;
-//   std::vector<double> data;
-// };
+std::string Gather::str() const {
+  return "Gather id: " + std::to_string(id);
+}
 
-// std::vector<double> read_double(const std::string &path, int nt, int nx) {
-//   std::ifstream file(path, std::ios::binary);
-//   if (!file) {
-//     throw std::runtime_error("Unable to open file at " + path);
-//   }
-//
-//   std::vector<double> data(nt * nx);
-//   file.read(reinterpret_cast<char *>(data.data()), nt * nx * sizeof(double));
-//   if (!file) {
-//     throw std::runtime_error("Failed to read data from file at " + path);
-//   }
-//
-//   return data;
-// }
-//
-// void write_double(const std::string &path, const std::vector<double> &data,
-//                   int nt, int nx) {
-//   std::ofstream file(path, std::ios::binary);
-//   if (!file) {
-//     throw std::runtime_error("Unable to open file at " + path);
-//   }
-//
-//   file.write(reinterpret_cast<const char *>(data.data()),
-//              nt * nx * sizeof(double));
-//   if (!file) {
-//     throw std::runtime_error("Failed to write data to file at " + path);
-//   }
-// }
-//
-// std::unique_ptr<Gather> gain_gather(const Gather &gather, double power) {
-//   auto new_gather = std::make_unique<Gather>(gather);
-//   new_gather->data.resize(gather.nt * gather.nx);
-//
-//   for (int trace_index = 0; trace_index < gather.nx; ++trace_index) {
-//     for (int sample_index = 0; sample_index < gather.nt; ++sample_index) {
-//       double time = static_cast<double>(sample_index) * gather.dt;
-//       double t_pow = std::pow(time, power);
-//       int data_index = trace_index * gather.nt + sample_index;
-//       new_gather->data[data_index] = gather.data[data_index] * t_pow;
-//     }
-//   }
-//
-//   return new_gather;
-// }
-//
-// void display_gather(const Gather &gather) {
-//   std::cout << "--------------------\n";
-//   std::cout << "Gather: id " << gather.id << '\n';
-//   std::cout << "Sample Rate: " << gather.dt << " seconds\n";
-//   std::cout << "Sampling Frequency: " << 1.0 / gather.dt << " Hz\n";
-//   std::cout << "Nyquist Frequency: " << 1.0 / (2 * gather.dt) << " Hz\n";
-//   std::cout << "Samples per trace: " << gather.nt << '\n';
-//   std::cout << "Number of traces: " << gather.nx << '\n';
-//   std::cout << "--------------------\n";
-// }
+std::ostream& operator<<(std::ostream& os, const Gather& gather) {
+  os << "Gather id: " << gather.id << std::endl;
+  return os;
+}
+
+Gather Gather::from_bin_file(const std::string& path, int nt, int nx, double dt) {
+  std::ifstream file(path, std::ios::binary);
+  if (!file) {
+    throw std::runtime_error("Unable to open file at " + path);
+  }
+  std::size_t total_values = static_cast<std::size_t>(nt) * nx;
+  std::vector<double> buffer(total_values);
+  file.read(reinterpret_cast<char*>(buffer.data()), total_values * sizeof(double));
+  if (!file) {
+    throw std::runtime_error("Failed to read data from file at " + path);
+  }
+  Gather result(0, nt, nx, dt);
+  result.data = std::move(buffer);
+  return result;
+}