Python binding generation with nanobind

python-binding-generation-with-nanobind

License: restricted — no clear open-source license detected for the underlying tool; verify licensing before commercial use or redistribution. <!-- asb-license-banner -->

Summary

Generate importable Python modules from C++ source code using nanobind binding specifications and CMake compilation. This skill bridges OpenMS C++ API functionality into Python environments by converting binding declarations into compiled extension modules.

When to use

When you have a C++ library (such as OpenMS) with nanobind binding specifications in a designated bindings directory and need to create a Python module that exposes C++ classes, functions, and data types to Python code. Use this skill when preparing a library for Python-first development workflows or when testing whether the full C++ API surface is accessible from Python.

When NOT to use

• Input is already a compiled Python wheel or conda package — skip to installation instead of rebuilding.
• Binding specification files are incomplete or missing for the C++ API surface you need to expose.
• Target C++ library cannot be compiled on the platform where you want to build the Python bindings (e.g., missing platform-specific toolchain or incompatible dependencies).

Inputs

• nanobind binding specification files (C++ source with nb::module declarations)
• CMakeLists.txt with nanobind build targets
• OpenMS C++ library headers and compiled object files
• Python development headers and interpreter (version matching the target binding)

Outputs

• Compiled Python extension module (.so on Linux, .pyd on Windows, .dylib on macOS)
• Importable pyOpenMS module accessible from Python sys.modules
• Verified binding completeness attestation (successful import + callable C++ functions/classes)

How to apply

First, review the nanobind binding specification files (typically in src/pyOpenMS/bindings/ for OpenMS) to understand which C++ types and functions are exposed. Configure the CMake build system to compile these bindings into a Python extension module by specifying nanobind as a build dependency and ensuring the binding files are included in the compilation targets. Execute the CMake configuration and build process to generate the compiled .so/.pyd/.dylib extension module. Import the resulting module in a Python interpreter using standard import syntax (e.g., import pyOpenMS), and verify module loading completes without ImportError or symbol resolution failures. Finally, execute a simple function call or attribute access on an exposed C++ class or function to confirm that the binding layer correctly translates between Python and C++ calling conventions.

Related tools

• nanobind (Python binding framework used to define and generate C++-to-Python conversion layer from binding specifications)
• CMake (Build system that configures compilation of nanobind binding sources into a linkable Python extension module)
• OpenMS (C++ mass spectrometry library being exposed to Python through pyOpenMS bindings) — https://github.com/OpenMS/OpenMS

Evaluation signals

• Module imports without raising ImportError, ModuleNotFoundError, or unresolved symbol errors when executed in a Python interpreter.
• Calling a bound C++ function or instantiating a bound class (e.g., obj = pyOpenMS.SomeClass()) succeeds and returns a Python object with expected attributes/methods.
• No segmentation faults or memory access violations occur during simple function calls or method invocations on bound C++ objects.
• Introspection on imported module reveals expected classes, functions, and type signatures matching the binding declarations (e.g., dir(pyOpenMS) and help(pyOpenMS.SomeClass) show expected items).
• CMake build log confirms all nanobind binding source files were compiled and linked without errors or warnings.

Limitations

• The provided document fragment lacks sufficient technical detail on binding file structure, module organization, and symbol export strategies; practitioners may need to consult nanobind official documentation and OpenMS-specific binding conventions beyond this skill description.
• Platform-specific toolchain compatibility (C++ compiler, Python development headers version match) is a hard prerequisite; binding generation will fail silently or with cryptic linker errors if toolchain versions are mismatched.
• Only C++ API surface for which binding specifications have been explicitly written will be accessible from Python; unbound C++ functionality remains inaccessible, even if present in the C++ library.

Evidence

• [other] Navigate to the src/pyOpenMS/bindings/ directory and review nanobind binding specifications according to CLAUDE.md wrapping instructions.: "Navigate to the src/pyOpenMS/bindings/ directory and review nanobind binding specifications according to CLAUDE.md wrapping instructions."
• [other] Configure the build system (CMake) to compile nanobind binding files into a Python extension module.: "Configure the build system (CMake) to compile nanobind binding files into a Python extension module."
• [other] Execute the build process to generate the compiled pyOpenMS module.: "Execute the build process to generate the compiled pyOpenMS module."
• [other] Import the generated pyOpenMS module in a Python environment and verify that the module loads without errors.: "Import the generated pyOpenMS module in a Python environment and verify that the module loads without errors."
• [readme] With pyOpenMS, OpenMS offers Python bindings to a large part of the OpenMS API to enable rapid algorithm development.: "With pyOpenMS, OpenMS offers Python bindings to a large part of the OpenMS API to enable rapid algorithm development."