Installation
============

Prerequisites
-------------
- Git
- Python 3
- C++23 compiler

Getting the Binary Distribution
-------------------------------

`libxdaq` uses `Conan 2` for package management. While using `libxdaq` without Conan is possible, it is not recommended due to the complexity of manually managing dependencies.
For more information, refer to the `Conan documentation`_.

.. _Conan documentation: https://docs.conan.io/2/

**To install Conan, execute:**

.. code-block:: bash

    pip install --user conan

**To set up the Conan profile:**

Prebuilt `libxdaq` binaries are compiled with the exact settings listed below. Using a
different compiler version, C++ standard, or standard library may result in ABI mismatches.
Do **not** rely on ``conan profile detect``; instead, save the profile matching your platform
into ``~/.conan2/profiles/`` (e.g. ``~/.conan2/profiles/macos-15-arm64``).

macOS (Apple Silicon) — ``macos-15-arm64``:

.. code-block:: ini

    [settings]
    arch=armv8
    compiler=clang
    compiler.cppstd=gnu20
    compiler.libcxx=libc++
    compiler.version=18
    os=Macos
    os.version=13.0
    [conf]
    tools.cmake.cmaketoolchain:generator=Ninja

macOS (Intel) — ``macos-15-x64``:

.. code-block:: ini

    [settings]
    arch=x86_64
    compiler=clang
    compiler.cppstd=gnu20
    compiler.libcxx=libc++
    compiler.version=18
    os=Macos
    os.version=13.0
    [conf]
    tools.cmake.cmaketoolchain:generator=Ninja

Linux (x86_64) — ``ubuntu-22.04-x64``:

.. code-block:: ini

    [settings]
    arch=x86_64
    compiler=gcc
    compiler.cppstd=gnu20
    compiler.libcxx=libstdc++11
    compiler.version=13
    os=Linux
    [conf]
    tools.cmake.cmaketoolchain:generator=Ninja

Windows (x86_64, MSVC) — ``windows-2022-x64``:

.. code-block:: ini

    [settings]
    arch=x86_64
    compiler=msvc
    compiler.cppstd=20
    compiler.runtime=dynamic
    compiler.version=194
    os=Windows
    [conf]
    tools.cmake.cmaketoolchain:generator=Ninja

Then pass the profile to every Conan command, e.g.:

.. code-block:: bash

    conan install . --profile:all macos-15-arm64 -s build_type=Release

**To add the kontex-neuro Conan repository:**

.. code-block:: bash

    git clone https://github.com/kontex-neuro/kontex-conan.git
    conan remote add kontex-neuro ./kontex-conan

After adding the remote, you can use `libxdaq` as a dependency in your project's `conanfile.txt` or `conanfile.py`, and Conan will automatically fetch it from the kontex-neuro repository.

**To update to newer package versions:**

.. code-block:: bash

    cd kontex-conan
    git pull