.. _c-api-core: C API reference =============== .. note:: This is the documentation for ``metatensor-core`` version |metatensor-core-version|. For other versions, look in the following pages: .. version-list:: :tag-prefix: metatensor-core-v :url-suffix: core/reference/c/index.html .. version:: 0.1.11 .. version:: 0.1.10 .. version:: 0.1.9 .. version:: 0.1.7 .. version:: 0.1.6 .. version:: 0.1.5 .. version:: 0.1.4 .. version:: 0.1.3 ``metatensor`` offers a C API that can be called from any language able to call C functions (in particular, this includes Python, Fortran with ``iso_c_env``, C++, and most languages used nowadays). Convenient wrappers of the C API are also provided for :ref:`Python ` users. The C API is implemented in Rust. You can use these functions in your own code by :ref:`installing the corresponding shared library and header `, and then including ``metatensor.h`` and linking with ``-lmetatensor``. Alternatively, we provide a cmake package config file, allowing you to do use ``metatensor`` like this (after installation): .. code-block:: cmake cmake_minimum_required(VERSION 3.16) project(my-project C) find_package(metatensor) # add executables/libraries add_executable(my-exe exe-source.c) add_library(my-lib lib-source.c) # Link to metatensor, this makes the header accessible and link to the right # libraries. # # The `metatensor` target will be an alias for `metatensor::shared` # or `metatensor::static` depending how you've installed the code. target_link_libraries(my-exe metatensor) target_link_libraries(my-lib metatensor) # alternatively, you can explicitly use the static or shared build of # metatensor. Unless you have a very specific need for a static build, we # recommend using the shared version of metatensor: this will allow to pass # data from your code to any other code using metatensor. # target_link_libraries(my-exe metatensor::shared) # target_link_libraries(my-exe metatensor::static) The functions and types provided in ``metatensor.h`` can be grouped in five main groups: .. toctree:: :maxdepth: 1 tensor labels block data misc