build: move to pyproject

docs/installing-oggm.rst

@@ -15,7 +15,7 @@ are not trivial to install. The instructions below provide all the required
 details and should work on Linux and Mac OS. See :ref:`install-troubleshooting`
 if something goes wrong.
 
-OGGM is fully `tested`_ with Python versions 3.9 to 3.11 on Linux.
+OGGM is fully `tested`_ with Python versions 3.11 to 3.13 on Linux.
 We do not test OGGM automatically on Mac OSX, but it should probably run
 fine there as well.
 
@@ -33,59 +33,33 @@ fine there as well.
         installing-oggm-windows.rst
 
 
-For most users we recommend to
-install Python and the package dependencies with the :ref:`conda package manager <conda-install>`,
-in particular with ``mamba`` and ``conda-forge``.
+For most users we recommend to install Python and the package dependencies with the :ref:`conda package manager <conda-install>`, in particular with ``mamba`` and ``conda-forge``.
 
 .. _tested: https://github.com/OGGM/oggm/actions/workflows/run-tests.yml
 .. _conda: https://conda.io/projects/conda/en/latest/user-guide/index.html
 .. _pip: https://docs.python.org/3/installing/
 .. _mamba: https://mamba.readthedocs.io
 
 
+OGGM can be installed:
+- as a library, if you don't want to modify its source code.This is recommended for most users.
+    - **stable**: this is the latest official release and has a fixed
+      version number (e.g. v1.6.2).
+    - **dev**: this is the development version.
+      It may contain new features and bug fixes, but will continue changing until its release.
+- as an editable, if you want to make changes or develop the model. This is recommended for developers.
+
+OGGM now supports installation with `pip`, `conda`, and `uv`.
+
 Dependencies
 ------------
 
-Here is a list of *all* dependencies of the OGGM model. If you want to use
-OGGM's numerical models only (i.e. no GIS or preprocessing tools), refer to
-`Install a minimal OGGM environment`_ below.
-
-Standard SciPy stack:
-    - numpy
-    - scipy
-    - scikit-image
-    - pillow
-    - matplotlib
-    - pandas
-    - xarray
-    - dask
-    - joblib
-
-Configuration file parsing tool:
-    - configobj
-
-I/O:
-    - netcdf4
-    - pytables
-
-GIS tools:
-    - shapely
-    - pyproj
-    - rasterio
-    - rioxarray
-    - geopandas
-
-Testing:
-    - pytest
-    - pytest-mpl (for image tests only: `OGGM fork <https://github.com/OGGM/pytest-mpl>`_ required)
-
-Other libraries:
-    - `salem <https://github.com/fmaussion/salem>`_
-    - `motionless <https://github.com/ryancox/motionless>`_
-
-Optional:
-    - progressbar2 (displays the download progress)
-    - bottleneck (might speed up some xarray operations)
+A full installation of OGGM requires GDAL.
+The easiest way is:
+.. code-block:: bash
+
+    sudo apt-get install gdal-bin libgdal-dev  # Linux (Debian distros)
+    brew install gdal  # MacOS
 
 .. _conda-install:
 
@@ -112,9 +86,8 @@ which explains how to install ``mambaforge`` and
 `this one <https://fabienmaussion.info/intro_to_programming/week_05/01-install-packages.html>`_
 for installing packages.
 
-We recommend to use `mamba`_ over conda as an
-installation command. Mamba is a drop-in
-replacement for all conda commands. If you feel like it, install mamba in your conda
+We recommend to use `mamba`_ over conda as an installation command.
+Mamba is a drop-in replacement for all conda commands. If you feel like it, install mamba in your conda
 environment (``conda install -c conda-forge mamba``)
 and replace all occurrences of ``conda`` with ``mamba`` in the instructions below.
 
@@ -128,24 +101,54 @@ and replace all occurrences of ``conda`` with ``mamba`` in the instructions belo
 .. _miniconda: http://conda.pydata.org/miniconda.html
 .. _mambaforge: https://github.com/conda-forge/miniforge#mambaforge
 
-The simplest way: with an environment file
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+The fastest way: `uv` tool
+~~~~~~~~~~~~~~~~~~~~~~~~~~
 
-Copy the content of the file below into a file called ``environment.yml`` on your system
-(alternatively, right-click -> "save as" on `this link <https://raw.githubusercontent.com/OGGM/oggm/master/docs/recommended_env.yml>`_).
+For users who just want to run a notebook and don't need a dedicated environment.
+This includes users who want to test OGGM before installing it, and teaching staff who want to use OGGM in a classroom.
 
-.. include:: recommended_env.yml
-   :literal:
+If you are new to Python and want to get started with OGGM as quickly as possible, we recommend using `uv`_.
+You can install and run OGGM on a Jupyter server without polluting your base system or conda environment.
 
-From the folder where you saved the file,  run ``mamba env create -f environment.yml``.
+.. _uv: _https://docs.astral.sh/uv/getting-started/installation/
 
-This will create a new environment called ``oggm_env`` in your conda installation.
-For more information about conda environments, visit the
-`conda documentation on the topic <https://conda.io/projects/conda/en/latest/user-guide/concepts/environments.html>`_. Similarly,
-visit `conda documentation on environment files <https://docs.conda.io/projects/conda/en/latest/user-guide/tasks/manage-environments.html#creating-an-environment-from-an-environment-yml-file>`_
-for more information about how to create an environment from a ``yml`` file.
+Install `uv`:
 
-Don't forget to :ref:`test-oggm` before using it!
+.. code-block:: bash
+
+    wget -qO- https://astral.sh/uv/install.sh | sh
+
+Launch a Jupyter notebook with OGGM fully installed:
+
+.. code-block:: bash
+
+    uvx --with oggm jupyter lab
+
+This installs a minimal version of OGGM into an isolated environment and runs a Jupyter server.
+
+.. note:: This will not pollute your base system or conda environment.
+
+The comfortable way: `conda` & `pip`
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+For users who don't want to use `uv` at all.
+
+Create and activate a conda environment:
+
+.. code-block:: bash
+    conda env create --n oggm_313 python=3.13
+    conda activate oggm_313
+
+Install the latest **stable** release:
+
+.. code-block:: bash
+    pip install oggm  # for a minimal install
+    pip install oggm[full]  # for a full install with all dependencies
+    uv pip install oggm  # if you already have uv installed
+
+Or install the latest **development** version:
+
+.. code-block:: bash
+    pip install --upgrade git+https://github.com/OGGM/oggm.git
 
 Install OGGM itself
 ~~~~~~~~~~~~~~~~~~~

docs/whats-new.rst

@@ -95,6 +95,10 @@ Bug fixes
 - Changed COPDEM data source (again) - this comes with good sides as the download
   is much easier now (:pull:`1773`).
   By `Fabien Maussion <https://github.com/fmaussion>`_
+- Fixed ``ModuleNotFoundError`` caused by calling ``distutils`` which has been deprecated since Python 3.12.
+  Reimplements ``strtobool`` natively.
+  By `Nicolas Gampierakis <https://github.com/gampnico>`_.
+
 
 Breaking changes
 ~~~~~~~~~~~~~~~~

oggm/cfg.py

@@ -8,7 +8,6 @@
 import sys
 import glob
 from collections import OrderedDict
-from distutils.util import strtobool
 
 import numpy as np
 import pandas as pd
@@ -892,3 +891,37 @@ def add_to_basenames(basename, filename, docstr=''):
     if '.' not in filename:
         raise ValueError('The filename needs a proper file suffix!')
     BASENAMES[basename] = (filename, docstr)
+
+
+def strtobool(value: str) -> bool:
+    """Convert a string representation of truth to True or False.
+
+    Reimplementation of distutils's ``strtobool``, which is deprecated
+    from Python 3.12.
+
+    Parameters
+    ----------
+    value : bool
+        Any value to convert.
+
+    Returns
+    -------
+    bool
+        True for "y", "yes", "t", "true", "on", and "1"; False for "n", "no", "f", "false", "off", and "0".
+
+    Raises
+    ------
+    ValueError if ``value`` is anything else.
+    """
+
+    trues = {"yes", "true", "t", "y", "on", "1"}
+    falses = {"no", "false", "f", "n", "off", "0"}
+
+    if isinstance(value, str):
+        value = value.lower()
+        if value in trues:
+            return True
+        if value in falses:
+            return False
+
+    raise ValueError('Expected "%s"' % '", "'.join(trues | falses))

oggm/tests/funcs.py

@@ -1,6 +1,5 @@
 import os
 import shutil
-from distutils.util import strtobool
 import hashlib
 
 import numpy as np
@@ -331,7 +330,7 @@ def patch_minimal_download_oggm_files(*args, **kwargs):
 
 def use_multiprocessing():
     try:
-        return strtobool(os.getenv("OGGM_TEST_MULTIPROC", "False"))
+        return cfg.strtobool(os.getenv("OGGM_TEST_MULTIPROC", "False"))
     except BaseException:
         return False
 

oggm/tests/test_utils.py

@@ -398,6 +398,23 @@ def test_cook_rgidf(self):
         assert_allclose(df.cenlon, cgidf['Glc_Long'])
         assert_allclose(df.rgi_area_km2, cgidf['Glc_Area'] * 1e-6, rtol=1e-3)
 
+    def test_strtobool(self):
+        trues = {"yes", "true", "t", "y", "on", "1"}
+        falses = {"no", "false", "f", "n", "off", "0"}
+        for value in trues:
+            assert isinstance(cfg.strtobool(value=value), bool)
+            assert cfg.strtobool(value=value)
+            assert cfg.strtobool(value=value.upper())
+
+        for value in falses:
+            assert isinstance(cfg.strtobool(value=value), bool)
+            assert not cfg.strtobool(value=value)
+            assert not cfg.strtobool(value=value.upper())
+
+        for value in {"ft", None, 2, True, False}:
+            with pytest.raises(ValueError):
+                cfg.strtobool(value=value)
+
 
 class TestInitialize(unittest.TestCase):
 

oggm/utils/_funcs.py

@@ -28,7 +28,10 @@
 import shapely.geometry as shpg
 from shapely.ops import linemerge
 from shapely.validation import make_valid
-from salem.gis import Grid
+try:
+    from salem.gis import Grid
+except ImportError:
+    pass
 
 # Optional libs
 try: